NOTE TO READER:
This is an abbreviated version of the thesis. Papers V and VII are missing from the
appendices as these are pending publication.
A list of errata can be found at the end of the appendices.
MR
Thesis for the degree of Doctorate in Biology, Östersund 2008
HEMATOLOGICAL CHANGES ARISING FROM SPLEEN
CONTRACTION DURING APNEA AND ALTITUDE IN HUMANS
Matt X. Richardson
Supervisors:
Prof. Erika Schagatay
Dr. Johan Andersson
Department of Natural Sciences, Engineering and Mathematics
Mid Sweden University, Östersund, Sweden
ISSN 1652-893X,
Mid Sweden University Doctora l Thesis 57
ISBN 978-91-86073-03-9
Akademisk avh andling som med tillstånd av Mittuniversitetet i Östersund
framläggs till offentlig granskning för avläggande av doktorsexamen i Biologi,
den 19e september 2008, kl. 10 i sal F234, Ridhuset, Mittuniversitetet, Östersund.
Seminariet kommer att hå llas på engelska.
HEMATOLOGICAL CHANGES ARISING FROM SPLEEN
CONTRACTION DURING APNEA AND ALTITUDE IN HUMANS
Matt X. Richardson
© Matt X. Rich ardson, 2008
Department of Natural Sciences, Engineering and Mathematics
Mid Sweden University, Östersund
Sweden
Telephone:
+46 (0)771-975 000
Printed by Kopieringen Mid Sweden University, Sundsvall, Sweden, 2008
Cover photographs by Annelie Pompe (World freediving championships, Egypt)
and Mike Callagh an (Lobuche Peak, Nepa l).
i
HEMATOLOGICAL CHANGES ARISING FROM SPLEEN
CONTRACTION DURING APNEA AND ALTITUDE IN HUMANS
Matt X. Richardson
Department of Natural Sciences, Engineering and Mathematics
Mid Sweden University, Östersund, Sweden
ISSN 1652-893X, Mid Sweden University Doctora l Thesis 57;
ISBN 978-91-86073-03-9
ABSTRACT
The spleen acts as a storage site for red blood cells, and contraction of the spleen
is for many mammals a way of coping with situations where increased blood
oxygen storage is beneficia l. This ability h as a lso recently been found in humans,
where most individuals respond with an increase in circulating erythrocytes in
response to environmenta l or physiologica l stress, th e majority of which can be
a ttributed to the spleen. The initiation of th is response, however, is unclear, as is
the variability of the response’s development among individuals. Th is thesis
identif ies th at there is a strong correla tion between spleen contraction and
circulating hemoglobin (Paper I), th a t release of these erythrocytes from the
spleen is more pronounced in those trained in breath- hold diving th an in elite
cross-country skiers and untra ined persons (Paper II), and th a t the response
develops even in oxygen-poor environments such as at h igh a ltitude (Paper III) .
Th is is due to the hypoxic stimulus th at exists such environments, wh ich acts as
an initia tor of the response (Paper IV), a lthough even high levels of carbon
dioxide are important (Paper V). The size of world ch ampionship apnea divers’
spleens is a lso predictive of the ir apneic performance, showing th a t spleen
contraction is an important factor in natural diving ability (Paper VI). In
conclusion, the contractile response of the spleen, a long with the previously
described “diving response” helps humans to better endure low-oxygen situations,
but despite the ir common function these responses are not initia ted and do not
develop in the same manner (Paper VII).
Keywords: hypoxia, hypercapnia, hemoglobin, hematocrit, breath- hold diving
ii
SAMMANDRAG
Mjälten innehå ller ett extra lager av röda blodkroppar, och mjälteskontraktion
är ett sätt för många däggdjur att anpassa sig till situa tioner där en ökad lagring
av syre i blodet är en fördel. Denna förmåga h ar rela tivt nyligen upptäckts hos
människan där flesta personer svarar på fysiologisk eller omgivningsrela terad
stress med en ökning av mängden erytrocyter i cirkula tionen vilket til l stor del
antas komma från mjälten. Det är dock inte känt hur responsen startas och varför
den är olika väl utvecklad hos olika personer. Denna avh andling visar att det
finns ett nära samband mellan mjälteskontraktionen och blodets Hb (Paper I), att
utsöndring av erytrocyter är större bland tränade apnédykare än bland
elitskidåkare och otränade personer (Paper II) och att responsen uppstår även i
syrefattig miljö t ex på hög höjd (Paper III). Det beror på att syrebristen i sig är
en viktig stimulus för att initiera responsen (Paper IV), men även höjd
koldioxidnivån är en viktig faktor (Paper V). En korrela tion upptäcktes mellan
storleken på apnédykares mjältar och deras resulta t på ett världsmästerskap
vilket visar att mjälteskontraktionen är viktig för människans naturliga
dykförmåga (Paper VI). S lutsatsen är att mjältesresponsen tillsammans med den
tidigare beskrivna ”dykresponsen” hjälper oss att bättre klara perioder med
syrebrist, och att dessa responser samverkar funktionellt men inte startas på
samma sätt (Paper VII).
Nyckelord: hypoxi, hyperkapni, hemoglobin, hematokrit, apnédykning
iii
TABLE OF CONTENTS
ABSTRACT ..............................................................................................................................II
SAMMANDRAG ...................................................................................................................III
TABLE OF CONTENTS......................................................................................................IV
LIST OF PAPERS ..................................................................................................................VI
INTRODUCTION.................................................................................................................... 1
What is apnea?.................................................................................................................... 1
What is hypobaria?............................................................................................................. 4
Differences in adaptation to hypoxia in humans ..............................................................4
Morphology of the spleen................................................................................................... 5
Functions of the spleen....................................................................................................... 6
Contraction of the spleen in non-human species ..............................................................8
Contraction of the spleen in humans............................................................................... 10
Initiation of spleen contraction........................................................................................11
Effects of the spleen on cardiovascular function............................................................13
AIM OF THE THESIS.......................................................................................................... 14
METHODS..............................................................................................................................14
MEASUREMENT OF BLOOD VARIABLES ................................................................................. 14
Venous capillarisation for Hb/Hct................................................................................... 14
MEASUREMENT OF SPLEEN SIZE............................................................................................15
MEASUREMENT OF OTHER PARAMETERS .............................................................................. 16
Arterial oxygen saturation ...............................................................................................16
Mean arterial pressure..................................................................................................... 17
Skin blood flow.................................................................................................................. 17
Respiration and gases....................................................................................................... 18
APNEA AND HYPOXIA EXPERIMENTAL PROTOCOLS .............................................................19
Apnea series ...................................................................................................................... 19
Inspired gases and hyperventilation................................................................................ 20
STATISTICAL APPROACHES.................................................................................................... 20
Publication statistical analysis method........................................................................... 20
Additional statistical analysis method............................................................................. 21
SUMMARY OF RESULTS ................................................................................................. 23
OCCURRENCE OF SPLEEN CONTRACTION .............................................................................. 23
iv
INITIATION OF SPLEEN CONTRACTION .................................................................................. 25
CONSEQUENCES AND APPLICATIONS OF SPLEEN CONTRACTION ......................................... 27
DISCUSSION..........................................................................................................................28
OCCURRENCE OF SPLEEN CONTRACTION .............................................................................. 28
INITIATION OF SPLEEN CONTRACTION .................................................................................. 29
Stimulus detection by chemoreception ............................................................................ 30
Output following chemoreception.................................................................................... 31
Other potential output mechanisms................................................................................. 34
CONSEQUENCES AND APPLICATIONS OF SPLEEN CONTRACTION ......................................... 34
In apneic performance...................................................................................................... 35
In acute adaptation to high altitude ................................................................................ 36
During exercise................................................................................................................. 37
Problems with Hct limits in competition......................................................................... 37
Relation to the diving response........................................................................................38
FUTURE DIRECTIONS.............................................................................................................. 39
Other measurement techniques........................................................................................39
Maximizing splenic contraction through training..........................................................39
Investigation of other possible sources of stored erythrocytes...................................... 40
ACKNOWLEDGEMENTS.................................................................................................. 40
REFERENCES.........................................................................................................................41
PAPER I. .................................................................................................................................. 50
PAPER II. ................................................................................................................................. 55
PAPER III.................................................................................................................................63
PAPER IV. ...............................................................................................................................73
PAPER V. ................................................................................................................................85
PAPER VI. ...............................................................................................................................98
PAPER VII.............................................................................................................................113
v
LIST OF PAPERS
Th is thesis is based on the following seven papers, hereinafter referred to by
their Roman numerals:
Paper I
Correlation between spleen size and hematocrit during apnea in
humans. MX Rich ardson, R deBruijn, S Pettersson, J Reimers and E Schagata y .
2006. In: Lindholm, P, Pollock NW, Lundgren CEG, eds. Breath- hold diving.
Proceedings of the Undersea and Hyperbaric Medica l Society/Divers Alert
Network 2006 June 20-21 Workshop, p 116-119.
Paper II
Increase of hemoglobin concentration after maxima l apneas in
divers, skiers and untra ined humans. 2005. MX Rich ardson, R deBruijn, HC
Holmberg, G Björklund, H Haughey and E Schaga tay. Canadian Journal of
Applied Physiology, 30(3): 276-281.
Paper III
S hort-term effects of normobaric hypoxia on the h uman spleen.
2008. MX Rich ardson, A Lodin, E Schagatay and J Reimers. European Journal of
Applied Physiology, published online Nov 28 2007, ahead of print.
Paper IV
Hypoxia augments apnea-induced increase in hemoglobin
concentration and hematocrit. 2008. MX Rich ardson, R deBruijn, and E
Schagatay. Submitted for publication in the European Journal of Applied
Physiology.
Paper V
Hypercapnia moderates hemoglobin increases during apnea. 2008.
MX Rich ardson, A Lodin, H Engan, and E Schagatay. In manuscript.
Paper VI
Spleen and lung volumes correla te with performance in elite apnea
divers. 2008. E Schagatay, MX Rich ardson and A Lodin. Submitted for
publication in the Journal of Sports Sciences.
Paper VII
Cardiovascular and hematological adjustments to apneic diving in
humans: Is the spleen response part of the diving response? 2006. E Schagatay,
MX Rich ardson, R deBruijn and JPA Andersson. In: Lindholm, P, Pollock NW,
Lundgren CEG, eds. Breath- hold diving. Proceedings of the Undersea and
Hyperbaric Medica l Society/Divers Alert Network 2006 June 20-21 Workshop, p
108-115.
vi
vii
INTRODUCTION
It is human nature to explore the environments around us, no matter how remote or
challenging. In these cases, humans voluntarily subject themselves to hostile or
dramatica lly changing environmenta l factors such as temperature extremes or
a lterations in atmospheric composition, despite the fact th a t such actions
threaten the ir normal functioning, or in some cases, their surviva l. Luckily,
humans are likely the only species capable of adapting to virtually every
environment on the planet, by virtue of our intelligent use of tools and therein the
ability to produce cloth ing, shelter, preserve foodstuffs and arrange effective
transport methods.
Strip away the largely “technological” advances and humans are considerab ly
less impervious to the ir surroundings. A naked diver in the cold North Atlantic,
for example, will h ardly last a tenth of the time of h is drysuit-clad equiva lent.
Other environments th a t pose particularly acute problems for the human are
those of reduced oxygen. Depending on the severity of the reduction, vita l bodily
functions can become compromised on a very rapid time scale (minutes), a lthough
humans show wide varia tion in their ability to adapt to such reductions. These
environments are generally encountered in two forms: 1) during apnea, whether
voluntary or involuntary, and 2) in hypobaria at h igh altitude.
Yet even the best mountaineers in the world, without bottled oxygen, begin to
drown their lungs and brains with oedema at our planet’s highest a ltitudes
with in hours. Some individuals can adapt to th ese situations on a purely
physiologica l level and may thus endure these environments for short periods.
Strong evidence for th is is provided in the current world record for apnea, wh ich
a t over 9 minutes is far in excess of commonly reported asphyxia surviva l times.
Those th a t cannot adapt face a considerably less auspicious outcome. These
differences in adapta tion th a t determine an individual’s fate are stil l large ly
unknown. The intent of th is thesis is to shed some light on one piece of th is
adaptive puzzle.
What is apnea?
Apnea is, simply put, the cessation of breath ing. When apnea is voluntary it is
a lso often referred to as breath holding; examples of th is behaviour among
humans occur in harvest divers, spear fishers, and competitive free divers. Gas
exchange with the environment is ha lted and the body must make do with t he
oxygen (O2 ) stored in the body i.e. in the lungs, the blood, and the tissues. Th is
1
reduces the amount of O2 in the body with time, and a t the same time increases
the amount of carbon dioxide (CO2 ). The ongoing metabolic processes in the body
will continue to use these O2 stores (and produce CO2 ) until, in voluntary cases,
breath holding ends due to discomfort with the increasing urge to breathe, and
gas exchange with the environment resumes.
Depending on the tenacity of the individual, the amount of O2 reaching the
various tissues of the body during breath holding may become insufficient in order
for them to continue the ir metabolic processes norma lly, known as hypoxemia.
The oxygen saturation level of the arteria l blood (S a O2 ) provides a good
indication of the severity of hypoxemia atta ined in the body [1], as the parti a l
pressure of O2 in the a lveoli (P AO2 ) and in the arteries (Pa O2 ) continues to
decrease in a constant relationship to each other throughout apnea [2]. Norma l
S a O2 values during breath ing are typically 97-99% but individuals tra ined in
breath holding (known as apneists or freedivers) can reduce values to less th an
50% during extended breath holding.1 The desaturation rate is more rapid when
the initia l lung volume is smaller [1] and when initia l mixed venous O2 saturation
is lower [4].
A marked slowing of heart rhythm (bradycardia) mediated via parasympatetic
vagus nerve activity and a progressive increase in mean arteria l pressure (MAP)
occur with in 30 s of breath holding [5], a long with a sympathetically mediated
reduction of blood flow to e.g. the extremities and skin via narrowing of the blood
vessels (periphera l vasoconstriction). Th is series of measures helps in conserving
the diminish ing O2 for extremely vita l processes such as brain and heart function
– at the expense of other tissues and organs – and are collectively known as the
human diving response.2
The lack of gas exchange during apnea a lso means th a t the CO2 produced by the
metabolic processes of the body cannot be removed from the venous blood. Th is
causes subsequent elevation of CO2 in the arteria l blood, known as hypercapnia .
1
To put th is in context, va lues of less th an 90% define clinica l hypoxemia, and in
a hospita l setting are typically considered an acute problem in patients [3].
2
Wh ile some experiments described in th is disserta tion refer to or compare the
parameters th a t make up the diving response in apneic situations, it is not t he
focus of th is work. For a more deta iled review, the theses by Schagatay [6] and
Andersson [7] and review by Foster and Sheel [8] provide thorough discussion of
the human diving response.
2
The hypercapnia is augmented by shrinking lung volume and the Ha ldane effect
from oxygenation of hemoglobin and liberation of hydrogen ions [9]. Venous blood
experiences a lesser rise in CO2 partia l pressure (PCO2 ) due to the Ha ldane effect,
which in venous blood works in the opposite direction to arteria l blood, and due
to buffering of CO2 by the tissues. CO2 output into the lungs reduces progressively
and eventually stops; continued breath holding reverses the cycle and sees CO2
move from the lungs into the arteria l blood [2, 9].3
As arteria l PCO2 is the main determinant of voluntary apneic duration, preapneic hyperventila tion can prolong apnea by reducing the CO2 content of t he
lungs, blood, and eventually the tissues (ref e.g. Lin). This allows the “break ing
point” of apnea, where PCO2 reaches a critica l level th at can no longer be
voluntarily tolerated, to be reached la ter. Pre-apnea breath ing of O2 can also
increase apneic duration, as the arteria l blood will remain normally saturated
until the lungs start to become depleted of oxygen [9]. Thus, the development of a
hypoxic stimulus and the onset of tissue hypoxemia are delayed. Klocke and
Ra hn [12] found th at combining hyperventilation and O2 breath ing could result in
breath- holding times up to 14 minutes with the entire lung vita l capacity volume
absorbed during apnea in some subjects.
Repetition of apnea with short rest intervals (2-3 minutes) a lso results in an
increase in apnea time and the time until the “physiological break ing point”,
defined as the onset of involuntary breath ing movements [13]. Th is occurs despite
unchanging pre-apneic arteria l PCO2 and lung volume, which indicates th at
other factors are responsible.
As a physiologica lly stressful experience, sympathetic activity is also increased
in various physiological systems during apnea. Renal sympathetic nerve activity
and MAP were shown to increase with apneic duration in cats, although these
could be reduced by 80% and 100%, respectively, when pre-ventila ted with O2
and a matched level of PaCO2 [14]. Resumption of breath ing caused an immedia te
fa ll in sympathetic nerve activity. Leunenberger and associates found th a t
sympathetic nerve activity was greatest during hypoxic apnea and attenuated
3
The abovementioned CO2 dynamics are only relevant for surface breath holding.
In the case of diving to depth, CO2 moves from the alveoli into the blood due to
lung compression, wh ich increases the PCO2 in the lung above th a t of the venous
blood; during ascent, the opposite occurs [10]. The extent to which the blood acts
as the CO2 store during continued breath- holding at the bottom is dependent upon
the rela tionships between the alveolar PCO2 and the PCO2 in the tissues [11].
3
during hyperoxic apnea [15], and was also increased during intermittent hypoxia
without apnea [16].
What is hypobaria?
Hypobaria occurs naturally as we ascend to altitude, where the pressure of the
a tmosphere (and thus the partia l pressures of the gases th a t it is composed of) is
lower th an at sea level. Humans exposed to such situations include high altitude
natives and residents, mountain climbers, and aviators, among others. It is also
possible to artif icia lly impose a reduced amount of O2 while mainta ining a
normal atmospheric pressure (i.e. 760 mmHg, or sea level) by filtering out O2 and
replacing it with another gas, typica lly nitrogen. This is called normobaric
hypoxia.
W h ile most known adverse effects of high a ltitude are ascribed to the reduction
in ambient O2 , there may be some differences between hypobaric and normobaric
hypoxia. The former leads to a slightly greater hypoxemia, reduction of CO2 in
the blood (hypocapnia), blood alka losis and lowering of Sa O2 ; however, the endtida l partia l pressure of O2 and CO2 are similar between the two conditions [17].
Despite these differences, it is common to study most effects of a ltitude by
simulating it in normobaric hypoxic environments.
W hen the onset of hypoxia is rapid, the human body follows a time dependent
series of adapta tions, which on the shortest time scale (minutes) begins wit h
increased ventila tion and abrupt changes in cardiovascular dynamics [18]. These
include an increase in sympathetic activity in the body, resulting in increased
cardiac output, centra lization of blood volume, and constriction of the venous
capacitance vessels [19-21]. An increase in hematocrit is a lso noted. Chronic
hypoxia further elevates hemoglobin via release of erythropoietin, and
production of new red blood cells with in days. However, the effect of t he
responses to both acute and chronic hypoxia is an attempt to increase O2 -carrying
capacity.
Differences in adaptation to hypoxia in humans
W h ile the physiological adapta tion to hypoxia is similar among most
individuals, the magnitude or effectiveness of some aspects of th is response are
not. Both acute and acclimatory adaptive responses to hypoxia appear to have
certain characteristics th at vary amongst individuals or groups of individuals.
According to Hochachka [22], these include vaoscontrictive responses mediated
by lung vasculature O2 sensors, production of new blood vessels, and mainta ined
4
Figure
1.
The
spleen,
il lustrated with vein and
artery. From Gray’s Anatomy,
1918.
erythropoeisis via kidney O2 sensor
output, leading to increased red blood cell
mass and O2 -carrying capacity. Andean
and Hima layan natives show blunted
responses to acute hypoxia compared to
lowlanders, suggestive of a h igh-a ltitude
phenotype in, for example, Quechuas and
S herpa populations. Although diff icult to
quantify, genetic and environmenta l
contributions to the variation in these
response characteristics are general ly
considered to be equal [23, 24], leaving a
fa ir amount of “flexibility” in the extent
to wh ich individuals sense and respond to
hypoxia [25, 26].
An increase in the amount of circulating O2
in the body is, at its most essentia l leve l,
necessary to improve oxygenation at t he
tissue level in oxygen-poor environments where breath ing is still permitted [27].
W h ile th is is possible at a ltitude, in the case of apneic hypoxia ventila tory
responses are not immediate ly possible during apneic hypoxia. Here, decreases in
heart rate are instead noted, a lthough an increase in hematocrit is still present.
The increase in hematocrit (Hct) and hemoglobin concentration (Hb) noted in
both cases of acute hypoxic exposure is not well understood. On such a short time
scale, where erythropoietin-stimulated red blood cell production is not ye t
rea lized, the intervention of a source of stored blood cells may be one possible
explanation yet to be explored in humans. And the spleen, an organ th a t
throughout history has had a multitude of afflictions and functions attributed to
it, may be an idea l candidate for such a storage function. The spleen’s prospective
role in th is adapta tion to these environmenta l stressors is the main focus of th is
thesis.
Morphology of the spleen
Found in humans beneath the diaphragm in the upper left abdomen, the spleen is
a uniquely complex organ with both immune and circulatory functions. According
to Gray’s Anatomy it normally makes up approximately 0.25% to 0.29% of tota l
body weight, and has a volume of about 200-250 cm3 in adults [28]. Two coats
surround the organ: a smooth external coat or tunica serosa, and an interna l
fibroelastic coat or tunica albuginea. The internal coat reflects inwards as sheaths
5
which in turn branch into numerous small f ibrous bands, or trabeculae, forming the
framework of the spleen. The small spaces created with in th is trabecular
framework constitute the spleen’s pulp.
The pulp itself can be grossly divided into white and red sections, the ir
respective colours obtained by the presence of a large number of lymphocytes and
erythrocytes. Not surprisingly, the wh ite pulp performs immunological functions
as part of the largest lymphoid organ in the body, while the red pulp is t he
largest filter of blood in the human [29]4. Arteria l blood, brought to the spleen
via the splenic artery, is transported through a series of cords and then into
venous sinuses of the red pulp. The endothelia l lining of the sinuses contains nonstriated muscular filaments, and with its cells in para llel formation has t he
ability to force blood from the cords to the sinuses through fibrous slits [31].
Older, less-flexible erythrocytes may become trapped at th is stage and are
“pitted” for recycling of the ir heme components [32]. However, the flexibility
and contractility of these fibres also appear to help reta in more viable
erythrocytes in the spleen, forming a reservoir. Via th is meshwork, the spleen
concentrates blood in the reservoir to about twice th at of the arteria l hematocrit
circulating in the rest of the body, from normally 40% to approximately 78% [33],
though others have estimated th is va lue as h igh as 96% [34]. Although t he
quantity of th is reservoir can vary greatly dependent on the size of the organ, it
is generally believed to conta in on average 200-250ml of concentrated blood [35,
36].
Functions of the spleen
The early medical practitioner Galen (circa 200 AD) identified the spleen as the
source of “black bile” in the body, up until the Middle Ages considered to the most
noxious substance in the body, the release of which both anger and melancholy
was often associated to5.
4
W h ile the spleen’s immune functions are considerable, it is not the focus of th is
disserta tion and therefore will not be discussed in deta il. For more information
about th is aspect of spleen function, Mebius [30] provides a thorough review.
5
Indeed, the contents of the spleen may appear black (for example, upon
splenectomy or autopsy) but th is is almost certa inly due to the elevated
hematocrit level, wh ich differentia tes it from normally circulating arteria l
blood tha t is more reddish on account of its h igher dilution with plasma.
6
As medical science modernized, late 19th and early 20th century experimentation
on the circulatory roles of the spleen involved domesticated animals, typical ly
dogs and cats. In these species, it was established th a t the spleen’s volume was
variable, capable of rapid contraction resulting in a discharge of its stored
contents, increasing both circulating red blood cells and arteria l pressure [37-39].
It was a lso found to contract and dila te rhythmica lly under normal physiologica l
conditions, albeit to a lesser extent [40, 41]. Later experimentation seemed to
confirm the role of the spleen as a storage site, as blood volume returned to normal
only minutes after large infusions of saline, dextrose and whole blood in intact
dogs, but not in splenectomised dogs [42]. There was a lso evidence for the
discharge of vasoactive substances from the spleen itself during its contraction,
which might further affect arteria l pressure above and beyond the effects of t he
spleen’s expelled volume [43]. More recently, in rats there is evidence th at it may
be capable of rapidly filtering out plasma from the blood and into the lymphatic
system when blood volume is expanded [44]. The storage function of the spleen is
a lso considered important in animals for survivability in cases of severe
h aemorrhage or shock, and restoring volume homeostasis when such
interventions were not required [45, 46].
Despite such research findings, however, the concept of similar functions in the
human spleen was disregarded in many medica l textbooks as la te as the 1980s.
Th is was undoubtedly due to the frequent observa tion tha t splenectomy, a
common procedure in traumatic injury, could be performed without noted adverse
long-term effects for the individual. Later long-term studies showed, however,
th a t while not necessary for human surviva l per se, remova l of the spleen
increases the risk for life-threatening bacteria l infections [47]. The more recent
observation th at the human spleen does contract [48-50], on the other hand, is
still largely unknown to most practitioners of medicine, and the phenomenon is
not well investigated or understood.
In humans, 70-80% of the tota l circulating blood volume is in the venous system a t
any given time, of wh ich the largest pools can be found in the cutaneous and
splanchnic circulation - the la tter composed of th e gastric, small intestina l,
colonic, pancreatic, hepatic, and splenic circulations, arranged in para lle l wit h
one another. As part of the splanchnic circulation, the spleen contributes to
homeostatic regulation of both blood volume and pressure and acts as an
“autologous blood bank” in times of depleted intravascular volume, and also
unloads the circulation by absorbing excess volume [51]. The spleen is also
considered important in regulating blood hematocrit, itself a major determinant
of blood viscosity and factor responsible for of the O2 -carrying capacity of t he
blood. Th is may occur both through filtration but also through plasma volume
7
regulation [51]. Possibly as an indication of these functions, the spleen receives a
rela tively large percentage of the cardiac output (up to 5%) for its size, with a
flow per gram of tissue 10 times h igher then th at of resting skeleta l muscle and
a lmost similar to the blood flow to the heart muscle [52]6. Only 75% of th is
volume is returned to the splenic vein at any given time, likely due to the
filtra tion of plasma into the lympha tic system, which occurs in the red pulp and
results in an elevated hematocrit in the spleen’s stored contents [51]. Blood
viscosity and hematocrit h ave been shown to increase in humans after
splenectomy [50, 70], likely due to the absence of the spleen’s filtering and storage
functions.
Contraction of the spleen in non-human species
Spleen size has been found to correlate with dive capacity i.e. the duration of
diving in pinnipeds, and th is also holds true across the phylogenetic tree –
species which h ave larger spleens have a lso longer dive times [53, 54]. The
Weddell seal is arguably most well endowed in terms of spleen size, wit h
approximately two-th irds of the sea l’s red blood cells sequestered in the spleen
during rest [55], and can also be considered one of th e most accomplished divers
amongst sea l species. Qvist and colleagues [55], as well as Thornton and
colleagues [56], the latter using magnetic resonance imaging (MRI), demonstrated
a fast contraction of the spleen during diving in seals, with concomitant increase
in red blood cells ejected into the circulation. Th is conceivably allows for an
increase in O2 storage capacity during diving and more effective O2 replenishment
during rest periods between dives [57]. Both the Qvist and Thornton groups also
noted th a t a re-uptake of the ejected blood occurs with in 15 minutes post-dive.
Th is recovery of the ejected blood by the spleen during non-diving periods would
prevent excessive circulatory viscosity when O2 was not limiting [58].
Many varied stimuli h ave been shown to cause a reduction in spleen volume in
mammals. In dogs, rebreath ing, injection of epinephrine, and fright from loud
noises caused significant transient decreases in spleen size [59]. Further
experimentation in dogs confirmed th at the blood volume increases seen after
fright, exercise and hypoxia were exclusively associated with spleen contraction,
and not the liver [46], although Carneiro and Donald [45] did find th a t
h aemorrhage caused decreased blood volume in the liver as well, a lbeit about
h a lf of the loss from the spleen. It was also found th a t an increase in blood
6
The spleen is also therefore particularly troublesome when ruptured or
damaged, as it can lead to massive internal blood loss and death.
8
viscosity a lso followed “emotional excitement” in rats and rabbits but th a t th is
increase was far smaller if the ir spleens were removed [60]. S heep [61], goats [62],
pigs [63] and horses [64] all show significant elevations of hematocrit during
exercise, which can be eliminated via splenectomy. Conceivably the purpose
underlying increases in circulating red blood cells is the same in virtually a l l
situations named above – a resultant increase in O2 uptake and/or an ability to
recover quickly during rest, thus reducing fatigue and improving performance
during physiologica lly stressful or demanding situations.
Kramer and Luft [65] found tha t hypoxia a lso caused appreciable reductions in
spleen mass in dogs, calculating blood expulsion in the order of 16-20% of
circulating blood volume. Th is effect was reversible with the remova l of
hypoxia. Tellingly, the hematocrit measured in the splenic vein during expulsion
was on average 80% h igher th an in arteria l blood, but during recovery from
hypoxia th is pattern was reversed, providing evidence for the storage of red cells
by the spleen7. Furthermore, the ejected blood in th e splenic vein maintained
consistently h igher O2 saturation th an the arteria l blood in the hypoxic dogs,
suggesting th a t stored cells in the spleen were well oxygenated. These findings
were la ter confirmed in cats, whereby rapid reduction in ambient O2 levels to
3.5% caused an increase in hematocrit as well as surviva l time, compared wit h
splenectomised cats [67]. Plasma protein values were not different between the
two conditions, suggesting circulating plasma volume was unaffected. Hoka et a l
[68] confirmed th a t severe hypoxia was accountable for reductions in spleen
volume, of wh ich 60% was accounted for by active spleen contraction due to
sympathetic efferent nervous discharge to the spleen. This sympathetic nervous
discharge increased six-fold over baseline activity during hypoxia in dogs.
Increases in Hb from the spleen were a lso noted in rats when exposed to hypoxia
after undergoing 1 h of hypoxia a day for 4 days, but not in rats exposed for only 1
day [69]. However, after the spleen contraction response was established, t he
ability to elicit it again with a single hypoxic stimulus remained for 60 days
thereafter.
7
Radioactive labelling in dogs showed that tagged cells rapidly reach an equilibrium
concentration in the spleen [66], making it difficult to distinguish changes in red cell
volume calculations attributable to spleen contraction when measured on a total-body scale.
Targeted or regional analysis of radioactivity when using this method is required to analyse
blood reservoir shifts.
9
Contraction of the spleen in humans
During exercise8
Sandler and associates [73] used radioactive labelling to demonstrate spleen
shrinkage during postural changes (from supine to standing) and during supine
exercise, resulting in a reduction in splenic radioactivity and increase in blood
radioactivity and red blood cell count. Spleen volume was found to decrease a lso
during upright exercise [74, 75] while kidney and liver volume showed mixed
results, with some showing no change [75] and others showing decreased volume,
though to a lesser extent th an the spleen [74]. Plasma norepinephrine and
epinephrine levels have a lso been shown to closely correlate with spleen volume
during graded exercise [49].
Wolski [34], in her investigation of spleen responses to exercise, noted an
approximately 9% reduction the tota l red cell volume of untra ined subjects’
spleens after exercise, with an overa ll organ volume reduction from 338 mL to 143
mL. The proportion of red blood cells released by the spleen did not differ
between untra ined or aerobica lly tra ined subjects and even exercise in hypoxia
did not increase the amount of red blood cells released.
Stewart and associates [72] found reductions in spleen volume from cycling at 60%
maximum O2 uptake after 5 min (28% reduction), 10 min (30%), and 15 minutes
(36%). An incrementa l ride to exhaustion resulted in a 56% reduction in spleen
volume. They suggested th at the spleen regulates its volume during exercise in
response to an intensity-dependent signal, in which plasma catecholamines may
play a role.
During apnea
The human spleen has a lso been found to contract during apnea or breath holding,
with a subsequent increase in both Hb and Hct [48, 50, 76, 77] th a t is not caused by
hemoconcentration [50, 76]. Splenectomised subjects do not demonstrate these
hematological changes after apnea, and also lacked the typica l prolongation of
8
General points regarding exercise: plasma volume decreases by 10 to 20% during exercise
[71], causing an increase in hematocrit and blood viscosity. This is caused by ultrafiltration
of plasma extravascularly and the movement of plasma to muscle, as well as through fluid
loss. However, according to Stewart, it appears unlikely that splenic release of erythrocytes
during exercise has any effect on indirect measures of plasma volume [72].
10
repeated apneas, suggesting the effect is mainly attributable to the spleen [50,
76].
Maximization of spleen contraction through apnea appears to require more th an
one apnea, with maximum contraction and Hb/Hct increase typica lly occurring
after 3-5 apneas [13, 78], a lt hough one study ha s suggested th a t maxima l
contraction is perhaps atta inable after a single apnea [35].
Even very short breath holds (30s) have been shown to elicit reductions in spleen
volume in humans, and the amount of hematological increase is related to t he
degree of contraction [77]. There is, however, uncerta inty regarding the influence
of apneic duration on degree of contraction of the spleen. One study has
demonstrated increased spleen contraction with increased apneic duration [77]
while other work has demonstrated powerful contraction also in time-limited
apneas [79].
Some benchmark studies of spleen contraction have reported mean reductions in
spleen volume of 18% [79], 20% [48], and 25% [35]. It has been shown tha t th is
contraction is likely actively initia ted, th a t is, not due to passive collapse or
reduced flow in the splenic artery [35]. The spleen remains contracted even
between apneas with short rest periods (2-3 minutes) in between [35, 50] but
typica lly returns to its relaxed size and volume with i n 10 minutes barring further
apneic intervention [35, 50, 77, 79]. Th is is a lso the duration known to reverse the
increase in apneic duration during repetitive attempts [79]. Wh i le the
observations by Hurford et a l [48] suggested th a t only trained Ama divers
responded with a spleen contraction, other studies reveal th a t both individuals
tra ined and untra ined in breath- hold diving demonstrate splenic contraction and
increases in hematological parameters [50, 77], although differences between
those trained or not tra ined in breath holding have also been demonstrated [76].
Initiation of spleen contraction
There is virtually no consensus about the “trigger(s)” of splenic contraction. Much
research points to the possibility of an underlying active, sympathoadrenergic
mechanism. The earliest work demonstrated splenic contraction after stimulation
of the splenic nerve in dogs [39]. Th is was aga in confirmed in later work [80, 81]
which observed th at stimulation of the sympathetic nervous input, and not
mechanica l restriction of inflow, to the dog spleen caused it to expel its stored
blood. Spleen size in Weddell sea ls has been reported to be correlated to plasma
catecholamine concentrations during diving and also contracts in response to
exogenous epinephrine infusion [82]. Innervation of the dog, cat, rat, mouse, and
11
human spleen are a ll adrenergically (sympathetica lly) controlled wh ile
cholinergic (parasympathetic) and sensory/tactile innervation is limited or nonexistent [83]. The spleen belongs to the most densely sympathetica lly innervated
organs, with a norepinephrine turnover severa l times higher th an the liver or
the lungs [84].
Accordingly, epinephrine, norepinephrine, and neurostimulation all cause
a lp h a-media ted vasoconstriction and contraction of the splenic capsule in both
non-humans and humans [82, 85-88]9. Indeed, Sh imizu and associates [89] using
microdialysis in conscious rats found th at norepinephrine recovered in the spleen
was derived from the nerve terminals of the splenic sympathetic nerve. Rogausch
and associates [52] found th a t removal of noradrenergic sympathetic transmission
to the spleen resulted in increased spleen blood flow and cell retention. The !1adrenergic receptors appear to be activa ted during spleen contraction in rats [90].
Numerous and varying stimuli ranging from corticotrophin-releasing factor from
the brain [89], cocaine administration [91], to vasoactive amines [92, 93] may have
the same catecholamine-releasing effects resulting in splenic contraction. In the
case of repetitive exposure to short-term hypoxia, observations in rats suggests
th a t splenic contraction is enabled by an increased response of the !2-adrenergic
receptors to their catecholamine agonists [69]. But there is little support for other
vasoactive substances such as nucleotides, ions, angiotensin II etc. influencing
spleen volume through direct effects in humans [83].
Another possibility is th a t the mechanisms underlying splenic contraction also
h ave the ir roots at least partia lly in chemoreception. Hoka and associates’
study [68] of hypoxia in dogs suggested th a t both “early” sympathetic nerve
activity (1 minute after initiation of hypoxia) and “la te” activity (2.5 minutes
after hypoxia) could be expla ined by chemoreceptor stimulation and centra l
nervous system (i.e. medullary) stimulation, respectively. The la tter, a lso known
as the central pressor effect [94] may be augmented by decreased cerebra l
perfusion pressure. Cutler and associates [95] found th a t periods of intermittent
hypoxic apnea could alter chemoreflex control of muscle sympathetic nerve
activity in humans, elevating the baseline sympa thetic output. Hypoxia
appeared to be the primary stimulus for th is response. Wh ile sympathetic
nervous output to the spleen was not directly investiga ted, in light of Hoka and
9
An additional rhythmic contraction of the spleen, at least in dogs, is known to
occur even after denervation, and is likely a result of varia tions in blood flow to
the splenic circulation [41]. These rhythms, however, are not present during
contraction of the spleen caused by rebreath ing or catecholamine injection.
12
associates [68] previous research in dogs suggests th at a ltered chemoreflex control
may also affect spleen volume.
Additional axoaxonal innervation along the length of the splenic nerve may
adjust its ability to influence spleen contraction. Qua tacker and associates [96]
found adrenergic, post-ganglionic inputs on bovine splenic nerves, a llowing direct
(and most likely inh ibitory) modulation of the splenic nerve axons. Direction
associations between noradrenergic fibres of the sympa thetic nervous system and
lymphocytes in the spleen’s wh ite pulp provide a route for direct influence of t he
autonomic nervous system [97]. Wh i le these are postulated to serve as
interactions between the nervous system and the spleen’s immune functions, it is
not improbable th a t collateral influence on the spleen’s contractile functions
could also occur.
Effects of the spleen on cardiovascular function
Herman and associates [88] found in dogs th at splenic afferent nerve activity
followed unmyelinated C-f ibres, a relative ly slow-firing nerve type th at lends
itself to tonic firing rates and over longer durations. Perh aps more interestingly
they were the f irst to demonstrate low-pressure baroreceptors in the spleen’s
venous side th a t responded to increases in spleen venous blood pressure. These
receptors in turn appeared to produce reflex a lterations in cardiopulmonary and
renal sympathetic nerve activity via spinal sympath etic pathways, resulting in
increases in heart rate and blood pressure. The existence of such spleen receptors
was la ter confirmed in cats by Ca laresu and associates [98], and when stimulated,
resulted in increased splenic efferent nervous activity and venous pressure and
furthermore elicited reflex cardiovascular responses. These reflex responses were
ach ieved through stimulation of the receptors by distension, bradykinin and
capsaicin, but not through contraction of the spleen by norepinephrine injection.
Both the Herman and Calaresu groups suggested th at stretch ing of the spleen, but
not contraction of the spleen, resulted in the reflex cardiovascular adjustments.
However, there was disagreement amongst these groups about the presence of
myelinated fibres in the splenic nerve, someth i ng tha t Ca laresu’s group
appeared to confirm.
Furthering the issue of baroreception in the spleen, Deng and Kaufman [99]
h igh ligh ted the effects of splenic control of arteria l pressure, where they found
th a t intrasplenic nitric oxide reflexly a ltered MAP, an effect th a t could be
abolished by splenic denervation. Further research by the same group [100]
showed th a t splenic venous pressure stimulated splenic afferent nerve activity
and activa ted a splenorenal reflex, wh ich might account for th is effect. Thus it
13
appears th a t the spleen, a long with the liver and kidneys, is integral ly
involved in cardiovascular homeostasis via afferent nervous activity.
AIM OF THE THESIS
Wit h such diverse roles indicated for it, and with such integrated physiologica l
mechanisms associated with it, it is surprising th at the spleen has not received
more recognition as a useful, functional organ. Th is may indeed be due to t he
elusiveness of identif iable mechanisms behind its various functions, and its
rela tively newly found applications for the hea lthy human.
One of these recently established applications is the spleen’s ability to contract
in certa in situations to inject a stored amount of red blood cells [48, 49, 101]. In
some non-human species, especia lly mammalian expert runners and divers, the
involvement of sympathetic pathways in th is contraction is fa irly established
and similar processes may be at work in humans as well. Wh a t is not we ll
understood is the range of situations in which spleen contraction may occur:
besides voluntary apnea and hard physica l work, there are a number of
situations where an increase in blood gas transportation capacity would be
beneficia l. The “triggers” of the spleen contraction a lso remain to be clarif ied.
Nor is it clear wh ich groups might draw benefits from the effects of spleen
contraction, or to wha t extent. The ability to alter, adapt or maximize the
contraction of the spleen to obta in greater effects for the individual is a lso
unclear.
Th is thesis addresses these issues through a series of experimenta l studies a imed
a t better understanding the triggers, effects, and potentia l uses of the spleen’s
hematological contributions in different environmental conditions and activities.
METHODS
Measurement of blood variables
Venous capillarisation for Hb/Hct
The main hematological measures used in experimenta tion were Hb (reported in
g/L) and Hct (reported in % blood volume, BV). Th ese measures were obta ined
through closed-system antecubita l venous sampling in a ll studies, although in
14
the study of tra ined divers, skiers and untra ined humans [102] they were also
obta ined through capillary puncture.
Hb is normally the most important factor in determining blood O2 content, and
accordingly the h igher the Hb, the greater the O2 content of blood at a given PO2 .
Normal values are 130-180 g/L for males and 120-160 g/L for females. Hct or
packed cell volume (PCV) measures the proportion of blood volume th a t is
occupied by red blood cells, and is a lso an important determinant of blood
viscosity and microcirculatory flow changes. Hct normal va lues (±SD) are 45±7%
BV for males and 42±5% BV for females.
Hb was measured via automated blood analyzer using the cyanomethemoglobin
method, where a lysing agent is added to a sample of diluted blood, disrupting
the red cells in the sample and releasing the hemoglobin into the fluid. The
hemoglobin is converted to cyanomethemoglobin and the concentration is read by
a spectrophotometer and calculated from the optica l density of the solution. Hct
is calculated by multiply ing the red cell count and the mean red cell volume, both
of wh ich were measured directly by automated blood analyzer. Studies th a t
analyzed blood samples in-house used the Micros 60 automated blood analyzer
(ABX Diagnostics, Montpellier, France) wh ile studies th at sent blood samples to
hospita l for analysis used a similar apparatus with the same measurement
mechanism.
Hemoglobin concentration and red blood cell mass are independent of blood
volume, and neither is associated with spleen size in mammalian hypobaric
hypoxic responses [103].
Measurement of spleen size
Spleen size was measured in four studies in the current thesis (Papers I, III, V and
VI). During spleen measurements, triaxia l ultrasound measurements were
obta ined via sta tionary ultrasound apparatus (HDI 3000, ATL A, Ph ilips
Company, Bothwell, WA, USA) or via portable ultrasound apparatus (Mindray
DP-6600, Shenzhen Mindray Bio-Medical Electronics Co., Ltd., S henzhen,
Ch ina). These measurements were then used as a basis for spleen volume
calculations, using a formula developed by S. Pilström: L!(WT-T2 )/3, where (L)
is spleen length, (W) width and (T) th ickness. The formula is based on the
average shape of the spleen observed on the ultrasonic images and describes t he
difference between two ellipsoids divided by two. This formula was compared to
one previously used by Breiman and associates [104] and Sonmez and associates
[105], where volume is calculated by spleen length x width x th ickness x 0.523,
15
and the two methods on average differed by less th an 4 mL, or approximately 1%
of the measured volume.
Ultrasound measurement of spleen volume has demonstrated low intra- and
interobserver variability as well as excellent correla tion with computerized
tomography [106, 107]. In vivo ultrasound measurements of spleen volume have
been shown to correla te well with the weight [108] and volume [109] of the
resected spleen.
Measurement of other parameters
Arterial oxygen saturation
Arteria l oxygen saturation (Sa O2 ) measures the percentage of hemoglobin binding
sites in the arteria l bloodstream occupied by O2 . This measurement was, in the
works of th is thesis, obtained through pulse oximetry, a lthough it can be
obta ined via other methods as well. Pulse oximetry uses a light sensor containing
two sources of light (red and infrared) th at are absorbed by hemoglobin and
transmitted through tissues to a photodetector. The amount of light transmitted
through the tissue is then converted to a digita l va lue representing the
percentage of hemoglobin saturated with O2 .
Pulse oximetry h as been shown to have a slightly positive bias in critica lly i l l
patients, whereas in healthy patients the magnitude of differences between
pulse oximetry Sa O2 readings and Sa O2 va lues obtained from blood samples are
small [110] and should be considered reliable. The averaging window for the raw
data logged from the pulse oximeters used in the current studies (Biox 3700e,
Ohmeda, Madison, WI, USA and WristOx 3100, Nonin Medica l Inc., Plymouth,
Minnesota, USA) was 2 seconds. Heart rate was a lso measured simultaneously
from the same mach ines.
Normal S a O2 values are 97% to 99% in the hea lthy individual, whereas 90% or
lower is defined as clinical hypoxemia [3]. Using the oxyhemoglobin dissociation
curve, an O2 saturation value of 90% is generally equa ted with a Pa O2 of 60 mm
Hg, depending on e.g. the PaCO2 .
The rate of arteria l oxyhemoglobin desaturation during apnea is correla ted to
lung gas volume at apnea onset, resting O2 consumption, and pre-apneic S a O2 [111].
The depth of desaturation is dependent upon the dura tion of the apnea and the
rate of desaturation, however these two items do not a lways appear to be
16
correla ted to each other [111]. Centra l venous blood O2 levels at apnea onset pla y
a major role in determining the rate of desaturation, however [111].
Mean arterial pressure
The mean arteria l pressure (MAP), or the average arteria l pressure during a
single cardiac cycle, depicts the average blood pressure of the individual,
according to the following equation at normal or low heart rates: MAP = DP + 1/3
(SP-DP), where SP and DP are systolic and diastolic pressures, respectively .
MAP is a lso considered to be the perfusion pressure experienced by organs in the
body, including the spleen.
Cardiac output and systemic vascular resistance affect MAP, wh ich is to say th a t
everyth ing th at affects these two parameters also affects MAP i.e. stroke volume
is affected by ventricular preload, venous compliance, and blood volume, thus
MAP as well.
In those studies where it was measured, MAP was measured continuously from the
finger via automated sphygmanometer (Finapres 2300, Ohmeda, Madison, WI ,
USA). Th is non-invasive device has shown good correla tion with invasive (i.e.
intra-radia l) measurements of MAP not occurring during anaesthesia [112], and
a lso for measurement of MAP according to the aforementioned equation at rest or
during light exercise [113]. Reliability of continuous MAP measurement from the
finger can be influenced by changes in limb posture (unpublished observations)
and limb cooling [114]. In our studies these two items were held constant and thus
should not have affected recording reliability. Th e Finapres does, however,
h ave a tendency to underestimate MAP during extreme periphera l
vasoconstriction [115] and thus it is recommended to monitor perip hera l
vasoconstriction simultaneously, using skin blood flow measurements, for instance.
Skin blood flow
Sk in blood flow (SkBF) measurements in the works included in th is thesis were
intended to represent peripheral vasoconstriction status. Vascular factors such as
nitric oxide, endothelin, and prostacyclin all h ave influences on vessel diameter,
as well as tissue factors such as adenosine, K+, h istamines, H+ etc. However,
myogenic mechanisms th a t, intrinsic to vascular smooth muscle, a lter vesse l
diameter are the intended parameter for measurement in skin blood flow
recordings. These a lterations are primarily regulated by arteria l baroreceptors,
and to a lesser extent, chemoreceptors via neurohumoral pathways, wh ich
further affect venous compliance, systemic vascular resistance, heart rate, and
17
stroke volume. Thus SkBF is almost certa inly integrally rela ted to effects on
MAP.
SkBF in the works of th is thesis was measured via laser Doppler flowmetry
(Perif lux System 5000, Perimed, Järfä lla, Sweden). The laser Doppler technique
measures blood flow in the very small blood vessels of the microvasculature,
primarily the low-speed flows in capillaries close to the skin surface. The
technique depends on the Doppler principle whereby low power light from a
laser is scattered by moving red blood cells and as a consequence is frequency
broadened, and then sent to the mach ine via f ibre-optic probe. In the current
studies, the laser-Doppler probe was media lly attached to the first joint of t he
thumb. Continuous recordings were made throughout the duration of the
experimenta l session.
Of a ll the recorded parameters in the studies in th is thesis, however, it is also
most unstable due to its substantia l temporal and spatia l variability .
Consequently, the comparatively small sampling volume (about 1 mm3) of fibreoptic-based laser-Doppler flowmetry instruments constitutes one of the main
limita tions of th is technology: with the spatia l variability in tissue blood
perfusion, gross differences in perfusion readings may appear at recordings from
adjacent sites. Temporal variability, however, is recorded effectively from
single sites, and thus gives a good picture of changes in perip hera l
vasoconstriction.
Respiration and gases
In the studies where it was measured, the mech anica l characteristics of
breath ing before, between and after apneas was measured via lab-developed
chest bellows, placed around the circumference of the thorax just below the
pectoralis major muscles. The pneumatic chest bellows produce a pressure change
with every chest movement th a t is then amplif ied and converted to a digit a l
signal for online recording. The chest bellows were used during apnea to designate
the onset of involuntary breath ing movements, often referred to as the
physiologica l break ing point of the apnea. They were also used for comparison of
respiration during non-apneic periods and could thus provide a gross measure of
ventilation rate and depth.
Some of the studies included in th is thesis also measured inspired and expired
percentages of O2 and CO2 via gas analyzer (Normocap Oxy, Datex-Ohmeda,
Helsinki, Finland). One of the main intentions with the measurement of expired
CO2 pre-apnea was to ensure primarily th a t its level was equivalent for repeated
18
apneas, thus ensuring th at hyperventila tion did not occur, or in the cases where it
was included in the protocol th at hyperventila tion occurred to a similar level in
each apnea. Similarly, inspired and expired O2 levels are measured in Papers I V
and V to confirm tha t any hypoxic influence was absent when it was not wanted.
Expired CO2 post-apnea a lso provided an indication of CO2 production and
transfer to the a lveoli during apnea.
Apnea and hypoxia experimental protocols
Apnea series
The experimenta l protocol for apnea in all of the current studies except Paper II I
followed essentia l ly the same steps. Subjects began with a rest period, typica l ly
20 minutes, in a prone or supine position (depending on the study) to ensure blood
mixing and stabilization of the transcapillary fluid exchange [116]. During th is
time a ll required equipment probes were placed, as well as catheterization of the
arm when venous blood sampling was required.
All apneas were preceded by a 2-minute countdown. During th is time before each
apnea, the “control” period was established, typica lly a 30s or 60s window,
which ended a minimum of 30s before the onset of apnea. Th is was used in later
analyses to denote changes occurring due to apnea, and also for comparison
between subsequent apneas i.e. recovery characteristics. Three apneas were
standard in the current studies, each spaced by 2 minutes. A nose clip was applied
in the fina l minute before apnea onset, and a mouthpiece or mask, dependent on
whether the protocol calls for pre-apneic inspiration of specia l gas mixtures, was
placed for measurement of inspired and expired gases. The subject determined the
onset of apnea at or near the end of an audible 10s countdown demarcating the end
of the 2-minute countdown period. Each apnea was preceded by a full
exha la tion, followed by a deep but not maximal inspiration, as instructed to f a l l
in the range of 85% range of vita l capacity. Self selection of the f inal inha la tion
depth h as been shown to yield reliable and repeatable results in our numerous
studies following the same protocol (see Schagata y et a l, 2001, Table 2) and
mainta ins subject comfort. If pre-apneic hyperventila t ion was required, the
subject was given audible feedback on rate and depth of breath ing by the
experimenter during the hyperventilation period, but the apnea began in the
same manner as in non-hyperventila tion tria ls.
The provision of time cues during apnea was dependent upon the protocol, but
maximum duration apneas typica lly occurred without time cues. Upon completion
of apneas, subjects expired fully through the mouthpiece or mask and then
resumed normal breath ing.
19
An online data acquisition system (BioPac MH100A CE multi-channel da t a
acquisition system, BioPac Systems Inc., Goleta, CA, USA) recorded some or all of
the following parameters continuously from the start of the first 2-minute
countdown until the end of the protocol: Sa O2 , HR, MAP, SkBF, O2 , CO2 , breath ing
movements from the chest bellows, and apnea time.
Inspired gases and hyperventilation
In some of the experiments included in th is thesis, inspired gases and breath ing
rates were a ltered in order to mainta in a specific physiologica l state or
environment. These experimental interventions included:
1) the use of 100% inspired O2 . Th is intervention was employed in Papers IV
and V to remove any hypoxic stimulus th at would normally result from
apnea. The intervention was achieved by providing medica l O2 through
a non-rebreath ing mask at 5-10L/minute.
2) the use of 12.8% inspired O2 . This intervention was employed in Paper III
to induce a hypoxic stimulus tha t was not present when breath ing normal
a ir. Th is intervention was achieved via hypoxicator, which is a
mach ine with a metabolic filter th a t replaces O2 with nitrogen without
changing the ambient pressure.
3) the use of 5% inspired CO2 . This intervention was employed in Paper V to
induce a greater hypercapnic effect during apnea. This intervention was
ach ieved through a mixture of CO2 in medica l O2 and via nonrebreath ing mask at 5-10L/minute.
4) the use of hyperventilation. This intervention is employed in Paper V to
reduce the body’s CO2 stores prior to apnea, to reduce the effect of
hypercapnia during apnea. Th is intervention is achieved by advising the
subject on increased rate and depth of breath ing according to online
feedback of breath ing movement and expired CO2 percentage, for 1 min
prior to each apnea.
Statistical approaches
Publication statistical analysis method
All experiments in the current study were controlled tria ls i.e. subjects were
measured at least once both pre- and post-treatment, and can also be considered
crossover tria ls, i.e. a ll subjects underwent all control and experimenta l
treatments. The final outcome statistic is thus the mean change between
treatments or from control. Sta tistica l approaches vary amongst the studies in
the current thesis dependent on the analysis required, however there are some
common methods between them th at can be described.
20
1) Intra- individual comparisons: these comparisons denote any changes
th a t occur with in a single tria l or between tria ls of each individual. For
example, a comparison of hemoglobin increase might occur for one
individual between their control sample taken 2 minutes prior to the first
apnea, and immediately following all subsequent apneas. Another
example would include the comparison of hemoglobin increase between a
tria l with inspired O2 and one without. Paired t-tests are used in all
comparisons with Bonferroni corrections for multiple comparisons when
applicable. The level accepted for statistically significant change is
P"0.05. Non-significant trends are denoted for P<0.1 but >0.05.
2) Inter-individual comparisons: these comparisons denote any changes
between groups of individuals. For example, a comparison of hemoglobin
increase during a tria l amongst trained divers might be compared with
increases amongst untrained individuals completing the same tria l.
Control va lues are also similarly compared between groups as well.
Unpaired Student t-tests (with corrections for multiple comparisons,
when appropria te) are used in these comparisons and the level accepted
for statistically significant change is P"0.05. Non-significant trends are
denoted for P<0.1 but >0.05.
Inter-individual comparisons of control values were usually compared as absolute
va lues i.e. using the units and scales in which the va lues themselves are
obta ined. Intra- individual comparisons on the other h and used relative values i.e.
percentage changes from control va lues where the established control va lue
represents zero change. In th is manner, subjects served as their own controls and
the non-uniformity of individual residuals was reduced.
Analyses of correla tions between various parameters e.g. spleen size and diving
performance were achieved using Pearson’s product-moment correla tions. The
level accepted in correla tion analyses for presence of significant change is P"0.05.
Additional statistical analysis method
An additional sta tistica l analysis method (a priori or post-hoc, depending on the
study) has been used in all experiments in the current thesis in addition to the
methods listed above or in the publications, regardless of whether th is second
method has been included in the published materia l or not. These analyses are
a lso based on paired t-sta tistics for crossover experimenta l design, but with some
devia tion from the above descriptions.
a) Firstly, the “rela tive” va lues are brought forth using log transformation
using the natural logs of the va lues of the variable – thus removing the
21
non-uniformity of the residuals entirely. Standard deviations in these
cases are a percent varia tion or coefficient of varia tion.
b) Secondly, the level accepted for presence of significant change is not
determined via P-value, but rather on Cohen’s thresholds for clinical
significance [117]. Based upon the smallest clinically relevant values of
the effect in question, the Cohen’s thresholds attribute size
characteristics to changes i.e. small (0.2-0.5), moderate (0.5-0.8), large
(0.8-1.2), very large (1.2 and greater), and can then be extrapola ted into a
percent probability th at the change in measured parameter is indeed
clinica lly signif icant. In the case of Hb and Hct, the smallest significant
change was deemed to be 2% and 0.5% respectively, based upon initia l
measures of normal responses of average subjects [78, 118] and clinica l
guidelines [119].
A relevant situation might be as follows: A group of subjects demonstrates a 2.3%
increase in Hb after 3 breath- holds. The P-value for the pa ired t-test comparison
between control and post-apnea Hb is 0.06, which according to our set criteria can
only conclude th a t there was no statistica lly significant change. The Cohen’s
threshold is 0.24, and indicates a small but clinicall y significant increase in Hb,
with a 90% confidence interva l of ±0.8% i.e. there is a 90% chance th a t the true
Hb increase is between 1.5% and 3.1%.
In th is case of the P-va lue, it must be concluded th at no change has occurred, as
the P-value fa lls outside the accepted level of statistica l significance. However,
a 2.3% change in Hb should not be considered as zero change. Indeed, a 2%
increase in Hb would by many physiologica l standards (and the author) be
considered a vita l improvement. The Cohen’s threshold illustrates th is, and the
confidence interva l gives an estimation of the uncertainty beh ind the statistica l
analysis.
Th is method is, unfortunately, not yet widely accepted by journal reviewers and
editors, and therefore does not constitute the majority of published statistica l
analyses. However, considering th a t statistical significance focuses on the
mathematica l null value of the effect, wh ile clinical significance provides an
measure of probability th a t any effect might h ave in terms of benefiting or
h indering the individual e.g. improving or reducing performance, it can be argued
th a t the latter approach is considerably more valuable and interesting to both
physiologists and experimenta l subjects. For th is reason, th is second method was
used for comparison with the “traditional” sta tistical analyses. Any devia tions
or differences between the two approaches are reported in the results section of
th is thesis, otherwise in can be assumed tha t the two methods are in agreement.
22
SUMMARY OF RESULTS
Occurrence of spleen contraction
PAPER I: Correlation between spleen size and hematocrit during apnea in
humans.
Main conclusion: Spleen contraction and circulating erythrocyte increase
develop in close association, with the latter arising primarily due to spleen
contraction.
In th is study, which a imed to further investigate the rela tion between spleen
volume, Hb and Hct during seria l apneas, 3 male subjects and 1 female subject
performed 3 maximal-duration apneas spaced by 2-minute pauses. Ultrasound
measurements revea led a mean reduction in spleen volume of 34% (p<0.01), from
338 ± 93 ml before apnea to 223 ± 66 ml after 3 apneas, wh ile simultaneous
increases in Hb and Hct from control values of 2.5 ±1.5 % and 3.1 ± 2.8%,
respectively (p<0.05 and p=0.07) were observed. The correla tion coefficient mean
(with 90% likely range va lues) between spleen volume and Hb was -0.88 (-1.0 to 0.26), and between spleen volume and Hct the correla tion coefficient was -0.85 (0.99 to -0.37). Splenic volume, Hb and Hct h ad returned to pre-apnea levels
with in ten minutes after the th ird apnea. Based on 80% splenic hematocrit
estimation, and blood volumes of 8% body weight, the rela tive contribution of
the observed spleen shrinkage to change in Hct was ca lculated to be 69%. It was
concluded th a t spleen contraction and Hb and Hct increases develop in close
association, with the erythrocyte increase arising due primarily to the spleen
contraction.
Two additional male subjects were added to the study after publication of Paper
I, which confirmed the findings with four subjects (Figure 2). Spleen volume
decreased 29% (p<0.01) from 353 ± 101 ml before apnea to 251 ± 76 ml after 3
apneas, while simultaneous increases in Hb and Hct from control va lues of 3.8 ±
2.3% and 4.3 ± 2.2%, respectively (both p<0.05) were observed. Splenic volume,
Hb and Hct had returned to pre-apnea levels with in ten minutes after the th ird
apnea. The correlation coefficient means between spleen volume and Hb was 0.81, and -0.82 between spleen volume and Hct (p<0.05).
23
Figure 2. Mean spleen volume (mL) and Hb (g/L) before apneas (Control), after
Apnea 1 and Apnea 3, and during the 10 minutes following Apnea 3. N=6.
Correlation between spleen volume and Hb is -0.81 (p<0.05)
PAPER II: Increase of hemoglobin concentration after maximal apneas in divers,
skiers and untrained humans.
Main conclusion: Non-diving athletes, untrained subjects and apneic divers all
respond with spleen related Hb increases during apnea. Trained divers have a
stronger response.
A comparison of 29 male and 4 female tra ined apneists, 13 male elite crosscountry skiers, and 23 male and 9 female untra ined individuals showed th a t the
apneists tended to have a h igher baseline Hb (150.1 g/L) th an the other groups
(145.5 g/L and 146.9 g/L, respectively). All groups showed an increase in Hb after
3 apneas, but the largest relative increase was in divers (2.7%) compared to the
other groups (2.1% and 1.4%, respectively). By 10 minutes after apneas, a l l
groups had returned to baseline Hb va lues.
The apneic duration for the first apnea was a lso longest in the divers (144s)
compared to the skiers and untra ined individuals (87s and 94s, respectively), and
while a ll groups increased the ir apnea duration across the series, the divers
showed the largest increase (20%, compared to 12.2% and 15.5%, respectively).
24
There was no correla tion between individual apneic duration and increase in Hb
over the series in any group.
PAPER III. Short-term effects of normobaric hypoxia on the human spleen.
Main conclusion: Eupneic hypoxia initiates spleen contraction and Hb increase,
suggesting a role of spleen contraction for Hb elevation in the early phase of
altitude acclimatization.
A pilot study at a ltitude was completed with three male subjects [120] revea ling
th a t the Hb increase normally associated with apnea was reduced and
eventually diminished at 5000 m.
In a following laboratory study using 20 min exposure to a low-O2 (12.8%) but
normocapnic environment, 4 males and 1 female untrained in apnea showed a
reduction of Sa O2 to 64.4% and a simultaneous decrease of spleen volume by 17.6%.
Hb and Hct simultaneously increased by 2.1% and 2.1% respectively, and a l l
parameters reverted to normal with in 10 minutes from removal of the hypoxic
environment. Negative correlations between spleen volume and both Hb and Hct
(r=-0.51 and r=-0.42, respectively) were found.
Initiation of spleen contraction
PAPER IV: Hypoxia augments apnea-induced increase in hemoglobin
concentration and hematocrit.
Main conclusion: Hypoxia augments spleen contraction during apnea, but there
is some residual contraction even without hypoxia indicating that other factors
are involved
After 3 apneas in oxygen and without desaturation, 10 male untrained subjects
showed lesser increases in Hb and Hct (2.0% and 2.0% respectively) th an when
they performed apneas of the same duration in air with desaturation to 84.7%
(3.1 and 3.0%, respectively). Expired CO2 after apneas was identica l in both
tria ls. The Hb values returned to normal with in 10 minutes after the fina l apnea.
There were no differences in the diving response between the tria ls.
Secondary statistical analysis using Cohen’s thresholds calculated th a t in the
a ir group, Hb and Hct were stil l elevated by a small amount after 10 minutes,
with 63% and 55% likeli hood, respectively.
25
PAPER V: Hypercapnia moderates hemoglobin increases during apnea.
Main conclusion: Hypercapnia during apnea causes increases in hemoglobin and
hematocrit in the absence of hypoxia, but not solely through breathing high
levels of carbon dioxide.
Four male and four female subjects conducted four separate 3-apnea series, each in
hyperoxia but with varying levels of CO2 – a series 1) with pre-apneic
hypercapnic inha la tion, 2) after normocapnic inha la tion, and 3) at hypocapnia
ach ieved via hyperventila tion. In addition three test-periods were performed 4)
without apnea but breath ing 5% CO2 in O2 for an equiva lent duration as the
apneas in the other tria ls. Hb increased most with pre-apneic hypercapnic
inh a la tion (by 4.0%) th an in the normocapnic, hypocapnic, and eupneic tria ls
(wh ich h ad Hb increases of 3.0%, 1.8% and 0.8%, respectively). The
physiologica l break ing point of apnea was shortest in the hypercapnia tria l,
while at hypocapnia it was the longest. Expired CO2 followed a similar pattern
as the Hb increases – greatest in the hypercapnia tria l, and lowest in the nonapneic tria l. No desaturation occurred in any tria l.
Three subjects had spleen measurement completed during all of the tria ls. The
change in spleen size after apnea 3 for the hypercapnic, normocapnic, hypocapnic
and eupneic tria ls was -32.5%, -9.4%, 30.4% and 13.0% from control, respectively
(Figure 3 a-d). Ten minutes following the final apnea, spleen size showed a return
to control va lues in all tria ls.
Secondary statistica l analysis using Cohen’s thresholds calculated th a t t he
increase in Hb and Hct in the hypercapnia tria l was greater th an the
hyperventila tion and eupnic tria ls by a large amount (94% and 97% likeli hood,
respectively), but only by a small amount in comparison with to the normocapnia
tria l (53% likeli hood). Hct in the hypercapnia tria l was greater by a large
amount compared to the hyperventilation and eupnea tria ls (87% and 89%
like li hood, respectively). It was lower th an in the normocapnia tria l by a sma l l
amount (53% likeli hood), however.
26
Figure 3 a-d. Mean spleen volume (mL) and Hb (g/L) before apneas (Control),
immediate ly following Apnea 1 and Apnea 3, and 10 minutes following Apnea 3,
for hypercapnic, normocapnic, hypocapnic, and eupneic series. N=3.
Consequences and applications of spleen contraction
PAPER VI: Spleen and lung volumes correlate with performance in elite apnea
divers.
Main conclusion: The size of the spleen correlated with competitive apnea
performance in elite divers. This study suggests that the spleen volume as well
as the lung volume may predict apnea performance.
In studying 14 male elite apneists competing at a world championship
competition, it was found th a t their spleen volume (both absolute and corrected
for anthropometric data) was correla ted with their combined three-discipline
competition score, but not with the ir he ight or weigh t. Their competition score
was not correla ted with height, weight, hours of tra ining or BMI, either. Lung
volume was also correla ted with performance.
27
The three h ighest scoring divers at the competition had a mean spleen volume of
538 ml, the largest volumes measured amongst the competitors. They a lso
appeared to have a very large contraction during a single 2-minute apnea,
showing a volume reduction to 258 ml. In contrast, the three lowest scoring divers
h ad a mean spleen volume of 270 ml. The extra apneic duration allowed by t he
spleen contraction of the 3 best divers was estimated to be 30s.
PAPER VII: Cardiovascular and haematological adjustments to apneic diving in
humans: Is the spleen response part of the diving response?
Main conclusion: The spleen response and the cardiovascular diving response to
apnea are not parts of the same response, and should be considered as
separately occurring entities.
W h ile the apnea-induced cardiovascular diving response and spleen contraction
both h ave dive-prolonging effects, these two phenomena should not be considered
as functionally connected parts of a single general response to apnea. The increase
in Hb and Hct from spleen contraction is not augmented by facia l ch il ling, and
does not develop fully during a single apnea, but rather over a series of apneas,
and is not reversed until 10 minutes without apnea. The cardiovascular diving
response, in contrast, is fully developed with in 30 s during each apnea, recovers
rapidly following apnea, and is also stronger with fa cia l ch ill ing. The effects of
these two phenomena are a lso realized through different mechanisms – the
diving response acts through increased O2 conservation, while the spleen response
exerts its influence through increased O2 storage and CO2 buffering capacity.
DISCUSSION
Occurrence of spleen contraction
The results presented in th is thesis clearly show th a t humans respond wit h
elevations in Hct and Hb during apnea, and these eleva tions correlate close ly
with splenic contraction. This response is a lso active during normobaric hypoxia ,
which may expla in the early elevation of Hb and Hct seen at altitude exposure.
The contribution of the spleen to these hematologica l changes is calculated to be
as high as 69%, wh ile the remaining contributions may possibly arise from other
pooled blood stores with h igh hematocrit in the abdomen, such as the liver [75],
or from plasma extravasation [44].
28
The presence of spleen-rela ted hematological increases in apnea divers is a
finding similar to th at of Hurford and associates [48]. However, the work in th is
thesis described significant levels of spleen contraction even in individuals wit h
no experience in apnea, in accordance with the findings of Schagatay et a l [50],
Espersen et al [77] and Bakovic et al [76]. However, the work in th is thesis a lso
demonstrated th a t the spleen response was more pronounced in apnea divers, not
only in comparison to sedentary subjects, but a lso elite aerobic endurance ath letes.
Th is suggests either a specif ic apnea training effect on the spleen, or
predisposition.
The novel finding th a t simulated a ltitude by normobaric hypoxia induces spleen
contraction is interesting as it suggests a role of the spleen in the Hb elevation
often observed in the early phase of a ltitude exposure. Sonmez et al [105] had
previously noted a reduction in spleen size in humans after prolonged residence a t
h igh a ltitude, but such a response would perh aps be even more beneficia l when
compensating for the acute effects of a transition to a lower O2 environment.
Identif ication of these situations and groups in which spleen contraction occurs
provided a background for further investigation of the mechanisms underlying
the response pattern.
Initiation of spleen contraction
Tha t both hypoxia and hypercapnia play a role in triggering spleen contraction
are a lso major novel f indings. Hypoxia was found to cause spleen contraction
regardless of breath ing state – both during apnea and normal ventila tion the
spleen responded to hypoxia. Hypercapnia, on the contrary, appeared to only
play a role during apneic states; normal breath ing of CO2 , at least under shorter
periods, does not appear to cause spleen contraction at a ll. It is, however, a
residual effect of apnea without excessive hyperventila t ion, and thus underlies
essentia lly a ll apneic periods even when hypoxia is eliminated. Combined,
hypoxia and hypercapnia may even ach ieve a simila r contractile state as either
can alone, suggesting th at there may be a threshold level of contraction i.e. t he
spleen becomes fully contracted by a certa in amount of either stimulus and no
further contraction can be obtained with additional stimulus load.
Tha t hypoxia and hypercapnia play a role in spleen contraction during apnea
infers th a t some form of chemoreception must a lso be involved in initia ting the
contraction process. Wh ile no direct evidence of the input of these physiologica l
systems can be derived from the current studies, the possible cha in of events from
stimulus detection to effect is a worthwh ile discussion point.
29
Stimulus detection by chemoreception
The body has at least two h igh ly sensitive receptor systems th a t initia te
system-wide nervous feedback during changes in CO2 and O2 respectively: t he
central chemoreceptors located in the pontomedullary region of the brain which
are stimulated by pH changes caused by diffusion of CO2 into the cerebrospina l
fluid, and the carotid bodies, themselves only a few micrograms in weigh t,
which are stimulated by changes in PO2 in the pla sma (though CO2 may a lso
h ave minor effects here as well, in the order of 10-20% [121]). Comparisons of the
speed of response and relative gains of these two systems have been studied for
decades and are still embroiled in controversy.
Hypoxia augments the sensory discharge of the carotid bodies before it reaches
the centra l nervous system, and then triggers the CNS to initia te autonomic
changes such as increased breath ing rate and blood pressure. Although there is
debate about the mechanisms behind arteria l PO2 sensing, it is currently thought
th a t activa tion of the afferent nerve ending in the carotid body is responsible for
detecting even modest changes (i.e. PO2 of 100 to 80 mmHg) with in seconds, vi a
ion transporters, mitochrondria l cytochromes or a combination of the two [122].
Cellular responses in the rest of the body often require far more severe hypoxia
(ca 40 mmHg) before initiation, and take minutes to hours to develop. Thus it can
be logically suggested th a t apneic interventions resulting in hypoxia or acute
hypoxic environments stimulate the carotid bodies, wh ich in turn elicit
subsequent systemic effects. The results from the studies in th is thesis examining
the effects of hypoxia during apnea and during normobaric hypoxia follow th is
line of reasoning.
Changes in alveolar CO2 , on the other h and, are reflected in brain extracellular
fluid pH on a time course consistent with circulation time i.e. a few seconds [123].
Nonetheless, when peripheral receptors are bypassed, ventila tory response is
about 58% slower in dogs [124]. The same research in dogs also demonstrates wide
individual varia tion in the rela tive contribution of central vs. periphera l
chemoreceptors to gain effects, a lthough centra l reception appears dominant i.e.
approximately 60%. Th is holds true in other studies of dogs [125] th a t
experienced hypercapnia in hyperoxia, a method similar those used in Paper V
included in th is thesis.
Earlier research demonstrated the cross-influence of hypoxia and hypercapnia
[126]. For example, the h igher the PCO2 , the greater the hypoxic sensitivity of
the carotid body [127]. Along the same lines, low O2 potentia tes the ventila tory
30
response to hypercapnia. The carotid body chemoreceptors are the site of
hypoxic-hypercapnic interaction [128] and also control the increased ventila tory
response post-apnea. Indeed, these receptors appear to be essentia l for the
initia tion of involuntary apnea, for example, during sleep apnea and periodic
breath ing due to low CO2 [129]. Thus a major role for carotid body chemoreceptors
could be theorized in terms of spleen contraction when its initiation follows
chemoreceptive stimuli. CO2 elevation always develops during apnea, even in
hyperoxia, and thus stimulation of the chemoreceptor system is inevitable.
Th is is not the case, however, for normobaric hypoxia without apnea. Even
without hypercapnia or cessation of breath ing a notable spleen contraction and
Hb/Hct elevation is evident. Here it is hypoxia th a t is the sole stimulus, and it
produces a spleen contraction-related elevation in Hb similar to those atta inable
during the apnea tria ls without hypoxia (but not with hypoxia). It is logica l
th a t peripheral chemoreceptor stimulation is the primary system affected in
these cases. However, neither normobaric hypoxia, nor apnea without hypoxia
reached the same levels of Hb/Hct elevation tha t apnea with hypoxia did. Th is
provides further evidence for the synergistic effect of hypoxia and hypercapnia
in producing maximal spleen-rela ted Hb/Hct eleva tion. Indeed, centra l and
peripheral mechanisms may act in a redundant fash ion to reconfigure ventila tory
responses after intermittent exposure to altered blood gases, a finding which
Cummings and Wilson [130] found in their own studies in rats.
The co-release of transmitters at the carotid body may also signal an interaction
of O2 and CO2 . ATP appears to play a critica l role in subsequent nerve activa tion
by hypoxia [122]. However, acetylcholine has also been implicated as a corelease mechanism with ATP, and the release of th is neurotransmitter appears to
be facilita ted by CO2 . Blockade of ATP receptors at the carotid body also
appears to prevent sensory excita tion by CO2 [131].
Output following chemoreception
Triggering of the chemoreceptors then requires an effector signal, in an attempt to
adapt to the environmental conditions, a response in which spleen contraction is
like ly included. Considering th at in most mammals the splenic nerve is 98%
composed of sympathetic nerve fibres [132], and th a t the sympathetic nerve
terminals in the spleen store, take up, and release norepinephrine in response to
stimulation [133], it is likely th a t neuronally-derived catecholamine release is
the end-point in the cha in of events resulting in spleen contraction. Wh ile the
onset of th is could potentia l ly occur relatively quickly, as the nervous system
works at greater speeds th an systemic circulating hormonal or other chemica l
31
release acting upon local receptors, the actual contractile effect may a lso follow a
slower time pattern more dependent on the contractil e properties of the spleen’s
musculature and the progressive development of asphyxia.
W hether the nervous input exerts its effects independent of, or as a result of,
changes in chemoreception, is debatable. It is possible th a t splenic contraction
signals come directly from higher neural input. The ventrolateral medull a ,
which conta ins a group of neurons projecting to spinal sympathetic preganglionic
neurons, has been indicated in splenic excita tory responses [134]. Such input migh t
be expected to be responsible if cessation of breath ing were enough to trigger
medullary impulses for splenic contraction, in the absence of changes in blood
gases.
But the evidence for influence of chemosensitive mechanisms on neural input is
far more extensive. The idea th a t CO2 -sensitive neuronal mechanisms might be
involved in the generation of sympathetic tone is not new [135]. More recent
studies have shown th a t 30s of intermittent hypoxic apneas for 20 minutes [136]
and 20s of intermittent hypoxic apneas for 30 minutes [16] increase muscle
sympathetic nervous activity (MSNA), an alteration of chemoreflex control th a t
appears to be mediated by hypoxia. In the la tter study, MSNA remained
elevated for 30 minutes post-apneas, a finding which h as been noted in
obstructive sleep apnea patients. Though, other recent work [137] found th is not to
hold true in elite breath- hold divers’ resting MSNA va lues, suggesting th a t t he
effect of apneas on sympathetic outflow may not be chronic.
Work by O’Donnell and associates [14] showed th a t renal sympathetic nerve
activity increased during apnea and was correla ted to directly recorded carotid
chemoreceptor activity. Potentiation of the chemoreflexes a lso appears to occur
when the inh ibitory influence of stretch receptors is eliminated, as during apnea,
and th is potentia tion is greater on the peripheral side [138].
W h ich of O2 and CO2 has the greater influence on eventual neural output to t he
spleen is a lso discussible. Earlier work in sleep apnea patients showed th at 100%
O2 results in reductions in central sympathetic outflow [139]. Augmentation of
nerve activity during apnea h as been shown to be much greater with hypoxia
th an with hypercapnia [121]. In Paper IV in th is th esis, hypoxia caused spleen
contraction even in the absence of apnea. However, similarly large increases in
Hb and Hct occurred in hypercapnia without hypoxia, suggesting th a t the
chemoreceptive output from either trigger is equally great. Nonetheless, recent
studies in th is author’s laboratory a lso showed th a t a 2 min period of apnea
( hypercapnic hypoxia) induced a spleen volume reduction twice as large as in
32
eupneic normobaric hypoxia (normocapnic hypoxia) despite similar levels of
hypoxia (in manuscript).
There is a lso little argument th a t catecholamines are released systemical ly
during hypoxia, but whether these are responsible for the initia tion of spleen
contraction and/or the maintenance of spleen contraction remains to be seen.
Earlier research [140] showed th a t in both arteria l a nd primary tissue hypoxia
the sympathetic nerves played a more important role in circulatory responses
th an the adrenal medullary hormones, and th a t sympathetic nerve control a lone
produced more perip hera l vasoconstriction th an increased adrena l
catecholamine secretion.
The time scale for systemic catecholamine release is not as fast as neura l
conduction. Th is also holds true for local catechola mine output, such as in the
carotid bodies, wh ich release catecholamines in response to hypoxic insult.
However, it h as been shown th a t th is release is not required for hypoxia- induced
chemosensory excita tion in the carotid body, as it typica lly occurs 1-3 minutes
following the initia l chemoexcita tion [141]. Sonmez and associates [105] study
showed a persistent reduction in spleen volumes after chronic exposure to a ltitude
after individuals moved there from sea level. The increase in catecholaminergic
discharge at h igh a ltitude [142] may be one expla nation, whereby a low-O2
environment may continue to stimulate catecholamine secretion and subsequent
spleen contraction even over long periods.
It seems likely th a t a combination of neural input and catecholaminergic release
resulting from chemoreceptive stimuli initia tes and prolongs the splenic
contraction, respectively. Ojiri and associates [90] showed th a t !1-adrenergic
blockers prevented increases in Hb and Hct in resting dogs induced by splenic
nerve stimulation and epinephrine and norepinephrine infusion. In humans, !adrenoreceptor blockers abolished the vascular responses of the spleen to
norepinephrine and direct neural stimulation [85]. Thus it appears th at both
neural and catecholaminergic innervation are susceptible to blockade when !1receptors are antagonized. Thus both may affect the spleen, but perhaps on
separate time scales. Catecholamines circulating systemica lly h ave a h a lf- life
of about 2 minutes, whereas nervous impulses can be ceased immediately. The
delay in complete recovery of the spleen following apnea (up to 10 minutes)
despite remova l of chemoreceptive stimuli could be accounted for by residua l
systemic catecholamine presence.
33
Other potential output mechanisms
Tha t other basic autonomic mechanisms could lead to spleen contraction is also a
possibility. Apneas held at inspiratory capacity h ave been shown to cause
increases in muscle sympathetic nerve activity [143], which may be related to the
unloading of low-pressure baroreceptors located in the atria and pulmonary
veins. These baroreceptors respond to small changes in absolute venous pressure –
like ly to occur during breath holding at larger lung volumes – and may send
signals to neural regions responsible for control of sympathetic activity. Spleen
contraction may be affected as a result of the increased sympathetic output,
through mechanisms already named. Macefield and associates [144] used brain
imaging techniques to study exactly th is phenomenon during 40s apneas, but used
calculation methods designed to eliminate the effects of hypercapnia, making
theoretica l interpretation of the results diff icult in th e context of th is thesis.
Palada and associates [145] also indicated downloaded baroreflex signals as a
possible mechanism of splenic contraction during apnea at large, but not sma l l
lung volumes (although spleen contraction still occurred even at small volumes).
Their own previous work suggesting an active contractile response during apnea
suggests th a t the neural functions are stil l involved via sympathetic output, but
th a t baroreception, similar to chemoreception, may a lso be implicated.
Kaufman’s group have a lso explored similar avenues of spleen function, showing
th a t the spleen may play a role in tonic maintenance of blood pressure in norma l
physiologica l conditions [99], and th a t afferent splenic nerve activity is
increased by augmented splenic venous pressure – and not blood flow [100]. Their
findings lend weight to the possibility of local baroreception playing not only a
role in systemic blood pressure regulation, but even sympathetic nerve activity in
loca l circuits between the spleen and the liver, kidneys, and perhaps itself .
W hether such pressure changes may affect spleen contraction via these routes is
not certa in, but certa inly justifies further investigation.
Consequences and applications of spleen contraction
Another novel finding in th is thesis is the associati on between spleen size and
apneic performance during competitive apnea, indicating the beneficia l effects of
a large spleen on overall diving performance. Th is is perhaps the most
applicable finding of the thesis, whereby the maximization of spleen contraction
during breath holding is of particular va lue. This author has previously shown
34
[78] th at the level of spleen contraction varies greatly between individuals, and
Paper II and Paper VI in th is thesis suggest th a t trained apneists are among
those who the most powerful elevation of hema tocrit rela ted to spleen
contraction during apnea. There are thus severa l ways in which apnea tra ining
could lead to performance-boosting hematologica l effects.
The presence of h igher resting Hb levels in tra ined apneists a lso points to a
possible tra ining effect specif ic to breath holding, most likely due to long-term
effects. The levels of hypoxia obtained during voluntary (and often involuntary)
breath holding are typica lly greater th an can be experienced even at maxima l
levels of exercise [146] and in many cases equal or exceed those atta inable at h ig h
a ltitude. 10 Thus, regular exposure to hypoxia through breath- hold tra ining may
h ave any number of the following effects: an increased stimulus for
erythropoeisis, an increase in the ability to contract the spleen, and/or an
increase in spleen size and therein storage capacity. The erythropoietic effect is
like ly, considering th at th is author’s research identified increases in endogenous
EPO after a single apnea tra ining episode [147] (paper not included in th is
thesis), wh ich in other circumstances such as altitude tra ining has been shown to
lead to increased novel Hb production. The second effect a lso appears to be
confirmed in the f irst two studies of th is thesis. The fina l possible effect, spleen
growth, is not as certa in. In the study of the world ch ampionship divers, apneic
tra ining hours leading up to the competition did not correlate with spleen size or
performance. Spleen size can indeed change in individuals, as shown in studies of
splenic regeneration after splenectomy and by splenomegolamy, but th a t its size
can be directly affected by tra ining is still uncerta in.
Nonetheless, the spleen size in divers likely confers an advantage in terms of
breath holding time, and the same might be said for ath letes of any sort th a t
consider maximal O2 -carrying capacity a limiting factor in performance. A
greater ejection of red blood cells in the spleen leads to increased oxygen stores
and an improved CO2 buffering capacity.
In apneic performance
The most obvious applications of spleen contraction are probably in apneic
situations. Here, enhanced convective O2 transport through release of stored Hb
10
Unpublished observations. However, the author has personally experienced an arterial
oxyhemoglobin saturation level of 44% during breath holding, and is not considered a
trained apneist. Even at altitudes about 6000m he has never attained this level of
desaturation in non-apneic situations.
35
and Hct results in increased O2 carrying capacity, and an improvement in CO2
buffering via increased ava ilability of carboxyl group binding on Hb. The work
conta ined in th is thesis has shown th a t apneists with larger spleens hold the ir
breath for longer, and th a t the contracted spleen in such cases offers a theoretica l
advantage of 30s of additional breath holding time, based solely on O2 -carrying
criteria i.e. with no respect to effects on CO2 . In untrained individuals, th is time
extension was in the order of 18s, itself a considerable advantage when
considered in surviva l terms.
However, the transient increases in Hb and Hct conferred through spleen
contraction should not only be considered in terms of O2 -carrying capacity. Dane
and associates [148] have shown in dogs th at the autologous blood infusion from
the spleen also enhances diffusive O2 transport in the lung, mainta ining alveolar
O2 transport in the presence of ambient hypoxia. Although dogs have
considerably larger splenic infusions th an humans (in the order of 20% BV
increase in Hct), it is conceivable th at even the lesser increases in Hct observed in
our lab of up to 5% may have positive effects in th is regard. Transient Hct
increases of up to 10% have been noted in some individuals during th is author’s
work.
In acute adaptation to high altitude
W h ile spleen contraction may convey advantages at sea level, the new discovery
of effects in normobaric hypoxia suggests th a t it may a lso play a role in “fast”
adaptation to a ltitude or other ambient hypoxic conditions, serving as a
compensatory mechanism by bridging the gap between non-acclimatization and
a ltitude-induced polycythemia. The latter can be achieved with in days to
weeks depending on the severity of the hypoxia but on its own would not assist
the individual in the immediate short-term. An early increase in Hb during
a ltitude exposure has often been reported, but it h as been expla ined to be a side
effect of increased ventila tion or to a ltitude-induced hormonal changes [149].
Th is thesis h as added a functional explanation to th i s phenomenon, which could
perh aps also expla in individual differences in susceptibility to altitude sickness.
Dogs, with the ir significant splenic contributions, acclimatize with in minutes to
a ltitude a llowing them “to chase the ir prey up and down a mountain wit h
minimal changes in hypoxic acclimatization during the exertion” [148]. Though
the human contractile response, comparatively speaking, may or may not occur to
the same magnitude as in dogs, it h as been shown to be progressive wit h
increasing altitude [120]. This small but likely appreciable advantage may
reduce the tissue hypoxia until new erythrocytes can be produced.
36
During exercise
Tha t exercise induces spleen contraction is certa inly known, but the effects are
not. Previous research regarding blood manipulation by venesection or autologous
transfusion in human subjects has shown variable increases in O2 uptake in the
order of 10-13 % [150, 151], and following splenectomy, maximal O2 uptake is
reduced in dogs [148] and horses [152]. A yet unpublished study by th is author’s
research group has observed spleen contraction during maximal O2 uptake tests in
elite cross-country skiers, similar to the findings of Froelich and associates [75]
and Laub and associates [49]. In th is study, skiers performing apneas prior to the
maximal test were compared to a second performed tria l without prior apneas.
Both tria ls ach ieved the same end-test increase in Hb/Hct as well as VO2
kinetics, a lthough the apnea tria l h ad a “head start” regarding these
parameters in the early phase of the tria l. Theoretica lly, a spleen contraction
th a t was “pre-activated” in th is regard could serve to improve performance over
shorter events i.e. 2-3 minutes, as it likely requires a t least th is long to become
fully initiated. Further investigation of th is possibili ty is certa inly warranted. 11
Problems with Hct limits in competition
One issue th a t may arise from the autologous “blood doping” effects of spleen
contraction is the presence of periods of naturally-occurring, but higher th an
normal (i.e. resting) Hct and Hb. The effects of hypoxia, prior exercise, stress
(resulting in catecholamine release), apnea and even body position could a l l
produce a spleen-rela ted temporary increase in blood parameters th a t may pose
problems in, for example, doping controls in sporting events. In both cycling and
cross-country skiing, maximum Hct and/or Hb levels are delineated in order to
reduce the risk of cardiovascular complication due to h igh blood viscosity.
W h ile the spleen contributes to such increases in viscosity it a lso eliminates such
increases rapidly when they are not required, a llowing a more “risk-free”
increase in red blood cell content. Nonetheless, such a transient increase would not
be desired when doping controls are being performed. In fact, any ath lete could
possibly be at h igher risk of being detected when tested, if he or she is stressed or
nervous during the testing! It may be judicious to inform ath letes and officia ls of
the risk of such effects.
11
As an interesting although perhaps coincidental aside, the individual with the clearly
largest spleen response and apneic duration during this study, is currently making his mark
on the world stage in cross-country skiing – those with lesser responses have yet to achieve
the same level of performance.
37
Relation to the diving response
The studies in th is thesis suggest th at the spleen contraction seen during brea t h
holding is not an integral part of the classical human “diving response” i.e.
bradycardia and an increase in periphera l vasoconstriction. This is a topic
currently under debate, where Espersen and associates [77] cla imed th a t spleen
contraction “should be included in the diving response on equal terms wit h
bradycardia, decreased periphera l blood flow and increased blood pressure”.
Palada and associates [145] have a lso suggested th a t the onset of spleen
contraction is immediate and is facilita ted by facia l cooling, similar to the
diving response.
However, th is author’s own research group has shown th a t facia l ch ill ing does
like ly not enhance spleen contraction-related increases in Hb and Hct, whereas
the diving response which can be elicited to a certa in extent in th is manner even
without breath holding. Elsner and associates [153] earlier showed th a t an
equally strong human diving response occurred irrespective of arteria l hypoxia or
hypocapnia. Th at the diving response thus appears directly at apnea onset, and
independent of blood gas changes and periphera l chemoreception further
dissociates it from spleen contraction, shown in th is thesis’ studies to be h igh ly
sensitive to and most likely initiated by such changes.
But perh aps the strongest evidence for dissociation between apneic spleen
contraction and the human diving response is the time scale of development of
these responses. Indeed, spleen contraction may begin immediately upon breat h
holding, but it does not develop fully until after 3-5 apneas, whereas the diving
response is fully ach ieved in the first and essentia lly every apnea thereafter.
Th is may be due to the fact th a t the spleen response serves a different function –
rather th an sparing O2 , wh ich is the conceivable function of the diving response,
it more likely serves to increase O2 storage and buffer CO2 , and functions most
effectively during repeated apneas as is seen in diving mammals. From an
functional perspective, these effects would be best ach ieved during surface
periods between dives where it facilita tes recovery.
Spleen contraction in other situations also varies from the typica l diving
response. High intensity exercise, for instance, as well as exposure to a ltitude or
eupneic normobaric hypoxia both elicit contractions of the spleen in the absence
of any sort of diving response – in fact, tachycardia is more preva lent in these
situations.
38
Future directions
Other measurement techniques
Splenic contraction in humans has now been measured using radionucleotide
imaging [77, 154], CT scan and multi- axia l ultrasound. Wh i le these measures
prove effective in measuring size and volume reduction, they still do not
constitute a “gold standard” in terms of three-dimensional quantification wit h
good temporal resolution. For th a t to be ach ieved, studies of spleen contraction
must turn to magnetic resonance imaging (MRI). MRI magnetizes its target tissue
or organ externally, and then passes a radio frequency pulse through the tissue,
exciting hydrogen ions in the process to produce images of great deta i l.
Furthermore, extremely quick modern MRI techniques such as single shot fastspin echo sequences can yield multiple images in fractions of a second, providing
an extremely accurate measure of organ volume and in some methods regiona l
blood flow characteristics as well. Thornton and associates [56] used such
methods in seals to ach ieve the most accurate determinations of spleen
contraction during diving in mammals to date.
The local chemical and neurotransmitter effects on the spleen are also an area not
yet examined in humans in vivo, a lthough th is h as been ach ieved in rats using
microdialysis [89]. In th a t study, th is technique, wh ich uses tiny needles wit h
membranes semi-permeable to extracellular chemical substances, was used th is
technique to show the presence and depletion of neuronally derived
catecholamines associated with the spleen. Similar work, if eth ica lly possible
in humans, would offer much information regarding local mechanisms at work
near the spleen.
Maximizing splenic contraction through training
Wit h the correla tion between apneic performance and spleen size now
established, an obvious direction for future work is investigating the potentia l
ability to increase spleen size through tra ining. W h ile th is still remains a
purely theoretica l pursuit, there may be ties to altitude- or hypoxia-based
tra ining in ach ieving th is. Th is author’s altitude field study [120] showed th a t
after a ltitude-induced polycythemia, descending to lower a ltitudes resulted in a
larger splenic contribution during apnea th an during the same a ltitudes during
ascent. Thus, the spleen’s role as a regulator of excess blood viscosity may a lso
render it “fuller” after periods of excess blood building. This could be viewed as
an augmentation of its autologous blood doping function – in effect, making the
spleen a larger source of stored erythrocytes. In either case, th is avenue has yet to
be sufficiently explored.
39
Investigation of other possible sources of stored erythrocytes
No study has found th a t 100% of the increase in Hct and Hb during apnea,
exercise or a ltitude exposure is spleen-derived. And although the liver and
kidneys have shown only minor or no reduction in blood volume during such
interventions, the door is still open to the possibility of additional reservoirs of
blood in the human body th a t can be remobilized when required. Due to its size,
the liver is probably the most likely candidate. More deta iled and temporally
accurate imaging techniques may assist in th is pursuit, and certa inly represent a
potentia l area for future exploration.
ACKNOWLEDGEMENTS
The author would like to express his sincere th anks to the following individuals
who assisted in the work leading to the production of th is document: to Professor
Erika Schagatay, for her patient guidance in times of both success and
frustration; to colleagues and co-authors Robert de Bruijn, Angelica Lodin, Dr.
Sofia Pettersson, Helena Haughey, Dr. H.-C. Holmberg, Dr. Jenny Reimers, Glenn
Björklund and Hara ld Engan for their unselfish assistance and attention to
deta il; to Torborg Jonsson, Håkan Norberg, and Siw Stenlund for their technica l,
logistical and administrative expertise; to Dr. Johan Andersson and Dr. Mats
Liner for their help in revising the text; to th e Jönsson, Norström, Borg,
Schagatay, Rich ardson and Grousset families for th eir exceptional hospita l ity
and kindness both home and away; and not least to our subject volunteers, who,
often under great physica l duress, a llowed the auth or to take pictures of the ir
internal organs and drain their vita l f luids for no more th an a h andshake and
perh aps a free lunch.
Specia l personal gratitude is extended to Sara Norström, whose sharing in these
endeavours has made them especia lly meaningful, and to Sarah, Paul and Mary
Anne Coderre Rich ardson for their life long moral support and encouragement in
mainta ining a balance.
40
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Findley, L.J., et al., Hypoxemia during apnea in normal subjects: mechanisms
and impact of lung volume. J Appl Physiol, 1983. 55(6): p. 1777-1783.
Hong, S.K., et al., Alveolar gas exchanges and cardiovascular functions during
breath holding with air. J Appl Physiol, 1971. 30(4): p. 540-547.
Schutz, S., Oxygen saturation monitoring by pulse oximetry, in AACN
procedure manual for critical care, D. Lynn-McHale and K. Carlson, Editors.
2001, WB Saunders. p. 77-82.
Fletcher, E.C., et al., Central venous O2 saturation and rate of arterial
desaturation during obstructive apnea. J Appl Physiol, 1989. 66(3): p. 14771485.
Perini, R., et al., Heart rate and blood pressure time courses during prolonged dry
apnoea in breath-hold divers. Eur J Appl Physiol, 2008. 104(1): p. 1-7.
Schagatay, E., The human diving response: effects of temperature and training.
1996, Department of Animal Physiology, Lund University: Lund, Sweden.
p. 108.
Andersson, J.P.A., Cardiovascular and respiratory effects of apnea in humans.
2001, Department of Animal Physiology, Lund University: Lund, Sweden.
p. 49.
Foster, G.E. and A.W. Sheel, The human diving response, its function, and its
control. Scandinavian Journal of Medicine and Science in Sports, 2005.
15(1): p. 3-12.
Mithoefer, J.C., Mechanism of pulmonary gas exchange and CO2 transport
during breath holding. J Appl Physiol, 1959. 14(5): p. 706-710.
Liner, M.H., M. Ferrigno, and C.E.G. Lundgren, Alveolar Gas-Exchange
During Simulated Breath-Hold Diving to 20 M. Undersea Hyperbar M, 1993.
20(1): p. 27-38.
Craig, A.B., Jr. and A.D. Harley, Alveolar gas exchanges during breath-hold
dives. J Appl Physiol, 1968. 24(2): p. 182-189.
Klocke, F.J. and H. Rahn, Breath holding after breathing of oxygen. J Appl
Physiol, 1959. 14(5): p. 689-693.
Schagatay, E. and J. Andersson, The respiratory, circulatory and hematological
effects of repeated apneas in humans. Undersea Hyperb Med, 1999. 26 (Suppl):
p. 29-30.
O'Donnell, C.P., et al., Reflex stimulation of renal sympathetic nerve activity
and blood pressure in response to apnea. Am J Respir Crit Care Med, 1996.
154(6 Pt 1): p. 1763-1770.
Leuenberger, U.A., et al., Hypoxia augments apnea-induced peripheral
vasoconstriction in humans. J Appl Physiol, 2001. 90(4): p. 1516-1522.
Leuenberger, U.A., et al., Short-term intermittent hypoxia enhances
sympathetic responses to continuous hypoxia in humans. J Appl Physiol, 2007.
103(3): p. 835-842.
Savourey, G., et al., Normo- and hypobaric hypoxia: are there any physiological
differences? Eur J Appl Physiol, 2003. 89(2): p. 122-126.
Hackett, P. and D. Rennie, High-altitude pulmonary edema. JAMA, 2002.
287(17): p. 2275-2278.
41
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
Anand, I.S., et al., Body fluid compartments, renal blood flow, and hormones at
6,000 m in normal subjects. J Appl Physiol, 1993. 74(3): p. 1234-1239.
Cruz, J.C., et al., Sustained venoconstriction in man supplemented with CO2 at
high altitude. J Appl Physiol, 1976. 40(1): p. 96-100.
Cunningham, W.L., E.J. Becker, and F. Kreuzer, Catecholamines in plasma
and urine at high altitude. J Appl Physiol, 1965. 20(4): p. 607-610.
Hochachka, P.W., Mechanism and evolution of hypoxia-tolerance in humans. J
Exp Biol, 1998. 201(Pt 8): p. 1243-1254.
Ama, P.F., et al., Fat distribution and adipose tissue metabolism in non-obese
male black African and Caucasian subjects. Int J Obes, 1986. 10(6): p. 503-510.
Fagard, R., E. Bielen, and A. Amery, Heritability of aerobic power and
anaerobic energy generation during exercise. J Appl Physiol, 1991. 70(1): p. 357362.
Bernardi, L., Heart rate and cardiovascular variability at high altitude. Conf
Proc IEEE Eng Med Biol Soc, 2007. 2007: p. 6679-6681.
Terblanche, J., et al., Measurement reliability of highly variable physiological
responses to experimentally-manipulated gas fractions. Physiol Meas, 2004.
25(5): p. 1189-1197.
Kolb, J., Physiological responses to intermittent hypoxia in humans, in Institute
for the theory and practice of training and movement. 2003, German Sport
University Cologne: Cologne.
Gray, H., Anatomy of the human body. 1918, Philadelphia: Lea and Febiger.
Steiniger, B. and P. Barth, Microanatomy and function of the spleen. Adv Anat
Embryol Cell Biol, 2000. 151: p. III-IX, 1-101.
Mebius, R.E. and G. Kraal, Structure and function of the spleen. Nat Rev
Immunol, 2005. 5(8): p. 606-616.
MacDonald, I.C., et al., Kinetics of red blood cell passage through
interendothelial slits into venous sinuses in rat spleen, analyzed by in vivo
microscopy. Microvasc Res, 1987. 33(1): p. 118-134.
Willekens, F.L.A., et al., Hemoglobin loss from erythrocytes in vivo results from
spleen-facilitated vesiculation. Blood, 2003. 101(2): p. 747.
MacDonald, I.C., E.E. Schmidt, and A.C. Groom, The high splenic hematocrit:
a rheological consequence of red cell flow through the reticular meshwork.
Microvasc Res, 1991. 42(1): p. 60-76.
Wolski, L., The impact of splenic release of red cells on hematocrit changes during
exercise [PhD thesis]. 1998, University of British Columbia.: Vancouver, B.C.
Bakovic, D., et al., Spleen volume and blood flow response to repeated breath-hold
apneas. J Appl Physiol, 2003. 95(4): p. 1460-1466.
Stewart, I.B. and D.C. McKenzie, The human spleen during physiological
stress. Sports Med, 2002. 32(6): p. 361-369.
Barcroft, J., Y. Nisimaru, and S.R. Puri, The action of the splanchnic nerves on
the spleen. J Physiol, 1932. 74(3): p. 321-326.
Roy, C.S., The Elastic Properties of the Arterial Wall. J Physiol, 1881. 3(2): p.
125-159.
Schafer, E.A. and B. Moore, On the Contractility and Innervation of the Spleen.
J Physiol, 1896. 20(1): p. 1-50.
Barcroft, J., L.C. Khanna, and Y. Nisimaru, Rhythmical contraction of the
spleen. J Physiol, 1932. 74(3): p. 294-298.
42
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
Grindlay, J.H., J.F. Herrick, and E.J. Baldes, Rhythmicity of the spleen in
relation to blood flow. Am J Physiol, 1939. 127(1): p. 119-126.
Roberts, G.M. and L.A. Crandall, JR., The role of the portal system in the
regulation of circulating blood volume. Am J Physiol, 1933. 106(2): p. 423-431.
Granaat, D., The spleen in the regulation of the arterial blood pressure. J Physiol,
1953. 122(2): p. 209-218.
Kaufman, S. and Y. Deng, Splenic control of intravascular volume in the rat. J
Physiol, 1993. 468: p. 557-565.
Carneiro, J.J. and D.E. Donald, Blood reservoir function of dog spleen, liver, and
intestine. Am J Physiol, 1977. 232(1): p. H67-72.
Guntheroth, W.G. and G.L. Mullins, Liver and spleen as venous reservoirs.
Am J Physiol, 1963. 204(1): p. 35-41.
Brender, E., The Spleen. JAMA, 2005. 294(20): p. 2660.
Hurford, W.E., et al., Splenic contraction during breath-hold diving in the
Korean ama. J Appl Physiol, 1990. 69(3): p. 932-936.
Laub, M., et al., Spleen emptying and venous hematocrit in humans during
exercise. J Appl Physiol, 1993. 74(3): p. 1024-1026.
Schagatay, E., et al., Selected Contribution: Role of spleen emptying in
prolonging apneas in humans. J Appl Physiol, 2001. 90(4): p. 1623-1629.
Isbister, J.P., Physiology and pathophysiology of blood volume regulation.
Transfus Sci, 1997. 18(3): p. 409-423.
Rogausch, H., et al., Local and systemic autonomic nervous effects on cell
migration to the spleen. J Appl Physiol, 2003. 94(2): p. 469-475.
Cabanac, A.J., Contracted spleen in seals, estimates of dilated organs, and diving
capacity. Polar Biol, 2002. 25: p. 1-4.
Hochachka, P.W., Pinniped diving response mechanism and evolution: a
window on the paradigm of comparative biochemistry and physiology. Comp
Biochem Physiol A Mol Integr Physiol, 2000. 126(4): p. 435-458.
Qvist, J., et al., Hemoglobin concentrations and blood gas tensions of free-diving
Weddell seals. J Appl Physiol, 1986. 61(4): p. 1560-1569.
Thornton, S.J., et al., Effects of forced diving on the spleen and hepatic sinus in
northern elephant seal pups. Proc Natl Acad Sci U S A, 2001. 98(16): p. 94139418.
Zapol, W.M., Diving adaptations of the Weddell seal. Sci Am, 1987. 256(6): p.
100-105.
Elsner, R. and H.J. Meiselman, Splenic oxygen storage and blood viscosity in
seals. Marine Mammal Science, 1995. 11: p. 93-96.
Hargis, E.H. and F.C. Mann, A plethysmographic study of the changes in the
volume of the spleen in the intact animal. Am J Physiol, 1925. 75(1): p. 180-200.
Nice, L.B. and D. Fishman, Changes in the viscosity of the blood in normal,
splenectomized, and adrenalectomized animals following emotional excitement.
Am J Physiol, 1933. 107(1): p. 113-119.
Turner, A.W. and V.E. Hodgetts, The dynamic red cell storage function of the
spleen in sheep. I. Relationship to fluctuations of jugular haematocrit. Aust J Exp
Biol Med Sci, 1959. 37: p. 399-420.
Anderson, R.S. and E.B. Rogers, Hematocrit and erythrocyte volume
determinations in the goat as related to spleen behavior. Am J Physiol, 1957.
188(1): p. 178-188.
43
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
Hannon, J.P., C.A. Bossone, and W.G. Rodkey, Splenic red cell sequestration
and blood volume measurements in conscious pigs. Am J Physiol, 1985. 248(3 Pt
2): p. R293-301.
Persson, S.G., et al., Circulatory effects of splenectomy in the horse. I. Effect on
red-cell distribution and variability of haematocrit in the peripheral blood.
Zentralbl Veterinarmed A, 1973. 20(6): p. 441-455.
Kramer, K. and U.C. Luft, Mobilization of Red Cells and Oxygen From the
Spleen in Severe Hypoxia. Am J Physiol, 1951. 165(1): p. 215-228.
Baker, C.H. and J.W. Remington, Role of the spleen in determining total body
hematocrit. Am J Physiol, 1960. 198(4): p. 906-910.
Smith, D.C. and F.C. Brown, Effects of acute decompression stress upon some
blood components, especially leucocytes, in intact and splenectomized cats. Am J
Physiol, 1951. 164(3): p. 752-765.
Hoka, S., et al., Regional venous outflow, blood volume, and sympathetic nerve
activity during severe hypoxia. Am J Physiol Heart Circ Physiol, 1989. 256(1):
p. H162-170.
Kuwahira, I., et al., Splenic contraction-induced reversible increase in
hemoglobin concentration in intermittent hypoxia. J Appl Physiol, 1999. 86(1):
p. 181-187.
Robertson, D.A., F.G. Simpson, and M.S. Losowsky, Blood viscosity after
splenectomy. Br Med J (Clin Res Ed), 1981. 283(6291): p. 573-575.
Neuhaus, D. and P. Gaehtgens, Haemorrheology and long term exercise.
Sports Med, 1994. 18(1): p. 10-21.
Stewart, I.B., et al., Cardiovascular and splenic responses to exercise in humans. J
Appl Physiol, 2003. 94(4): p. 1619-1626.
Sandler, M.P., et al., Dynamic fluctuations in blood and spleen radioactivity:
splenic contraction and relation to clinical radionuclide volume calculations. J Am
Coll Cardiol, 1984. 3(5): p. 1205-1211.
Flamm, S.D., et al., Redistribution of regional and organ blood volume and effect
on cardiac function in relation to upright exercise intensity in healthy human
subjects. Circulation, 1990. 81(5): p. 1550-1559.
Froelich, J.W., et al., Redistribution of visceral blood volume in upright exercise
in healthy volunteers. J Nucl Med, 1988. 29(10): p. 1714-1718.
Bakovic, D., et al., Effect of human splenic contraction on variation in
circulating blood cell counts. Clin Exp Pharmacol Physiol, 2005. 32(11): p.
944-951.
Espersen, K., et al., The human spleen as an erythrocyte reservoir in divingrelated interventions. J Appl Physiol, 2002. 92(5): p. 2071-2079.
Richardson, M., et al. Hematological response pattern associated with maximalduration apnea series in untrained subjects. in XXIX Annual Scientific Meeting
of the European Underwater and Baromedical Society. 2003. Copenhagen,
Denmark.
Schagatay, E., H. Haughey, and J. Reimers, Speed of spleen volume changes
evoked by serial apneas, in European Journal of Applied Physiology. 2005. p. 447452.
Davies, B.N., J. Gamble, and P.G. Withrington, Frequency-dependent
differences in the responses of the capsular and vascular smooth muscle of the
44
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
spleen of the dog to sympathetic nerve stimulation. J Physiol, 1973. 228(1): p. 1325.
Donald, D.E. and L.L. Aarhus, Active and passive release of blood from canine
spleen and small intestine. Am J Physiol, 1974. 227(5): p. 1166-1172.
Hurford, W.E., et al., Splenic contraction, catecholamine release, and blood
volume redistribution during diving in the Weddell seal. J Appl Physiol, 1996.
80(1): p. 298-306.
Reilly, F.D., Innervation and vascular pharmacodynamics of the mammalian
spleen. Experientia, 1985. 41(2): p. 187-192.
Felten, D.L., Neural influence on immune responses: underlying suppositions
and basic principles of neural-immune signaling. Prog Brain Res, 2000. 122: p.
381-389.
Ayers, A.B., B.N. Davies, and P.G. Withrington, Responses of the isolated,
perfused human spleen to sympathetic nerve stimulation, catecholamines and
polypeptides. Br J Pharmacol, 1972. 44(1): p. 17-30.
Cabanac, A., L.P. Folkow, and A.S. Blix, Volume capacity and contraction
control of the seal spleen. J Appl Physiol, 1997. 82(6): p. 1989-1994.
Davies, B.N., D.A. Powis, and P.G. Withrington, The differential effect of
cooling on the responses of splenic capsular and vascular smooth muscle to nerve
stimulation and noradrenaline. Pflugers Archiv, 1978. 377(1): p. 87-94.
Herman, N.L., D.R. Kostreva, and J.P. Kampine, Splenic afferents and some of
their reflex responses. Am J Physiol Regul Integr Comp Physiol, 1982. 242(3):
p. R247-254.
Shimizu, N., T. Hori, and H. Nakane, An interleukin-1 beta-induced
noradrenaline release in the spleen is mediated by brain corticotropin-releasing
factor: an in vivo microdialysis study in conscious rats. Brain Behav Immun,
1994. 8(1): p. 14-23.
Ojiri, Y., et al., Effects of adrenergic stimulants on the splenic diameter,
haemoglobin content and haematocrit in anaesthetized dogs: determination of the
adrenoceptor subtype responsible for changes in the splenic diameter. Acta
Physiol Scand, 1993. 149(1): p. 31-39.
Kaufman, M.J., et al., Cocaine administration induces human splenic
constriction and altered hematologic parameters. J Appl Physiol, 1998. 85(5): p.
1877-1883.
Reilly, F.D. and R.S. McCuskey, Studies of the hemopoietic microenvironment.
VII. Neural mechanisms in splenic microvascular regulation in mice. Microvasc
Res, 1977. 14(3): p. 293-302.
Reilly, F.D. and R.S. McCuskey, Studies of the hemopoietic microenvironment.
VI. Regulatory mechanisms in the splenic microvascular system of mice.
Microvasc Res, 1977. 13(1): p. 79-90.
Heistad, D.D. and F.M. Abboud, Dickinson W. Richards Lecture: Circulatory
adjustments to hypoxia. Circulation, 1980. 61(3): p. 463-470.
Cutler, M.J., et al., Periods of intermittent hypoxic apnea can alter chemoreflex
control of sympathetic nerve activity in humans. Am J Physiol Heart Circ
Physiol, 2004: p. 00377.02004.
Quatacker, J., P. Partoens, and W. De Potter, Postganglionic, direct axo-axonal
contacts on the splenic nerve. Brain Res, 2001. 891(1-2): p. 213-217.
45
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
Felten, D.L., et al., Noradrenergic sympathetic innervation of the spleen: I. Nerve
fibers associate with lymphocytes and macrophages in specific compartments of the
splenic white pulp. J Neurosci Res, 1987. 18(1): p. 28-36, 118-121.
Calaresu, F.R., et al., Splenic and renal sympathetic responses to stimulation of
splenic receptors in cats. Am J Physiol, 1984. 247(5 Pt 2): p. R856-865.
Deng, Y. and S. Kaufman, Splenorenal reflex regulation of arterial pressure.
Hypertension, 2001. 38(3): p. 348-352.
Moncrief, K. and S. Kaufman, Splenic baroreceptors control splenic afferent
nerve activity. Am J Physiol Regul Integr Comp Physiol, 2006. 290(2): p.
R352-356.
Schagatay, E., J. Andersson, and B. Nielsen. Does facial chilling augment the
hematocrit increase seen at human apneic diving? in International Thermal
Physiology Symposium. 2001. Wollongong, Australia: IUPS Commission on
Thermal Physiology.
Richardson, M., et al., Increase of hemoglobin concentration after maximal
apneas in divers, skiers and untrained humans. Canadian Journal of Applied
Physiology, 2005. 30(3): p. 276-281.
Winslow, R. and C. Monge, Hypoxia, polycythemia and chronic mountain
sickness. 1987, Baltimore, MD: John Hopkins University Press. 255.
Breiman, R.S., et al., Volume determinations using computed tomography. AJR
Am J Roentgenol, 1982. 138(2): p. 329-333.
Sonmez, G., et al., Effects of altitude on spleen volume: sonographic assessment. J
Clin Ultrasound, 2007. 35(4): p. 182-185.
Lamb, P.M., et al., Spleen size: how well do linear ultrasound measurements
correlate with three-dimensional CT volume assessments? The British Journal of
Radiology, 2002. 75(895): p. 573.
Picardi, M., et al., Measurement of spleen volume by ultrasound scanning in
patients with thrombocytosis: a prospective study. Blood, 2002. 99(11): p. 4228.
Ishibashi, H., et al., Sonographic assessment and grading of spleen size. J Clin
Ultrasound, 1991. 19(1): p. 21-25.
Koga, T., Correlation between sectional area of the spleen by ultrasonic
tomography and actual volume of the removed spleen. J Clin Ultrasound, 1979.
7(2): p. 119-120.
Perkins, G.D., et al., Do changes in pulse oximeter oxygen saturation predict
equivalent changes in arterial oxygen saturation? Crit Care, 2003. 7(4): p. R67.
Fletcher, E.C., C. Costarangos, and T. Miller, The rate of fall of arterial
oxyhemoglobin saturation in obstructive sleep apnea. Chest, 1989. 96(4): p. 717722.
Petersen, M.E., T.R. Williams, and R. Sutton, A comparison of non-invasive
continuous finger blood pressure measurement (Finapres) with intra-arterial
pressure during prolonged head-up tilt. Eur Heart J, 1995. 16(11): p. 1641-1654.
Raamat, R., et al., Beat-to-beat measurement of the finger arterial pressure pulse
shape index at rest and during exercise. Clin Phys Funct Imaging, 2003. 23(2):
p. 87-91.
Raamat, R., K. Jagomagi, and J. Talts, Different responses of Finapres and the
oscillometric finger blood pressure monitor during intensive vasomotion. J Med
Eng Technol, 2000. 24(3): p. 95-101.
46
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
Jagomagi, K., R. Raamat, and J. Talts, Effect of altering vasoactivity on the
measurement of finger blood pressure. Blood Press Monit, 2001. 6(1): p. 33-40.
Lundvall, J. and P. Bjerkhoel, Failure of hemoconcentration during standing to
reveal plasma volume decline induced in the erect posture. J Appl Physiol, 1994.
77(5): p. 2155-2162.
Cohen, J., Statistical power analysis for the behavioural sciences. 2nd edition ed.
1988, New Jersey: Lawrence Erlbaum.
Richardson, M., Holmberg, H.-C., de Bruijn, R., Björklund, G., Haughey,
H., Schagatay, E. Hemoglobin concentration after serial apneas in divers, skiers
and untrained humans. in 7th Scandinavian Congress on Medicine and Science in
Sports. 2004. Stockholm, Sweden.
Wiesen, A.R., et al., Equilibration of hemoglobin concentration after transfusion
in medical inpatients not actively bleeding. Ann Intern Med, 1994. 121(4): p.
278-230.
Richardson, M. and E. Schagatay. Altitude attenuates apnea-induced increase
in hemoglobin concentration. in 12th International Conference on Environmental
Ergonomics. 2007. Piran, Slovenia: Biomed.
Somers, V.K., et al., Contrasting effects of hypoxia and hypercapnia on
ventilation and sympathetic activity in humans. J Appl Physiol, 1989. 67(5): p.
2101-2106.
Prabhakar, N.R., O2 sensing at the mammalian carotid body: why multiple O2
sensors and multiple transmitters? Experimental Physiology, 2006. 91(1): p.
17-23.
Ahmad, H.R. and H.H. Loeschcke, Transient and steady state responses of
pulmonary ventilation to the medullary extracellular pH after approximately
rectangular changes in alveolar PCO2. Pflugers Arch, 1982. 395(4): p. 285-292.
Smith, C.A., et al., Response time and sensitivity of the ventilatory response to
CO2 in unanesthetized intact dogs: central vs. peripheral chemoreceptors. J Appl
Physiol, 2006. 100(1): p. 13-19.
Rodman, J.R., et al., Carotid body denervation in dogs: eupnea and the
ventilatory response to hyperoxic hypercapnia. J Appl Physiol, 2001. 91(1): p.
328-335.
Loeschchke, H.H. and K.H. Gertz, [Effect of oxygen pressure in inspired air on
respiratory activity of the human, tested under the constant behavior of alveolar
carbon dioxide pressure.]. Pflugers Arch, 1958. 267(5): p. 460-477.
Lahiri, S. and R.G. DeLaney, Stimulus interaction in the responses of carotid
body chemoreceptor single afferent fibers. Respir Physiol, 1975. 24(3): p. 249266.
Daristotle, L., A.D. Berssenbrugge, and G.E. Bisgard, Hypoxic-hypercapnic
ventilatory interaction at the carotid body of awake goats. Respir Physiol, 1987.
70(1): p. 63-72.
Nakayama, H., et al., Carotid body denervation eliminates apnea in response to
transient hypocapnia. J Appl Physiol, 2003. 94(1): p. 155-164.
Cummings, K.J. and R.J.A. Wilson, Time-dependent modulation of carotid body
afferent activity during and after intermittent hypoxia. Am J Physiol Regul
Integr Comp Physiol, 2005. 288(6): p. R1571-1580.
47
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
Zhang, M. and C.A. Nurse, CO2/pH chemosensory signaling in co-cultures of
rat carotid body receptors and petrosal neurons: role of ATP and ACh. J
Neurophysiol, 2004. 92(6): p. 3433-3445.
Klein, R., et al., Opioid peptides and noradrenaline co-exist in large dense-cored
vesicles from sympathetic nerve. Neuroscience, 1982. 7(9): p. 2255-2261.
Elenkov, I.J., et al., The sympathetic nerve--an integrative interface between two
supersystems: the brain and the immune system. Pharmacol Rev, 2000. 52(4): p.
595-638.
Beluli, D.J. and L.C. Weaver, Areas of rostral medulla providing tonic control of
renal and splenic nerves. Am J Physiol, 1991. 261(6 Pt 2): p. H1687-1692.
Trzebski, A. and L. Kubin, Is the central inspiratory activity responsible for
pCO2-dependent drive of the sympathetic discharge? J Auton Nerv Syst, 1981.
3(2-4): p. 401-420.
Cutler, M.J., et al., Hypoxia-mediated prolonged elevation of sympathetic nerve
activity after periods of intermittent hypoxic apnea. J Appl Physiol, 2004. 96(2):
p. 754-761.
Dujic, Z., et al., Central chemoreflex sensitivity and sympathetic neural outflow
in elite breath-hold divers. J Appl Physiol, 2008. 104(1): p. 205-211.
Kara, T., K. Narkiewicz, and V.K. Somers, Chemoreflexes - physiology and
clinical implications. Acta Physiol Scand, 2003. 177(3): p. 377-384.
Narkiewicz, K., et al., Baroreflex control of sympathetic nerve activity and heart
rate in obstructive sleep apnea. Hypertension, 1998. 32(6): p. 1039-1043.
Korner, P.I. and S.W. White, Circulatory control in hypoxia by the sympathetic
nerves and adrenal medulla. J Physiol, 1966. 184(2): p. 272-290.
Iturriaga, R., J. Alcayaga, and P. Zapata, Dissociation of hypoxia-induced
chemosensory responses and catecholamine efflux in cat carotid body superfused in
vitro. J Physiol, 1996. 497 (Pt 2): p. 551-564.
Berglund, B., High-altitude training. Aspects of haematological adaptation.
Sports Med, 1992. 14(5): p. 289-303.
Macefield, V.G. and B.G. Wallin, Effects of static lung inflation on sympathetic
activity in human muscle nerves at rest and during asphyxia. J Auton Nerv
Syst, 1995. 53(2-3): p. 148-156.
Macefield, V.G., S.C. Gandevia, and L.A. Henderson, Neural sites involved
in the sustained increase in muscle sympathetic nerve activity induced by
inspiratory capacity apnea: a fMRI study. J Appl Physiol, 2006. 100(1): p. 266273.
Palada, I., et al., Spleen and cardiovascular function during short apneas in
divers. J Appl Physiol, 2007. 103(6): p. 1958-1963.
Turcotte, R., et al., Exercise-induced oxyhemoglobin desaturation and pulmonary
diffusing capacity during high-intensity exercise. European Journal of Applied
Physiology, 1997. 75(5): p. 425.
de Bruijn, R., M. Richardson, and E. Schagatay, Increased erythropoietin
concentration after repeated apneas in humans. Eur J Appl Physiol, 2008.
102(5): p. 609-613.
Dane, D.M., et al., Splenectomy impairs diffusive oxygen transport in the lung of
dogs. J Appl Physiol, 2006. 101(1): p. 289-297.
Windsor, J. and G. Rodway, Heights and haematology: the story of hemoglobin
at altitude. Postgrad Med J, 2007. 83(977): p. 148-151.
48
150.
151.
152.
153.
154.
Ekblom, B., A.N. Goldbarg, and B. Gullbring, Response to exercise after blood
loss and reinfusion. J Appl Physiol, 1972. 33(2): p. 175-180.
Ekblom, B., G. Wilson, and P.O. Astrand, Central circulation during exercise
after venesection and reinfusion of red blood cells. J Appl Physiol, 1976. 40(3): p.
379-383.
Wagner, P., et al., Maximum oxygen transport and utilisation before and after
splenectomy. Equine Vet J Suppl, 1995. 18: p. 82-89.
Elsner, R., B.A. Gooden, and S.M. Robinson, Arterial blood gas changes and
the diving response in man. Aust J Exp Biol Med Sci, 1971. 49(5): p. 435-444.
Hoefs, J.C., et al., A Novel, Simple Method of Functional Spleen Volume
Calculation by Liver-Spleen Scan. The Journal of Nuclear Medicine, 1999.
40(10): p. 1745.
49
PAPER I.
Correlation between spleen size and hematocrit during apnea in humans. MX
Rich ardson, R deBruijn, S Pettersson, J Reimers and E Schagatay. 2006. In:
Lindholm, P, Pollock NW, Lundgren CEG, eds. Breath- hold diving. Proceedings
of the Undersea and Hyperbaric Medica l Society/Divers Alert Network 2006
June 20-21 Workshop, p 116-119.
Correlation between spleen size and hematocrit during apnea in
humans
Richardson M1 , de Bruijn R1 , Petterson S1 , Reimers J2 , Schagatay E1
1) Environmenta l Physiology Group, Mid Sweden University, Sundsvall, Sweden.
2) X-ray Clinic, Härnösand Hospita l, Härnösand, Sweden.
Introduction
Increases in hemoglobin concentration (Hb) and hematocrit (Hct) h ave been
observed after both solitary (3) and repeated apneas spaced by short interva ls (7;
2). In the la tter case, the increase in Hb and Hct reaches a maximum after 3 – 4
maximal effort apneas (7). This increase is reversed with in 8-9 minutes after
cessation of apneas (8). The return to baseline of Hb shows th a t the response is not
due to increased diuresis or sweating, and constancy of protein levels shows th a t
it is not due to extravasation of plasma by increased filtration (7). In diving
mammals, such as sea ls, increases in Hb and Hct h ave a lso been observed during
dives and th is has been attributed to a simultaneous reduction in spleen size (4).
It is thus conceivable th a t the increase in Hb and Hct seen during apneas in
humans is, at least partly, caused by a similar contraction of the spleen. In
humans, as in other mammals, the spleen apparently reta ins erythrocytes and
forms a reservoir of red blood cells while reducing blood viscosity during rest (10),
but the size of th is reservoir stil l needs to be determined. Contraction of t he
spleen during apnea has been shown to be an active process, largely independent
of arteria l inflow, suggesting the presence of a centrally-mediated feed-forward
mechanism (1). No direct assessment has been made of splenic volume changes in
rela tion to increases in hematocrit during apnea, and the indirect estimations
h ave varied. In th is study we a imed to further investigate the rela tion between
spleen volume, Hb and Hct during seria l apneas.
Methods
50
Four healthy subjects volunteered to perform a series of 3 maximal apneas, spaced
by 2 min pauses. Ultrasonic measurements of spleen diameter and venous blood
samples, for Hb and Hct analysis, were taken pre-series, after the 1 st and 3 rd
apnea, and at 2, 4, 6, 8, and 10 minutes following apnea 3. Ultrasonic
measurements of spleen maximal length (L), th ickness (T) and dorsoventra l
width ( W) were used to calculate spleen volume using the following equation:
L!(WT-T2 )/3. Comparisons between pre- and post-series measurements were made
for splenic volume, Hb and Hct using paired Students t-test. Correlations between
spleen volume and Hb and Hct across the series were calculated using Pearsons
test. An estimation of the spleen contribution to observed Hct changes was made,
based on an estimated blood volume of 8% of body weight, a spleen hematocrit of
80% (5) and the spleen volume and Hct observed before and after apneas.
Results
A reduction in spleen volume during apneas was seen in all subjects, on average
from 338 ml pre-series to 223 ml directly after apnea 3 (by 34%; P=0.01).
S imultaneously increases in Hb, from 133 g/L to 136 g/L (P=0.02), and in Hct, from
38.6% to 39.9% (P=0.07), were observed (Figure 1). The Pearson´s correlation
coefficients for spleen volume and Hb and Hct were -0.88 and -0.85, respectively
(both P<0.01). The reduction in spleen volume was estimated to account for 70% of
the observed increase in Hct. The apnea-induced changes of Hb, Hct and spleen
volume were completely reversed with in 10 minutes after the last apnea.
51
Figure 1. Mean spleen volume (ml) and hematocrit (Hct: %) from 4 subjects before
apneas, after apnea 1, apnea 3, and during 10 min recovery after apneas. * *
denotes significance at P=0.01 for spleen volume after apneas, compared to
control. The P value for Hct was 0.07.
Discussion
The inverse correlation between spleen volume and both Hb and Hct during the
apnea series suggests a causative rela tion between the parameters. The
estimation th a t the observed reduction in spleen volume could account for 70% of
the observed Hct increase, further supports th is conclusion. Th is is a h igher va lue
compared to previously reported estimations of 60 % (7) and 37 % (8). One
difference may be the timing of post apnea spleen volume measurement, wh ich in
the present study is directly after apnea termination. We speculate th a t t he
remaining increase in Hb and Hct could arise from either extravasation of
plasma, or possibly from venous blood pools with h igh hematocrit in the
abdomen such as the liver. The transient increases in circulating erythrocyte
volume will increase oxygen storage capacity as well as carbon dioxide buffering
during subsequent apneas, both with a dive prolonging effect. The spleen related
increases in Hb and Hct may be an explanation for the increases in apneic
duration in successive apneas (7), previously thought to be due to
52
hyperventila tion. Some observations indicate th a t th is spleen and Hb response
may be trainable: The increase in apneic duration across seria l apneas tended to
be stronger after apnea tra ining (9), and trained apneists were found to have a
larger increase in Hb during apneas, compared to untra ined subjects and elite
cross-country skiers (6), which is linked to spleen contraction by the present
study. Further studies will reveal whether the spleen response can be trained,
and what benefits it may offer in different sports.
Conclusion: We conclude th at spleen contraction and Hb and Hct changes develop
in close association and th a t the main cause of the rise in Hb and Hct observed
after repeated apneas is spleen contraction.
References
1. Bakovic D, Eterovic D, Saratlija-Novakovic Z, Pa lada I, Va lic Z,
Bilopavlovic N and Dujic Z. Effect of human splenic contraction on varia tion in
circulating blood cell counts. Clin Exp Pharmacol Physiol 2005; 32: 944-951.
2. Bakovic D, Va lic Z, Eterovic D, Vukovic I, Obad A, Marinovic-Terzic I and
Dujic Z. Spleen volume and blood flow response to repeated breath- hold apneas. J
Appl Physiol 2003; 95: 1460-1466.
3. Espersen K, Frandsen H, Lorentzen T, Kanstrup IL and Christensen NJ. The
human spleen as an erythrocyte reservoir in diving-rela ted interventions. J Appl
Physiol 2002; 92: 2071-2079.
4. Hurford WE, Hochachka PW, Schneider RC, Guyton GP, Stanek KS, Zapol
DG, Liggins GC and Zapol WM. Splenic contraction, catecholamine release, and
blood volume redistribution during diving in the Weddell sea l. J Appl Physiol
1996; 80: 298-306.
5. Laub M, Hvid-Jacobsen K, Hovind P, Kanstrup IL, Christensen NJ and Nielsen
SL. Spleen emptying and venous hematocrit in humans during exercise. J App l
Physiol 1993; 74: 1024-1026.
6. Rich ardson M, de Bruijn R, Holmberg H-C, Björklund G, Haughey H and
Schagatay E. Increase of hemoglobin concentration after maximal apneas in
divers, skiers, and untra ined humans. Can J Appl Physiol 2005; 30: 276-281.
7. Schagatay E, Andersson JP, Ha llen M and Palsson B. Selected contribution: role
of spleen emptying in prolonging apneas in humans. J Appl Physiol 2001; 90: 16231629; discussion 1606.
8. Schagatay E, Haughey H and Reimers J. Speed of spleen volume changes
evoked by seria l apneas. Eur J Appl Physiol 2005; 93: 447-452.
53
9. Schagatay E, van Kampen M, Emanuelsson S and Holm B. Effects of physica l
and apnea tra ining on apneic time and the diving response in humans. Eur J Appl
Physiol 2000; 82: 161-169.
10. Stewart IB and McKenzie DC. The human spleen during physiological stress.
Sports Med 2002; 32: 361-369.
54
PAPER II.
Increase of hemoglobin concentration after maximal apneas in divers, skiers and
untrained humans. 2005. MX Rich ardson, R deBruijn, HC Holmberg, G Björklund,
H Haughey and E Schagatay. Canadian Journal of Applied Physiology, 30(3):
276-281. Th is article is reproduced with the kind permission of NRC Research
Press.
Increase of Hemoglobin Concentration After Maximal Apneas in
Divers, Skiers, and Untrained Humans
Matt Richardson1 , Robert de Bruijn1 , H.-C. Holmberg2,3 , Glenn Björklund1,4 ,
Helena Haughey1 , and Erika Schagatay1
1
Dept. of Natural and Environmenta l Physiology, Mid Sweden University, 85170
Sundsva ll, Sweden; 2 Dept. of Physiology & Ph armacology, Karolinska Institut,
Stockholm,
Sweden; 3 Swedish Olympic Committee, Sofia tornet, Olympiastadion,
Stockholm, Sweden;
4
National Winter Sport Center, Östersund, Sweden.
Abstract/Résumé
Human splenic contraction occurs both during apnea and maximal exercise,
increasing the circulating erythrocyte volume. We investigated the
hematological responses to 3 maximal apneas performed by elite apneic divers,
elite cross-country skiers, and untra ined subjects.
Post-apnea hemoglobin concentration had increased in all groups, but especia l ly
in divers.
The increases disappeared with in 10 min of recovery. Apneic duration across
apneas a lso increased the most in divers. Responses in divers could be more
pronounced as a result of apnea training.
La contraction de la rate chez l’être humain qui se produit pendant l’apnée et au
cours d’un exercice maximal augmente le nombre d’érythrocytes en circulation.
Nous avons analysé les ajustements hématologiques de plongeurs en apnée de haut
niveau, de skieurs de fond de haut niveau, et de sujets non entraînés au cours de
trois apnées maximales. On remarque après une période d’apnée une augmentation
de la concentration d’hémoglobine dans les trois groupes, mais de façon plus
marquée chez les plongeurs. L’augmentation s’estompa pendant les 10min de
55
récupération. La période apnéique fut plus marquée chez les plongeurs. Les
adaptations plus importantes chez les plongeurs sont peut-être dues à
l’entra înement à l’apnée.
Key words: breath hold, seria l apneas, spleen contraction, cross-country skiing
Mots-clés: apnée, apnée en série, contraction de la rate, ski de fond
Introduction
Oxygen storage capacity is a vita l component both for endurance performance in
a th letes and for apneic duration in apneic divers. One way to increase oxygen
storage capacity is to increase the circulating hemoglobin volume by various
means such as a ltitude tra ining or even blood doping (Ekblom, 2000). Severa l
studies have shown th a t spleen contraction is associa ted with near-maximal or
maximal exercise (Froelich et al., 1988; Laub et a l., 1993; Wolski, 1998). This
spleen contraction during exercise elicits a transient increase in hemoglobin
concentration (Hb) and hematocrit (Laub et a l., 1993; Wolski, 1998), wh ich may,
without changes in other blood components, increase oxygen carrying capacity. An
increase in circulating Hb can also be ach ieved by apnea-induced splenic
contraction (Schagatay et a l., 2001), leading to an increase in oxygen carrying
capacity. Repeated apneas in series with short rest interva ls increase in duration
(Schagatay et a l., 1999), wh ich h as been attributed to the increase in Hb due to
spleen contraction (Schagatay et a l., 2001). Th is prolongation may reflect both
increased oxygen storage and improvements in CO2 buffering capacity on
repeated apneic attempts. Hb returns to normal va lues with in 10 minutes of
normal breath ing, wh ich is a lso the time known to reverse the increase in apneic
duration at repetition. The return of Hb to baseline shows th at the response is not
due to increased diuresis or sweating, and the constancy of protein levels shows
th a t it is not due to extravasation of plasma by increased filtration (Schagat a y
et a l., 2001). Apnea-induced splenic contraction has been shown to be present in
tra ined apneists (Bakovic et al., 2003; Hurford et al., 1990), and a spleena ttributable increase in Hb after apneas has been documented in untra ined
subjects (Schagatay et a l., 2001), but it is unknown whether th is response can be
augmented through apneic tra ining. It is also unknown whether spleen
contraction resulting in increased Hb can be triggered in endurance ath letes by
apnea alone, and how th is compares to the response in apneic divers and
untrained subjects. The present study was a imed at describing the hematologica l
response pattern associated with repeated maxima l apneas in hea lthy nondivers (untra ined), nondiving elite cross-country skiers (skiers), and elite apneic
divers (divers).
56
Methods
Subjects
A tota l of 78 volunteers from specific populations, ages 18 years and older, were
recruited into three groups: elite divers, elite skiers, and untrained subjects. Of
the 33 elite divers, 29 men and 4 women (30.3 ± 10.6 yrs, 75.9 ± 12.1 kg, 181.5 ± 8.5
cm), 17 were national team members in apnea sports from six countries, 9 were
Swedish underwater rugby players, and 7 were apnea or free diving instructors
from Sweden. They trained an average of 5.5 ± 4.0 hrs a week in apneic diving
and apnea rela ted activities, and also an average of 6.5 ± 6.1 hrs a week of
physica l tra ining. No divers had been residing or tra ining at a ltitude in the 6
months prior to the study, but some of the national team members had been
competing in breath- holding in the month leading up to the testing period.
The 13 skiers, a ll male (20.2 ± 3 yrs, 72.1 ± 5.5 kg, 180.6 ± 6.2 cm), were a l l
members of the Swedish national senior (n = 4), under-23 (n = 4), and junior (n = 5)
cross-country ski teams. They were tested in their off-season during which they
tra ined 13.5 ± 3.2 hrs per week of ski tra ining or skiing rela ted physica l
activities. They h ad little or no experience in breath- holding. There h ad been no
competi- tions in the previous month before testing, and none of the skiers had
been living or training at a ltitude in the 6 months prior to the study.
The 32 untra ined subjects (30.0 ± 6.3 yrs, 71.6 ± 8.4 kg, 177.6 ± 8.0 cm) were 9 women
and 23 men who had little or no experience in breath holding. On average th is
group completed 2.5 ± 3.1 hrs per week of physica l training or ath letic activities,
but none were competitive ath letes in any sport, and none were living at altitude.
Experimental Procedure
The subjects reported to the lab after 2 hrs without heavy meals and 1 hr without
any meals or liquid intake. They signed informed consent documents for the test
protocol, wh ich was approved by the regional human eth ics board. Following 20
min of horizonta l rest, each subject performed 3 apneas of maximal, voluntary
duration without time cues, spaced by 2 min of rest and normal breath ing. Subjects
were instructed not to hyperventila te and to immedia te ly precede apneas by a
tota l expiration and a deep but not maximal inspiration.
Measurements
Blood samples for hemoglobin values were taken by tra ined personnel via
intravenous catheter (Venflon Pro, Beckton-Dickson AB, Helsingborg, Sweden)
and analyzed via automated blood analysis unit (Micros 60 Analyzer, ABX
Diagnostics, Montpellier, France), or via capillary blood samples and analyzed
by hemoglobin analyzer (B-Hemoglobin Photometer, Hemocue AB, Ängelholm,
57
Sweden), a ll analyzed in triplicate. These two methods have been shown to
produce similar va lues in previous studies (Neufeld et a l., 2002; Pa iva Ad Ade et
a l., 2004) and in our own pilot study data (unpublished). Methods were consistent
with in individuals and each person served as h is or her own control. Blood
samples were drawn immediately before the onset of the 2-min countdown for t he
first apnea, immediately following the final apnea, and 10 min after the fina l
apnea. Maximum volume drawn for catheterized subjects was below 15 ml, and
fingertip samples consisted of several drops from a new fingertip each time
collected in the Hemocue cuvettes.
Analysis
The va lues of Hb and apneic duration were compared with in each group via a
Student paired t-test, and the effects of apneas between groups were compared
via unpaired t-test. T-tests were corrected for multiple comparisons using
Bonferroni correction. Sta tistical signif icance was accepted at p < 0.05. Va lues
are expressed as absolute va lues and, when appropria te, a lso as percent changes
from reference. A Pearson correla tion was used to analyse the relationship
between apneic duration and Hb increase across the apnea series.
Results
Baseline Hb was similar and normally distributed in all groups, with divers
tending to have a h igher va lue (divers 150,1 g/L; skiers 145,5 g/L; untrained 146.9
g/L) (Figure 1). After the apneas, all groups responded with an increase of Hb (p
< 0.001; Figure 1). The largest increase was seen in divers (4.0 g/L–1; 2.7%)
compared to skiers (3.0 g/L–1; 2.1%), and untra ined (2.1 g/L–1; 1.4%, p < 0.05),
while skiers did not differ from either divers or untra ined (Figure 1). All groups
returned to baseline Hb va lues with in 10 minutes. Wh en the data from skiers and
untrained subjects was pooled into a single non-diver group, Hb was h igher in
divers th an in the non-diver group after 3 apneas (p < 0.05). Duration of the
initia l apnea was h igher for divers (144 s) th an for both skiers (87 s; p < 0.001)
and untrained (94 s; p < 0.01; Figure 2), but the durations were the same for skiers
and untra ined.
The apneic duration increased in all groups across th e 3-apnea series (p < 0.01 in
skiers, and p < 0.001 in divers and untra ined; Figure 2). The increase was more
pronounced in divers (43 sec, 20%) compared to skiers (26 sec, 12.2%, p < 0.05) and
untrained (24 s, 15.5%, p < 0.01; Figure 2). The apneic duration of the last apnea of
the series was 187 s in divers, 111 s in skiers, and 121 s in untra ined. Four subjects in
the divers’ group performed near-maximal apneas instead of maximal apneas
and were excluded from duration analysis. However, no correla tion was found
58
between apneic duration and Hb increase across the apnea series in any group.
Figure 1. Hb va lues (mean ± SE) before the first apnea (A1), after Apnea 3 (A3),
and 10 min after apneas in divers (n = 33), skiers (n = 13), and untrained subjects (n
= 32). ***S ignif icant changes of Hb from pre-apnea value with in groups, p <
0.001; a Significant difference between divers and pooled non-divers, p < 0.05.
59
Figure 2. Duration of apneas performed by divers (n = 29), skiers (n = 13), and
untrained subjects (n = 32). Significant changes of duration across 3 apneas with in
groups: ***p < 0.001; **p < 0.01. Difference between: a divers and untra ined; b
divers and skiers.
Discussion
Th is study shows th at elite divers, elite skiers, and untra ined subjects a ll respond
with elevated Hb to 3 maximal effort apneas, but with divers showing a more
powerful response. Since Hb increase after apneas h as previously been shown to
be absent in splenectomized subjects (Schagatay et a l., 2001), we suggest th a t
splenic contraction is likely to be the major factor beh ind the increase in Hb seen
in the current study. Th is is the first observation of an apnea-induced elevation of
Hb in non-diving endurance ath letes, a lthough the response is no greater th an
th a t of untra ined subjects. The higher increase in Hb after apneas in the divers
suggests th a t regular apnea practice could impart a specific tra ining effect,
affecting haematological responses to apnea in a manner th at differs from th a t of
exercise tra ining. It may be th a t repeated attempts to a level of transient
hypoxia, wh ich is not ach ieved during sub-maximal or even maximal exercise,
are needed to evoke a greater splenic response.
The apneic duration was h igher in the trained divers th an in other groups, and it
could be argued th a t longer apneas caused the further increase in Hb. However, it
h as previously been observed in untrained subjects th a t a longer duration of
60
individual maximal apneas did not affect the magnitude of Hb and Hct increases
(pilot study reported in Schatagay et a l., 2001). The lack of correlation between
duration of apneas and magnitude of Hb increase with in groups is in accordance
with these earlier findings. The pronounced hematological response in divers
could conceivably be a function of increased splenic contractility or a greater
amount of stored red blood cells prior to splenic contraction after apnea tra ining,
or both. However, differences could a lso be due to specif ic genetic tra its of persons
who choose diving as a sport. Known effects of apnea training include increased
cardiovascular diving response and apneic duration (Schagatay and Andersson
1998), but it is not known whether the spleen response can similarly be tra ined.
Longitudinal studies of apnea tra ining in previously untra ined subjects are needed
in order to explore th is possibility further.
In conclusion, a ll groups responded to apnea with an increased Hb, likely due to
spleen contraction, but th is was most preva lent in divers, suggesting th at these
responses may be enhanced by apnea tra ining but not by endurance ski tra ining.
Acknowledgments
Th is study was supported by The Swedish Na tiona l Center for Research in
Sports, Länsstyrelsen Västernorrlands län and Sundsvalls Sjukhus. All subjects
participated voluntarily and the study was conducted in compliance wit h
Swedish laws and eth ica l standards. We would like to th ank a ll participating
subjects, Dr. Johan Andersson, the Na tional Winter Sports Center in Östersund,
and all our laboratory personnel.
References
Bakovic, D., Va lic, Z., Eterovic, D., Vukovic, I., Obad, A., Marinovic-Terzic, I.,
and Dujic,
Z. (2003). Spleen volume and blood flow response to repeated breath- hold
apneas. J. Appl. Physiol. 95: 1460-1466.
Ekblom, B.T. (2000). Blood boosting and sport. Baillieres Best Pract. Res. Clin.
Endocrinol. Metab. 14: 89-98.
Froelich, J.W., Strauss, H.W., Moore, R.H., and McKusick, K.A. (1988).
Redistribution of viscera l blood volume in upright exercise in hea lt hy volunteers.
J. Nucl. Med. 29: 1714-1718.
Hurford, W.E., Hong, S.K., Park, Y.S., Ahn, D.W., Sh irak i, K., Mohri, M., and
Zapol, W.M. (1990). Splenic contraction during breath- hold diving in the Korean
ama. J. Appl. Physiol. 69: 932-936.
Laub, M., Hvid-Jacobsen, K., Hovind, P., Kanstrup, I.L., Christensen, N.J., and
Nie lsen, S.L. (1993). Spleen emptying and venous hematocrit in humans during
61
exercise. J. Appl. Physiol. 74: 1024-1026.
Neufeld, L., Garcia-Guerra, A., Sanchez-Francia, D., Newton-Sanchez, O.,
Ramirez- Villa lobos, M.D., and Rivera-Dommarco, J. (2002). Hemoglobin
measured by hemocue and a reference method in venous and capillary blood: A
va lidation study. Salud. Publica Mex. 44: 219-227.
Paiva Ad Ade, A., Rondo, P.H., Silva, S.S., and La torre Md Mdo, R. (2004).
Comparison between the HemoCue(R) and an automa ted counter for measuring
hemoglobin. Rev. Saude Publica 38: 585-587.
Schagatay, E., and Andersson J. (1998). Diving response and apneic time in
humans. Undersea Hyperb. Med. 25: 13-19.
Schagatay, E., Andersson, J.P., Ha l len, M., and Pa lsson, B. (2001). Selected
contribution: Role of spleen emptying in prolonging apneas in humans. J. Appl.
Physiol. 90: 1623-1629.
Schagatay, E., van Kampen, M., and Andersson, J. (1999). Effects of repeated
apneas on apneic time and diving response in non-divers. Undersea Hyperb. Med.
26: 143-149.
Wolski, L. (1998). The impact of splenic release of red cells on hematocrit
changes during exercise. PhD thesis. Vancouver: University of British Columbia.
62
PAPER III.
S hort-term effects of normobaric hypoxia on the human spleen. 2008. MX
Rich ardson, A Lodin, E Schagatay and J Reimers. European Journal of Applied
Physiology, published online Nov 28 2007, a head of print. Th is article is
reproduced with kind permission of Springer Science and Business Media. The
European Journal of Applied Physiology can be found online a t:
h ttp://www.springer.com.
Short-term effects of normobaric hypoxia on the human spleen
Matt X. Richardson1 , Angelica Lodin1 , Jenny Reimers2 and Erika Schagatay1
1) Department of Natural Sciences, Mid-Sweden University, Sundsvall, Sweden
2) X-ray clinic, Härnösand Hospita l, Härnösand, Sweden
Abstract
Spleen contraction resulting in an increase in circulating erythrocytes has been
shown to occur during apnea. Th is effect, however, h as not previously been
studied during normobaric hypoxia wh ilst breath ing. After 20 min of horizonta l
rest and normoxic breath ing, 5 subjects underwent 20-minutes of normobaric
hypoxic breath ing (12.8% oxygen) followed by 10 minutes of normoxic breath ing.
Ultrasound measurements of spleen volume and samples for venous hemoglobin
concentration (Hb) and hematocrit (Hct) were taken simultaneously at short
intervals from 20 min before until 10 min after the hypoxic period. Heart rate,
arteria l oxygen saturation (SaO2 ) and respiration rate were recorded
continuously. During hypoxia, a reduction in SaO2 by 34% (P<0.01) was
accompanied by an 18% reduction in spleen volume and a 2.1% increase in both Hb
and Hct (P<0.05). Heart rate increased 28% above baseline (P<0.05). With in 3
min after hypoxia SaO2 had returned to pre-hypoxic levels, and spleen volume,
Hb and Hct had a ll returned to pre- hypoxic levels with in 10 min. Respiratory
rate remained stable throughout the protocol. Th is study of short-term exposure
to eupneic normobaric hypoxia suggests th a t hypoxia plays a key role in
triggering spleen contraction and subsequent release of stored erythrocytes in
humans. This response could be beneficia l during early altitude acclimatization.
Keywords: hypoxia, spleen, hemoglobin, hematocrit, ultrasound
63
Introduction
The spleen is known to serve as a dynamic reservoir for red blood cells in many
species (Barcroft and Stephens 1927; Kramer and Luft 1951; Hurford et a l 1996). In
humans, the splenic reservoir is rela tively small compared to body size but sti l l
augments the body’s circulating hematocrit (Hct) and hemoglobin (Hb)
concentration when the spleen contracts. These spleen-associated increases have
been shown to occur during apnea (Schagatay et a l 2001; Bakovic et a l 2005) and
are reversible with in minutes (Schagatay et a l 2005). Similar effects occur
during high- intensity exercise (Laub et a l 1993). The benefit of such a transient
increase is an improved oxygen carrying capacity during situations where it
would be particularly he lpful, wh ile avoiding polycythemia when it is not
necessary through expedient storage of extra red blood cells by the spleen.
One such situation where fast Hb elevation would be helpful is acute hypoxia .
Rats h ave shown an increase in Hb originating from the spleen in response to
acute intermittent hypoxia, wh ich is a lso reversible and mediated by alp h a2adrenoreceptors (Kuwah ira et a l 1999). Kramer and Luft (1951) a lso showed in a
study on dogs th a t Hb increased directly correlated to the size of the spleen
following severe hypoxia. In humans, Richardson et al (2005a) h ave suggested
hypoxia to be involved in initia ting the spleen-associated Hb and Hct increase
during breath holding, together with one or more other stimuli.
Human splenic contraction is not well- investigated in general hypoxic situations,
and previous work has focused on chronic exposures to altitude. Using
ultrasonography, Sonmez et al (2007) confirmed th a t 3-6 months exposure to
a ltitude through relocation of lowlanders resulted in reduced splenic volume and
concomitant increases in Hb and Hct, but no specific focus was directed toward t he
ascent and descent phases. Rich ardson and associates (2007) discovered th at t he
transient elevation in Hb during apnea was attenuated and eventual ly
disappeared as individuals travelled to progressively h igher altitudes, during
which time an overa ll polycythemia occurred. During descent, the Hb increase
during apnea was greater th an at similar a ltitudes during ascent. Th is suggests
th a t the spleen may contract tonica lly with increasing altitude, eventually
reach ing a “fully contracted” state wh ich would not produce any further Hb
elevation even during additional hypoxic insult i.e. apnea. During descent, after
three weeks of increasing altitude, splenic reuptake of unnecessary red blood cells
may have resulted in a greater Hb increase through expulsion of its contents
during apnea.
64
There has been no previous study of the short-term effects of a hypoxic stimulus
during regular breath ing on the spleen. The a im of th is study was therefore to
revea l whether spleen contraction and related increases in Hb and Hct occur
during eupneic exposure to short-term normobaric hypoxia.
Methods
Subjects
Five volunteer subjects (4 men, 1 woman), of mean±SD age 27±3 years, heigh t
1.85±0.1 m, and mass 89±21 kg participated in the study. All were healthy nonsmokers who trained 5±3 hours per week for physica l f itness, and had a vit a l
capacity of 6.1±0.7 L. None had been at h igh a ltitude for at least three months
prior to the study. The study protocol was approved by the eth ics committee a t
Umeå University, Sweden and complied with the 2004 Declaration of Helsink i
for research involving human subjects.
Experimental protocol
Testing took place at Härnösand regional hospita l. After 20 min of horizonta l
rest and normoxic breath ing at sea level, subjects underwent 20-minutes of hypoxic
breath ing (hypoxia) followed by 10 minutes of normoxic breath ing. The hypoxia
was induced via mask (Hypoxico, Hypoxico Inc., New York USA) to an oxygen
content of 12.8%, wh ich simulated approximately 4100m above sea level. Al l
subjects remained prone and resting during the entire protocol. For analysis of Hb
and Hct 1.5 mL blood samples were taken via intravenous catheter (Venflon Pro,
Beckton-Dickson AB, Helsingborg, Sweden) placed at the antecubita l region, a t
the time periods illustrated in Figure 1. Ultrasound measurements of the spleen
(via HDI 3000, ATL A, Ph ilips Company, Bothwell, WA, USA) were taken in
three axes by a tra ined professional, simultaneously with the blood samples.
Arteria l oxygen saturation (SaO2 ) and heart ra te (HR) were measured
continuously throughout the protocol via pulse-oxymeter (WristOx 3100, Nonin
Medical Inc., Plymouth, Minnesota, USA). Breath ing movements were a lso
recorded through the protocol using a lab-developed pneumatic chest bellows.
The tota l volume of blood (including waste volume) obta ined from each subject
was approximately 40 mL, and the injected saline was approximately 20 mL. The
analysis of Hb and Hct was conducted via automated blood analysis unit (Micros
60 Analyzer, ABX Diagnostics, Montpellier, France).
65
Figure 1. The time course of the test protocol, ill ustrating the hypoxic and
normoxic periods and sampling points for blood and ultrasound measurements.
Data analysis
Spleen volume was calculated using a formula developed by S. Pilström (personal
communication): L!(WT-T2 )/3, where (L) is spleen length, ( W) width and (T)
th ickness obta ined from the ultrasound measurements. The formula is based on
the average shape of the spleen observed on the ultrasonic images and describes
the difference between two ellipsoids divided by two. Control values for spleen
volume, Hb and Hct were established immediate ly prior to the onset of hypoxia ,
and then compared with va lues from hypoxia and post- hypoxia. For SaO2 and
HR, the control va lue was established as a 60 s average recorded 90-30 s before
the onset of hypoxia, and compared with 30 s average va lues taken at the same
time points as the ultrasound and blood measurements. All va lues were logtransformed before comparison via t-test. 90% confidence intervals and p-va lues
were calculated for a ll comparisons. Pearson r correlations were performed
between spleen volume and blood parameters, and both p-va lues and the
like li hoods for true values based on Cohen’s threshold (Cohen 1988) were
calculated via purpose-made spreadsheet (Hopkins 2002).
Individual observed reductions in spleen volume during the hypoxic period were
used to represent expelled blood volumes, with a Hct of 80% (Laub et a l 1993).
The effects of adding th is to the circulating blood was calculated based on the
measured individual Hct before hypoxia and an estimated circulating blood
volume based on height and weight (Nadler 1962). The mean estimated Hct
66
increase for all measurements during the hypoxic period was then compared to
the measured increase in Hct to calculate the contribution from the spleen.
Results
Results are reported as mean±SD for anthropometric data, and as mean (90%
confidence limits) for changes and comparisons, with p-values.
Pre-hypoxia
Control va lues were established for SaO2 at 96.6±2.0%, HR at 62±17 bpm, and
respiration rate at 9.5 breaths per minute. During the pre-hypoxic 20 min period
of horizonta l rest, spleen volume increased and Hb and Hct decreased as the
subject’s blood volume and capillary exchange equilibrated as a result of the
change in position. Prior to the onset of hypoxia, spleen volume was 405±133 mL,
Hb was 134.4±9.0 g/L and Hct was 39.8±3.2 %BV.
Hypoxia
SaO2 fell and HR increased promptly during the initia l 5 min of hypoxic
exposure, to reach 33.9% (17.2) below baseline for SaO2 (64.4%; P<0.01) and 28.1
(19.5)% increase above baseline for heart rate by 20 min (P<0.05). Respiration
rate was 10.3 breaths per minute by the end of the hypoxic exposure (NS).
Mean spleen volume was reduced by 13.3% (6.4) after 3 min of hypoxic breath ing
(P<0.01; Figure 2). At 15 min it h ad been reduced by 17.9% (8.7) and after 20 min
the volume was reduced by 17.6% (8.7) below the control value (both P<0.01).
Hb had increased by 2.1% (1.0) to 137.2±9.0 g/L at 3 min (P<0.05), and was
136.2±9.9 g/L after 20 min of hypoxic breath ing (Figure 2). Hematocrit followed
the same response pattern as Hb, increasing by 2.1% (0.9; P<0.01) to 40.6±3.4 %BV
a t 3 min hypoxia and was 40.2±3.3 %BV after 20 min of hypoxic breath ing.
67
Figure 2. Spleen volume and Hb during pre- hypoxia, hypoxia and post-hypoxia
for N=5. For comparison to control va lues, P<0.05 is indicated by * and P<0.01 by
* *.
Post- hypoxia
There was a prompt return of SaO2 and HR to the control va lues following the
onset of the post-hypoxic period, and HR va lues were -3.8% (10.6; NS) and 12.7% (11.5; NS) compared to control at 5 and 10 minutes post- hypoxia ,
respectively. Respiration rate was 9.2 breaths per minute ten minutes after
exposure.
Spleen volume was stil l reduced by 22.2% (11.3; P<0.01) and 13.6% (11.0, P<0.05)
a t 1 and 5 min post- hypoxia, respectively (F igure 2). Corresponding SaO2 va lues
were
-6.1% (3.6; p=0.05) and -2.9% (4.5; NS) compared to the control va lue. At 10 min
post- hypoxia spleen volume was -5.8% (9.7; NS) and SaO2 was -1.4% (1.8; NS) ,
compared to control va lues.
Compared to the control va lue, Hb was still elevated by 3.4% (2.0; P<0.05) and
3.3% (2.0; P<0.05) at 1 and 5 min post- hypoxia, respectively, but gradual ly
decreased throughout the post- hypoxic period and by 10 min post- hypoxia was
1.8% (2.2; NS) above control va lue (Figure 2). Hct followed a similar pattern,
68
with an increase of 3.2% (2.1; P<0.05) and 2.6% (±2.1; NS) at 1 and 5 min posthypoxia, respectively, gradually decreasing to 0.9% (1.7; NS) from the control
va lue at 10 minutes post- hypoxia.
Correlation between spleen volume and blood parameters
The Pearson r correla tion between spleen volume and Hb was -0.51 (P<0.05),
representing 74% likelihood th at the true va lue of the correla tion is clinica lly
negative according to Cohen’s threshold of 0.1. For th e correlation between spleen
volume and Hct of -0.42 (P<0.1), th is represents 69% likeli hood th at the true
va lue of the correlation is clinica lly negative according to Cohen’s threshold of
0.1. The spleen was estimated to contribute to 60% of the observed overa l l
increase in Hct during hypoxic exposure.
Discussion
Th is study shows th at spleen contraction develops rapidly upon onset of
normobaric hypoxia in humans. An ejection of stored erythrocytes results, as
demonstrated by the corresponding elevations in Hct and Hb. These responses are
reversed with in approximately 10 minutes upon return to normoxia. This result
suggests, together with other recent observations (Rich ardson et a l 2007) th a t
spleen contraction may be part of the early acclimation to a ltitude in humans,
whereby increases in circulating hematocrit could he lp overcome the init i a l
phase of hypoxia before erythropoietic effects may develop. These early
elevations in Hct and Hb were previously suggested to be associated wit h
changes in plasma volume caused by the hypoxia- induced responses of sodiumand water-regulating hormones (Schmidt 2002) or dehydration due to increased
ventilation (Heinicke et al 2003). However, in the current study it appears th a t
the spleen is an important source of extra circula ting erythrocytes during
hypoxia.
The spleen contraction generally follows the development of arteria l hypoxia
suggesting, in accordance with earlier findings, th a t hypoxia may be an
important input for its development (Rich ardson et al 2005a). In th a t study,
removal of the apneic hypoxia by inspiration of oxygen attenuated the rise in Hb
and Hct, but some response remained showing tha t other factors are also
involved. A possible contributing factor in the apneic situation could be
hypercapnia, but in the current context hypercapnia is h ighly unlikely, as the
respiration rate was unchanged during the hypoxic exposure. Catecholamines or
sympathetic nervous system activity rela ted to stress might a lso contribute to
splenic contraction, a lthough the conduct of the current experiment was not
69
considered by subjects to be stressful (personal communication). Apneic situations
are likely perceived as more stressful, but changes in Hb and Hct in some previous
apneic studies (Schagatay et a l 2005; Rich ardson et a l 2005b) were of a similar
magnitude as in the present study.
The decrease in SaO2 was most intense during the initia l 5 min, wh ich may
expla in the powerful spleen responses during th i s period. Interestingly, it
appears th a t both the initia tion and interruption of the hypoxic exposure
initia l ly evoke a spleen contraction, as indicated both by the spleen volume and
blood boosting response. We speculate th a t th is could either be a response to the
change in inspired oxygen, irrespective of direction, or a general arousal response
associated with the placement and removal of the breath ing mask. However,
the direction of the response after the initia l ph ase in these two situations is
clearly different, showing tha t during hypoxia spleen contraction preva ils.
The question if there are any physiologica l differences between hypobaric
hypoxia compared to normobaric hypoxia for the same ambient PO2 is still under
debate. In addition to lowered SaO2 , Savourey and associates (2003) reported
changes in mainly respiratory parameters at 4500 m a ltitude, such as increased
breath ing frequency, tidal volume and minute ventila tion, resulting in
hypocapnia and a higher blood alka losis. Wh i le blood chemistry and tida l
volume were not measured in the current protocol, respiration rate remained
unchanged during the hypoxia, suggesting th at no hypercapnia occurred.
Th is laboratory study supports the observations of a previous study in 3 climbers
(Rich ardson et a l 2007) where apnea-induced eleva tions in Hb were found to
diminish with increased a ltitude, conceivably due to emptying of the spleen
with increasing hypoxia. The present study suggests th a t even during normobaric
eupnea, hypoxia is a trigger for spleen contraction and increased volume of
circulating erythrocytes. Th is may provide a rapid means of increasing oxygen
carrying capacity during early phases of a ltitude adaptation as well as during
acute hypoxic situations.
Acknowledgements
The authors would like to th ank our volunteer subjects for participating, Mr. KaYu Law, Ms. Torborg Jonsson and Mr. Håkan Norberg for technica l assistance, and
the Swedish Na tional Center for Research in Sports (CIF) for financia l support.
Th is study complies with Swedish laws and eth ical standards.
70
References
Bakovic D, Eterovic D, Saratlija-Novakovic Z, Pala da I, Va lic Z, Bilopavlovic
N, Dujic, Z (2005) Effect of human splenic contraction on varia tion in circulating
blood cell counts. Clin Exp Pharmacol Physiol 32: 944-951
Barcroft J, Stephens J (1927) Observations upon the size of the spleen. J Physiol
64: 1-22
Cohen J (1988) Sta tistica l power analysis for the behavioral sciences. 2nd
edition. Lawrence Erlbaum, New Jersey
Cook S, Alaf i M (1956) Role of the spleen in acclima tization to hypoxia. Am J
Physiol 186(2): 369-372
Heinicke K, Prommer N, Cajiga l J, V iola T, Behn C, Schmidt W (2003) Long-term
exposure to intermittent hypoxia results in increased h aemoglobin mass, reduced
plasma volume, and elevated erythropoietin plasma levels in man. Eur J Appl
Physiol 88: 535-543
Hopkins W (2002) Ca lculating likely (confidence) limits and likelihoods for true
va lues (Excel spreadsheet). In: A new view of statistics, sportsci.org: Internet
society for sport science, sportsci.org/resource/stats/xcl.xls
Hurford W, Hochachka P, Schneider R, Guyton G, Stanek K, Zapol D, Liggins G,
Zapol W. Splenic contraction, catecholamine release, and blood volume
redistribution during diving in the Weddell seal. J Appl Physiol 80(1): 298- 306
Kramer K, Luft U (1951) Mobilization of red cells and oxygen from the spleen in
severe hypoxia. Am J Physiol 165: 215-228
Kuwa h ira I, Kamiya U, Iwamoto T, Moue Y, Urano T, Ohta Y, Gonzalez N (1999)
Splenic contraction-induced reversible increase in hemoglobin concentration in
intermittent hypoxia. J Appl Physiol 86(1): 181-187
Laub M, Hvid-Jacobsen K, Hovind P, Kanstrup I, Christensen N, Nie lsen S (1993)
Spleen emptying and venous hematocrit in humans during exercise. J Appl
Physiol 74: 1024-1026
Nadler S, Hidalgo J, Bloch T (1962) Prediction of blood volume in normal human
adults. Surgery 51: 224-232
Rich ardson M, de Bruijn R, Schagatay E (2005a) Hypoxia – a trigger for spleen
contraction? In: Desola J (ed) ICHM-EUBS Proceedings, Barcelona, Spa in.
Rich ardson M, de Bruijn R, Holmberg HC, Björklund G, Haughey H, Schagatay E
(2005b) Increase of hemoglobin concentration after maximal apneas in divers,
skiers and untra ined humans. Can J Appl Physiol, 30(3): 276-281
Rich ardson M, Schagatay E (2007, in publication) Altitude attenuates apneainduced increase in hemoglobin concentration. In: Mejkavic E (ed) ICEE conference
proceedings, Portoroz, Slovenia.
71
Savourey G, Launay J, Besnard Y, Guinet A, Travers S (2003) Normo- and
hypobaric hypoxia: are there any physiological differences? Eur J Appl Physiol
89: 122-126
Schagatay E, Andersson J, Ha llén M, På lsson B (2001) Selected contribution; Role
of spleen emptying in prolonging apneas in humans. J Appl Physiol 90: 1623-1629
Schagatay E, Haughey H, Reimers J (2005) Speed of spleen volume changes
evoked by seria l apneas. Eur J Appl Physiol 93: 447-452
Schmidt W (2002) Effects of intermittent exposure to high altitude on blood
volume and erythropoietic activity. High Alt Med Biol 3: 167-176
Sonmez G, Ozturk E, Basekim C, Mutlu H, Kilic S, Onem Y, Kizilkaya E (2007)
Effects of a ltitude on spleen volume: sonograph ic assessment. J Clin Ultrasound
35:182-185
72
PAPER IV.
Hypoxia augments apnea-induced increase in hemoglobin concentration and
hematocrit. 2008. MX Rich ardson, R deBruijn, and E Schagatay. Submitted for
publication in the European Journal of Applied Physiology.
Hypoxia augments apnea-induced increase in hemoglobin
concentration and hematocrit
Matt X. Richardson, Robert de Bruijn, and Erika Schagatay
Environmenta l Physiology Group, Mid Sweden University, Östersund, Sweden.
Abstract
Increased hemoglobin concentration (Hb) and hematocrit (Hct), attributable to
spleen contraction, raise blood gas storage capacity during apnea, but the
mechanisms th at trigger th is response have not been clarif ied. We focused on the
role of hypoxia in triggering these Hb and Hct elevations. After horizonta l rest
for 20 min, 10 volunteers performed 3 maximal apneas spaced by 2 min, each
preceded by a deep inspiration of air. The series was repeated using the same
apneic durations but after 1 min of 100% oxygen (O2 ) breath ing and O2 inspiration
prior to each apnea. Mean apneic durations were 150s, 171s, and 214s for apneas 1,
2, and 3, respectively. Relative to pre-apnea values, the mean post-apneic
arteria l O2 saturation nadir was 84.7% after the a ir tria l and 98% after the O2
tria l. A more pronounced elevation of both Hb and Hct occurred during the a ir
tria l: after apnea 1 with a ir, mean Hb had increased by 1.5% (P<0.01), but no
clear increase was found after the first apnea with O2 . After the th ird apnea
with air Hb had increased by 3.0% (P<0.01), and with O2 by 2.0% (P<0.01). After
the first apnea with a ir Hct had increased by 1.9% (P<0.01) and after 3 apneas
by 3.0% (P<0.01), but Hct did not change significantly in the O2 tria l. In both
tria ls, Hb and Hct were at pre-apneic levels 10 min after apneas. Diving
bradycardia during apnea was the same in both tria ls. We conclude th at hypoxia
contributes to spleen contraction during apnea, likely through chemosensorrela ted sympathetic output. There are, however other factors involved th a t
trigger spleen contraction even in the absence of hypoxia.
73
Key Words: Breath - hold, oxygen saturation, spleen contraction, diving
response, diving reflex
Introduction
Diving mammals, e.g. the Weddell seal, are known to expel red blood cells stored
in the spleen during dives, which ra ises the circulating hemoglobin concentration
(Hb; [82]. In humans, an increase in circulating Hb and hematocrit (Hct) can also
be achieved by apnea- induced splenic contraction [155-157]. Th is increase results
during both solitary [77] and repeated apneas spaced by short intervals [50, 76]. In
the la tter case, the increase in Hb and Hct reach es a maximum after 3 – 4
maximal effort apneas [50]. Previous work [78] described a normal distribution of
Hb and Hct responses to a three-apnea series in untrained breath- holders, where
the 25th to 75th percentile showed a 0.7% to 3.7% increase in Hb, and 0.4 to 4.0% in
Hct. Spleen contraction has been shown to account for approximately 60% of these
increases (Rich ardson et a l. 2008; Schagatay et a l. 2001). A reversa l of the spleen
contraction is seen 8-9 min after cessation of apneas [158], with a concomitant
return to baseline of Hb and Hct. Th is shows th a t th e Hb and Hct increase is not
due to increased diuresis or sweating, and constancy of protein levels shows th a t
it is not due to extravasation of plasma by increased filtration [50]. The increased
circulating Hb and Hct may a lso expla in the prolonged apneic duration in
subsequent attempts, possibly through an increased capability to take up and
store oxygen (O2 ) and also to buffer carbon dioxide (CO2 ).
Bakovic and associates [35] have shown th at the mobilization of blood from the
spleen during apnea is an active process in humans largely independent of
arteria l inflow, suggesting the presence of a centrally-media ted feed-forward
mechanism. Their later work suggested th a t the initi a l contraction was, at least
during apneas at large lung volumes, caused by downloaded baroreflex during
decreases in blood pressure and cardiac output (Palada et a l. 2007). Eupneic
hypoxia in cats and dogs appears to have similar effects as electric stimulation
of the splenic nerve which results in an emptying of the spleen of its stored blood
contents [81, 159]. Hoka and associates [68] suggested th a t the sympathetic
efferent nerve activity, augmented by the stimula tion of chemoreceptors,
contributes greatly to those hypoxic changes. Human spleens possess
adrenoreceptors in the vasculature and capsule and injection of epinephrine has
shown to cause contraction (Ayers et a l. 1972). Still, there is little consensus about
the actual trigger of human splenic contraction during apnea.
74
The decrease in SaO2 during prolonged apnea could be a possible mechanism th a t
triggers spleen-related increases in Hb and Hct. Qvist and associates [160]
completed studies of Korean ama divers’ hematological characteristics both
during diving to depth and breath holding in air. Although it was not the main
focus of their study, during the repeated working dives to depth th a t produced a
reduction in SaO2, they noted increases in Hb and Hct in the divers. However,
these increases were not present in resting breath- holds of similar duration in air,
which showed no decrease in SaO2 .
Pre-breath ing of oxygen has been shown to extend brea th- hold duration [12, 161]
and thus delays the development of a hypoxic stimulus. The tota l body storage of
O2 , including tissue, blood, and lung compartments, of a 70 kg male, is
approximately 1500 mL, with ~370 mL in the lungs and ~280 mL in the blood
(Ra hn 1964). Breath ing 100% O2 increases these tota l body stores through
replacement of nitrogen in the lungs with O2 , wh ich would be reflected as a
slower depletion of SaO2 . O2 -enhanced breath holding does not, however, appear
to affect CO2 production, wh ich appears to develop similarly as during breat h
holding without O2 , as demonstrated by Corkey and associates [162]. They
observed th at the amount of CO2 added to the lungs during resting apneas was
rela tively constant at a rate of 0.32 L min-1 , independent of inha led O2
concentration (21%, 50% or 100%).
In th is study, we examined the possibility tha t desa turation during prolonged
apnea is involved in triggering the spleen related Hct increase. In order to test
th is, we eliminated the desaturation during apnea by pre-apnea O2 breath ing in
one tria l, and compared the hemoglobin response to a tria l with subjects
breath ing air before apneas. The cardiovascular response to apnea was also
compared between these two conditions.
Methods
Subjects
Ten hea lthy male subjects of mean ± SD age 24 ± 5 years, mass 80.6 ± 14.8 kg, and
he ight 1.83 ± 0.72 m volunteered for th is study. Subjects signed a consent form
after being informed of the experimenta l protocol, wh ich h ad been approved by
the regional human research eth ics board at Umeå University, Sweden and in
accordance with the Declaration of Helsinki. All were non-smokers and tra ined
7.3 ± 7.0 h per week for general fitness. Subjects had experience in breath holding
a t a level averaging 3 on a scale of 1 to 5 (1=no breath- holding experience,
3=regular breath- hold diving but less th an 2h per week, 5=world-class apneist;
Schagatay 1996). V ita l capacity was 6.2 ± 1.0 L in the standing position and 5.8 ±
1.1 L in the experimenta l, prone position.
75
Experimental procedure
The subjects completed 2 experimenta l tria ls spaced by at least 4 hours. The first
tria l consisted of apneas of maximal voluntary duration after breath ing air prior
to each apnea (a ir). In the second tria l, the apneic times from the first session
were reproduced, with the subjects breath ing 100% O2 at a normal ventilatory
rate for 1 min prior to each apnea (oxygen). This pre-breath ing of oxygen was
intended to mainta in a h igher SaO2 during apnea th a n in the a ir tria l. Due to t he
like li hood of greatly prolonged breath- holding durations in oxygen [12, 161], the
a ir tria l was completed first so tha t the apneic duration could be reproduced.
Subjects reported to the laboratory without having ea ten or consumed caffeine a t
least 2 h prior to testing. The vita l capacity of th e subject was measured via
spirometer (Compact II, Vita lograph, Buckingham, England) and an intravenous
catheter (Venflon Pro, Beckton-Dickson, Helsingborg, Sweden) was placed in the
antecubita l region.
Subjects were prone for the entire duration of the tria ls, beginning with a 20 min
period of horizonta l rest, followed by a 2 min countdown to the f irst of three
apneas. In the a ir tria l, a noseclip was placed at 30 s remaining in the countdown,
and with 15 s remaining in the countdown the subject was administered a
mouthpiece for measurement of inspired and expired gases. In the O2 tria l, a nose
clip was administered just prior to 60 s remaining in the countdown, after which
subjects were administered a mask for breath ing oxygen with flow rate at ~10 L
min-1. For both tria ls, at the end of the 2 min countdown, the subject exha led fully
followed by a deep but not maximal inspiration and then began the apnea.
Hyperventilation was not permitted in either of the series. Upon completion of
the apnea, subjects expired fully through the mouthpiece and then resumed
normal breath ing. All apneas were spaced by 2 min of rest. In the a ir tria l,
apneas were conducted without time cues and to maximum duration, wh ile the O2
tria l reproduced the times ach ieved during the air tria l and the subject was
informed when to stop the apneas.
Blood samples (2 ml) were taken via the intravenous catheter 2 min before the
first apnea, immediate ly after the first apnea, and immediately after as well 10
min following the th ird apnea. The catheter was rinsed with 2 ml sa line
following each sample. The tota l volume of blood (including waste volume)
removed for each subject was approximately 25 ml, and the injected saline was
approximately 12 ml. Blood samples were analyzed for Hct and Hb via
76
automated blood analysis unit (Micros 60 Analyzer, ABX Diagnostics,
Montpellier, France).
From 2 min prior to apneas until 2 min post apneas SaO2 and heart rate (HR) were
measured via pulse oximeter (Biox 3700e, Ohmeda, Madison, WI, USA) and using
a BioPac MH100A CE multi-channel data acquisition system (BioPac Systems,
Goleta, CA, USA. Inha led and exha led O2 and CO2 were measured before and
after each apnea via gas analyzer (Normocap Oxy, Datex-Ohmeda, Helsink i,
Finland). A lab-developed chest bellows was used to monitor the depth and
frequency of breath ing movements.
Analyses
The va lues used for nadir SaO2 were the lowest obta ined 2-second averages after
apneas. For respiratory parameters O2 and CO2 percentages in the last inspired
breath before and the first expired breath after the apneas were used. Hb and
Hct obta ined 2 min before the first apnea were used as control. Mean changes from
control were used to compare changes with in each tria l, and pooled mean changes
from apneas were used to compare between tria ls.
Control va lues for cardiovascular and respiratory parameters were calculated as
averages from the period 60-30 s before each apnea. Apneic values for HR were
obta ined from the period from 30 s after breath- hold onset until the cessation of
the breath- hold, in order to obta in a fully developed diving response (Jung and
Stolle 1981). The mean changes from all apneas were used to calculate individua l
means.
Subjects served as their own controls and effects were expressed as percent changes
from control. All variables were log transformed before analysis to reduce nonuniformity of error. Excel templates were used for all calculations, purposedesigned for analyses using physiological data [163]. With in-tria l comparisons
for changes resulting from apneas were completed with control va lues using
ANOVA-based pair-wise comparisons with Bonferroni corrections for both a ir
and oxygen tria ls. Separate between-tria l comparisons were conducted using
ANOVA-based pair-wise comparisons with Bonferroni corrections for Hb and
Hct after the first and th ird apneas. Cardiovascular ch anges were averaged over
the three apneas and then compared between tria ls using the same method.
Acceptable differences were set at p<0.05 for in the comparisons.
Results
Apneic duration
77
All subjects increased their apneic duration (P<0.01) across the three air tria ls
with mean ± SD apnea times of 150 ± 68 s, 171 ± 63 s, and 214 ± 106 s for apneas 1,
2, and 3, respectively. All subjects successfully repea ted these durations during
the O2 tria l.
Arteria l desaturation
Mean ± SD control va lues for SaO2 in the a ir and oxygen tria ls were similar prior
to the start of apneas, at 96.8 ± 1.1 % and 98.5 ± 0.6 %, respectively (NS). The
mean saturation nadir for apneas with a ir was 84.7 ± 9.1 % (P<0.01), but SaO2 did
not change from control in the O2 tria l, at 98.4% ± 0.8. During apneas, SaO2 va lues
were lower in the air tria l compared to the O2 tria l (P<0.01).
Hematological response
One subject’s blood samples are not included in the hematological analysis due to
catheter coagulation during the air tria l, and N=9 for th is section. Mean ± SD
control va lues for Hb before apneas for a ir and oxygen tria ls were not different, a t
134.1 ± 0.8 g/L and 131.7 ± 0.8 g/L respectively. After apnea 1 with a ir, mean Hb
h ad increased (P<0.01), but no clear increase was found after the f irst apnea in
the O2 tria l (Figure 1). After the th ird apnea Hb had increased from control in
both the a ir tria l and the O2 tria l (P<0.01). The increase in Hb resulting from
apneas was greater in the air tria l (p<0.05). 10 min after apneas, Hb was not
different from control values for either air or oxygen tria ls, nor were they
different from each other.
Control va lues for Hct were 40.9 ± 2.0 %BV and 40.6 ± 2.4 %BV for air and oxygen
respectively, and not different from each other. After apnea 1 with a ir Hct h ad
increased (P<0.01) and continued th is increase after 3 apneas (P<0.01), but Hct
did not change significantly in the O2 tria l (F igure 2). In both a ir and oxygen
tria ls Hct va lues were at control levels 10 min after the fina l apnea, and va lues
were the same for the air and oxygen tria ls.
78
Figure 1. Mean ± SD percent changes in Hb from control pre-apneic va lues for 9
subjects. Blood sampling time points are labelled on the horizonta l axis.
S ignificant differences from control are indicated by ** for p<0.01.
Figure 2. Mean ± SD percent changes in Hct from control pre-apneic values for 9
subjects. Blood sampling time points are labelled on the horizonta l axis.
S ignificant differences from control are indicated by ** = p<0.01.
Cardiovascular Responses
Mean ± SD control va lues of HR for the a ir and oxygen tria ls were not different a t
77.5 ± 16.9 bpm and 76.0 ± 8.5 bpm , respectively. The HR showed simil ar
reductions from control va lues during apnea in the air tria l, at 14.8 ± 8.8 %, and in
79
the O2 tria l, at 14.6 ± 9.8 %. Between apneas HR was restored to the resting
eupneic level in both tria ls, and there were no changes of apneic or recovery
levels throughout the series. During apneas, there was no difference in HR
between the air and oxygen tria ls.
Respiratory Parameters
Inspired and expired O2 and CO2 percentages before and after apneas are listed in
Table 1. Both inspired and expired O2 were h igher in the O2 tria l (p<0.01).
Inha led and exha led CO2 were similar between the two tria ls.
Table 1. Inspired and expired O2 and CO2 percentages before and after apneas for
10 subjects.
Inspired O2
Expired O2
Inspired CO2
Expired CO2
* *P<0.01
AIR
20.7% (0.04)
10.0% (1.0)
0.1% (0.02)
7.2% (0.2)
OXYGEN
87.0% (1.4)**
42.4% (11.9)**
0.1% (0.03)
7.5% (0.2)
Discussion
The greater increase in Hb and Hct after apnea with a ir suggests th a t the
associated decrease in SaO2 imposes an essentia l stimulus for the increase in
circulating erythrocytes. Previous studies have shown th a t these Hb and Hct
elevations during apnea are mainly due to spleen contraction (Bakovic et al. 2005;
Espersen et al. 2002; Schagatay et a l. 2001). The recovery after 10 min shows th a t
these elevations are not an effect of increased diuresis, which is in accordance
with earlier findings [50]. A constant tota l protein level across seria l apneas
indicates th at changes in Hb and Hct are not a result of hemoconcentration caused
by extravasation of plasma due to increased filtrati on (Schagatay et a l. 2001).
Thus, a decrease in SaO2 appears to be a major stimulus evoking spleen contraction
in humans.
In mammals, it h as been shown th at the spleen contracts in response to
sympathetic nervous stimulation (Donald and Aarhus 1978; Greenway and Stark
1968; Hurford et al. 1996), someth ing th a t can also be induced by hypoxia (Hok a
et al. 1989). As an example, renal sympathetic nerve activity h as a lso been
shown to increase during apnea, and was directly correla ted to carotid
chemoreceptor activity (O’Donnell et a l. 1996). The splenic nerve, itself
composed almost entirely of sympathetic nerve fibers, may thus be affected by
increased hypoxia- induced sympathetic output, resulting in contraction. The
80
carotid bodies are thought to react with in seconds to even small changes in PO2,
whereas other hypoxic rela ted cellular responses in the rest of the body work on
a considerably longer time scale i.e. hours (Prabh ak ar 2006), making any
hypoxia-rela ted spleen contraction more likely to be associated wit h
chemoreceptor-related output.
Th is would not expla in the increased in Hb in the O2 tria l, however, as any
prospective carotid chemoreceptor stimulation would have been removed by
virtue of the hyperoxic intervention and lack of Sa O2 change. An additiona l
factor such as arteria l pCO2 may play a role however, since a residual carbon
dioxide level is a lways present during apnea, even in hyperoxia (Lin et a l. 1983,
Corkey et al. 1998), and thus stimulation of CO2 -sensitive centra l chemoreceptors
may also occur, even in the absence of peripheral chemoreceptor responses. CO2
sensitive neuronal mechanisms have been suggested to generate sympathetic tone
(Trzebski and Kubin 1981). Lin and associates [164] hypothesized th a t
hypercapnia may enhance sympathoadrenal catecholamine release. Thus it is
a lso conceivable th at sympathetic output from central chemosensation of
increasing CO2, wh ich a lso occurs on an rapid time-scale. can result in splenic
stimulation resulting in contraction.
The function of the spleen rela ted Hct and Hb release in diving mammals is
like ly to prolong aerobic dive duration (Cabanac et a l. 1997) and there is a lso
evidence of an apnea prolonging effect in humans across repeated apneas
(Schagatay et al. 2001; Bakovic et a l. 2003). The grea ter increase in Hb and Hct
found in the a ir tria l could thus be assistive in subsequent apneic attempts. The
release of additional red blood cells from the spleen during sensed reductions in
hemoglobin saturation would provide additional O2 storage and buffering
capabilities to prolong subsequent apneas, which is typica l in repeated apnea
a ttempts (Schagatay et al. 1999; 2001; 2005). The circulating red blood cells
would a lso accelerate recovery between apneas and allow apneas to be spaced by
short pauses. Without hypoxia, as during apnea in th e O2 tria l, an injection of red
blood cells of similar magnitude would not be required, as O2 is no longer a
limiting factor. In th is way an unnecessary increase in blood viscosity and cardiac
stra in would be avoided.
In th is study the tota l apneic duration increased by 64 s between the first and
th ird apnea, but it h as previously been shown th at a prolonged “struggle phase”
of the apnea accounts for a large portion of the apnea time increase, due mainly to
psychologica l factors (Schagatay et a l. 1999). Nonethe less, a typica l ejection of
stored blood in the spleen in a human with 5L blood volume could increase
arteria l O2 content by 2.8 to 9.6% (Stewart and McKenzie 2002), an appreciable
81
effect for those undertak ing acute hypoxic activity. The 3% increase in Hct
observed in the present study falls with in th is range, and assuming a 400 ml
spleen volume (Richardson et a l. 2008) with 50% reduction after apnea 3, and a
VO2 of 215 ml/min (Byrne et a l. 2005), the increase in apneic duration
a ttributable to the increase in arteria l O2 supply could be approximately 18 s.
Added to th is would be the effect of the increased CO2 buffering capacity, thus a
significant part of the increase in apneic duration could be attributable to the
observed spleen contraction induced Hct increase.
Conclusions
We conclude th a t hypoxia is an important factor for development of the spleenrela ted Hb and Hct increase during apnea, beyond additional apneic costimulus/stimuli. Hypercapnia is one such possible co-stimulus, which may a lso
provide an initia ting or moderating influence on these increases.
Acknowledgements
The authors would like to th ank our volunteer subjects for participating, Dr. Jenny
Reimers, Ms. Torborg Jonsson and Mr. Håkan Norberg for technica l assistance, and
the Swedish National Center for Research in Sports (CIF) and the County
Administrative Board of Västernorrland, Sweden for financia l support. Th is
study complies with Swedish laws and eth ical standards.
References
Ayers AB, Davies BN, Wit hrington PG (1972) Responses of the isolated, perfused
human spleen to sympathetic nerve stimulation, catecholamines and
polypeptides. Br J Pharmacol 44:17-30
Bakovic D et a l (2003) Spleen volume and blood flow response to repeated
breath- hold apneas. J Appl Physiol 95:1460-1466
Bakovic D et a l (2005) Effect of human splenic contraction on variation in
circulating blood cell counts. Clin Exp Pharmacol Physiol 32:944-951
Byrne NM et al (2005) Metabolic equiva lent: one size does not fit a ll. J App l
Physiol 99:1112-1119.
Cabanac A, Folkow LP, Blix AS (1997) Volume capacity and contraction control of
the sea l spleen. J Appl Physiol 82:1989-1994
Corkey WB et a l (1998) PaO2 and PaCO2 during brea th hold. In: Proceedings of
the UHMS 1998 Annual Meeting, Seattle, WA. http://archive.rubiconfoundation.org/648
82
Donald DE, Aarhus LL (1974) Active and passive release of blood from canine
spleen and small intestine. Am J Physiol 227:1166-1172
Engel GL et al (1946) Voluntary breath holding. II. The rela tion of the maximum
time of breath holding to the oxygen tension of the inspired a ir. J Clin Invest
25:729-733
Espersen K et a l (2002) The human spleen as an erythrocyte reservoir in divingrela ted interventions. J Appl Physiol 92:2071-2079
Greenway CV, Stark RD (1968) Vascular responses of the spleen to sympathetic
nerve stimulation in the anaesthetized cat. J Physiol 194:1-2
Hoka S et al (1989) Regional venous outflow, blood volume, and sympathetic
nerve activity during severe hypoxia. Am J Physiol Heart Circ Physiol 256:162170
Hopkins W (2000) A new view of statistics. Internet Society for Sport Science
h ttp://www.sportsci.org/resource/stats/xcrossover.xls
Hurford WE et al (1996) Splenic contraction, catecholamine release, and blood
volume redistribution during diving in the Weddell seal. J Appl Physiol 80:298306
Jung K, Stolle W (1981) Be h avior of heart rate and incidence of arrhythmia in
swimming and diving. Biotelem Patient Monit 8:228-239
K locke FJ, Ra hn H (1959) Breath holding after breath ing of oxygen. J App l
Physiol 14:689-693
Lin YC, Sh ida KK, Hong SK (1983) Effects of hypercapnia, hypoxia, and
rebreath ing on heart rate response during apnea. J Appl Physiol 54:166-171
O’Donnell CP et a l (1996) Ref lex stimulation of renal sympathetic activity and
blood pressure in response to apneas. Am J Respir Crit Care Med 154:1763-1770.
Palada I et a l (2007) Spleen and cardiovascular function during short apneas in
divers. J Appl Physiol 103:1958-1963
Prabhakar, NR (2006) O2 sensing at the mammalian carotid body: why multiple
O2 sensors and multiple transmitters? Exp Physiol 91:17-23.
Qvist J et a l (1993) Arteria l blood gas tensions during breath- hold diving in the
Korean ama. J Appl Physiol 75:285-293
Ra hn H (1964) Oxygen stores of man. In: Oxygen in the Animal Organism, Dickens
E (ed), Pergamon/MacMillan, Oxford, pp 609-618
Rich ardson MX et a l (2003) Hematologica l response pattern associated wit h
maximal-duration apnea series in untra ined subjects. In: Proceedings of the 24th
Annual Scientif ic Meeting of the European Underwater and Baromedical Society ,
Jansen EC, Mortensen CR, Hyldegaard O (eds), EUBS, Copenhagen, pp 39
Rich ardson MX et a l (2008) Short-term effects of normobaric hypoxia on the
human spleen. Eur J Appl Physiol, DOI 10.1007/s00421-007-0623-4
83
Schagatay E (1996) The human diving response: effects of temperature and
tra ining [Doctoral thesis]. Lund: Department of Animal Physiology, Lund
University
Schagatay E, van Kampen M, Andersson JPA (1999) Effects of repeated apneas on
apneic time and diving response in non-divers. Undersea Hyperb Med 26:143-149
Schagatay E et a l (2001) Role of spleen emptying in prolonging apneas in humans.
J Appl Physiol 90:1623-1629
Schagatay E, Haughey H, Reimers J (2005) Speed of spleen volume changes
evoked by seria l apneas. Eur J Appl Physiol 93:447-452
Stewart IB, McKenzie DC (2002) The human spleen during physiological stress.
Sports Med 32:361-369
Trzebski A, Kubin L (1981) Is the centra l inspiratory activity responsible for
pCO2-dependent drive of the sympathetic discharge? J Auton Nerv Syst 3:401420.
84
PAPER VII.
Cardiovascular and hematologica l adjustments to apneic diving in humans: Is the
spleen response part of the diving response? 2006. E Schagatay, MX Richardson,
R deBruijn and JPA Andersson. In: Lindholm, P, Pollock NW, Lundgren CEG, eds.
Breath- hold diving. Proceedings of the Undersea and Hyperbaric Medica l
Society/Divers Alert Network 2006 June 20-21 Workshop, p 108-115.
Cardiovascular and hematological adjustments to apneic diving
in humans – is the ‘spleen response’ part of the diving
response?
Schagatay E1 , Richardson M1 , de Bruijn R1 , Andersson J2
1) Department of Natural Sciences, Mid Sweden University, Sundsvall, Sweden.
2) Department of Cell and Organism Biology, Lund University, Lund, Sweden.
Abstract
Adaptive mechanisms, involving both the cardiovascular diving response and
the red blood cell-boosting spleen contraction may serve to facilita te apneic
diving in humans. Both mechanisms have been shown to have apnea-prolonging
effects. The more recently discovered spleen response may or may not be part of
the general cardiovascular diving response, discussed based on published and new
data. Methods used were mainly simulated apneic diving by seria l apneas or
apneas with facia l immersion, spaced by short pauses, with the registration of
cardiorespiratory and haematological parameters. The volunteer participants
were hea lthy untra ined subjects and divers of various levels of tra ining. We
found th a t the hematologica l response was not fortif ied by facia l ch ill ing by
immersion like the cardiovascular diving response, but appears to be triggered by
the apnea stimulus alone, possibly in part by the hypoxic stimulus. Wh i le t he
diving response reaches its full magnitude during each subsequent apnea with full
recovery between apneas, the hematological response developed progressively
across severa l apneas, and is not fully reversed until a t least 10 min after apneas.
Taken together the results suggest th a t the hematologica l response may be
closely associated with , but not neurally linked to the human diving response,
indicating tha t there are two different responses associated with human apneic
diving. The main functions of the two adjustments may a lso be different: Wh i le
the cardiovascular diving response has an oxygen-conserving effect during apnea,
the blood boosting spleen contraction will increase blood gas storage capacity,
increasing dive oxygen storage as well by facilita ting recovery between apneas.
113
Introduction
Cessation of breath ing (apnea), whether voluntary or involuntary, is one of the
most acutely stressful situations for the organism, wh ich can with in short become
a threat to surviva l. Th is ra ises great demands on an effective regulation of
limited supplies of oxygen in lungs, blood and other tissues, and any increase of
the gas-storage potentia l would also be beneficia l. Functions known to serve these
regulatory purposes in diving mammals are the cardiovascular diving response
and the red blood cell-boosting spleen contraction, wh ich may facilita te apnea
and apneic diving also in humans. Previous studies revealed the presence of an
oxygen-conserving diving response also in humans, and more recently a
hematological response from spleen contraction was described.
Diving response
The diving response, first observed in human divers by Irving (12) can be viewed
as a priority system triggered during the threat of apneic asphyxia ( hypoxia ,
hypercapnia and acidosis). The most pronounced physiological adjustments are a
reduction of the heart rate (bradycardia) and selective vasoconstriction. The
function is to guarantee the oxygen supply to the heart and brain (7), wh ich
cannot survive without a constant access to oxygen, while less sensitive organs
rely on stored oxygen (e.g., myoglobin in muscles), anaerobic metabolism, or,
possibly, hypometabolism. Physiologica l mechanisms for reducing metabolism
during apnea are selective vasoconstriction, causing local reduction in oxygen
consumption and the bradycardia in itself, leading to a reduction of the demand
of oxygen of the cardiac muscle (15). Both mechanisms have been considered to
contribute to prolonged apneic durations in diving mammals (7). It h as previously
been shown th a t these factors are at work with a dive-prolonging (19) and
oxygen-conserving effect a lso in humans during apnea (1) and during apnea wit h
exercise (2,3,16).
Spleen response
In diving seals, spleen contraction has been shown to contribute to prolonged
apneas by release of erythrocytes (11,17). Spleen contraction has been observed in
humans both during physica l exercise (14) and diving (10). Hurford observed a
decrease in spleen size after long series of breath- hold dives in the Korean Ama
by measurements with ultrasonic imaging (10). Little is known about its initia tion
and function, but the response appears to develop across a number of apneas. In
both diving mammals and naturally diving humans, dives are performed in series
114
with the duration of apneas and surface interva ls adapted to the working depth .
In elite apneists, 'work-up' apneas are used to increase performance, and may
yield apneic durations exceeding eight minutes. Repeated apneas with short
(<10 min) interva ls h ave been shown to prolong human apneic time (9). The
underlying mechanisms were poorly understood until it was discovered th a t t he
hemoglobin concentration (Hb) and hematocrit (Hct) were reversibly increased
over series of apneas and not accompanied by an increase in plasma proteins,
which suggests th a t spleen contraction had occurred (22). The conclusion was
supported by a lack of th is response in splenectomized subjects (subjects wit h
spleen removed), wh ich a lso lacked the increase in apneic time observed in the
subjects with spleen (22). A close association was observed between changes in
spleen size and Hct and Hb across a series of apneas, which supports the
conclusion tha t spleen contraction is the origin of the hematological effects (18).
Th is indicates th at the spleen response can be held responsible for an increase in
apneic time at repeated apneas in humans (22).
It seems clear th a t both the cardiovascular diving response and the spleen
response have potentia lly dive-prolonging effects in diving mammals as well as
in humans. Repeated diving may elicit the diving response and spleen contraction
a t similar or different time frames, by the same or by different mechanisms. If
mechanisms for initiation and time for establishment of the adjustments are
similar or identica l, they could be considered parts of the same general response,
and we have to accept a wider definition th an the classical of the diving
response. But if initia tion and development are different, they may be two
different responses occurring in para llel but with different physiological origins.
The a im is to further study the mechanisms responsible for initiation of the
responses and the ir development, to revea l whether the spleen-response is part
of the classical diving response.
Methods
A model for 'simulated diving' by apnea or apnea with facia l immersion in cold
water has been developed in our laboratory, a llowing a h igh level of control of
the behaviora l and environmental factors which can affect the associated
responses (20). Hea lthy adult volunteers of both sexes, with categorized diving
experience and physica l-tra ining background, participated. They performed
series of apneas, simulated dives or immersed dives, at rest or during exercise,
usually with short (2 min) recovery pauses. Apneas were typically performed
after a deep but not maximal inspiration, and preceded by normal respiration.
Durations and deta ils in performance depended on the factor being studied. By
manipulating the factor of interest we could elucidate the physiologica l
115
mechanisms responsible for the cardiorespiratory or haematological adjustments.
The method developed is based on continuous registra tion of cardiovascular and
respiratory parameters using mainly non-invasive techniques. These methods are
combined with the use of invasive techniques, ma inly blood sampling and
ultrasonic imaging for measurements of spleen size. As a complement to the
simulated diving model we have in some studies used field registrations from
divers, or experiments in an indoor tank. For deta ils in methods see previous
studies and below.
Results and Discussion
Initia tion and development of the cardiovascular diving response
The main neural inputs eliciting the cardiovascular diving response are 1)
initia tion of apnea, most likely media ted via lung stretch receptors, or possibly
by direct radia tion from the respiratory center to the cardiovascular control
centers (7), and 2) chil ling of the face (13), mediated by cold receptors in the
upper part of the face (24). About ha lf of the response is present at apnea alone,
compared to apnea with face immersion (Figure 1). The influence of hypoxia and
hypercapnia appear to be of minor importance. The diving response is thereby a
preventive system, as the response is triggered prior to the occurrence of
asphyxia. The diving response develops rapidly upon apnea onset, and is fully
developed with in 30 s (Figure 3). The recovery after apnea is a lso nearly
immediate, and the diving response does not change in magnitude across seria l
apneas or dives (21; Figure 3). The response is, however, affected by the intensity
of the thermal stimulus, i.e., the difference between ambient air and water
temperatures (20). It was concluded from th a t study th a t the tropical diver wil l
a lso have a well developed response, as long as there is a temperature difference
between air and water. A more recent study revea led th a t an oxygen conserving
diving response resulted with apneic face immersion a lso in the diver with the
body being constantly immersed in 23°C water, despite cold-induced
vasoconstriction and other circulatory effects of the immersion (6). The diving
response was also found to dominate over the tach ycardia response normally
associated with immersion of the arm, when both arm and face were immersed in
cold water during apnea (4). It is present also during exercise and apnea (2,3).
Th is powerful response thus seems to have the h ighest priority in cardiovascular
regulation, overruling the influence of other strong cardioregulatory stimuli.
116
Figure 1. Mean (±SE) heart rate and skin blood flow reduction during apneas (A)
and apneas with face immersion (FIA) of the same duration in eight subjects.
S ignificance is indicated by *, for p<0.05, and *** for p<0.001.
Figure 2. Mean (±SE) change in hematocrit (Hct) and hemoglobin concentration
(Hb) from eight subjects during apneas (A) and apneas with face immersion (FIA)
of the same duration. No differences were observed between A and FIA.
117
Figure 3. Spleen volume (mL) and heart rate (beats*min -1) va lues from one
subject across three maximal effort apneas (A1-A3) performed in series
interspaced by 2 min pauses. Spleen volume is calcula ted from three diameters,
measured using ultrasonic imaging.
Initia tion and development of spleen contraction and hematological response
Spleen contraction and the associated increases in Hct and Hb appear to develop
slowly compared to the cardiovascular diving response, and severa l apneas are
required for the full response development (22; Figure 3). The hematologica l
response has been suggested to be initia ted by apnea and facia l immersion, i.e., by
the same factors as the cardiovascular diving response (8). However, we found
th a t the response was the same irrespective of face immersion in cold water or not
(Figure 2). It appears th a t apneic hypoxia augments the response, but some of the
Hb response occurs also during normoxic apnea (18), indicating th a t hypercapnia
or the apnea in itself may a lso be involved. There are indications th a t
catecholamines are involved in inducing the exercise-related spleen contraction,
but also th a t these hormones do not account for a ll of the changes seen (25). The
conclusions from different studies concerning recovery have been very different,
like ly depending on both the method used for triggering the response and for
detecting the spleen volume changes. The recovery process is described as
completed with in 2 min after apneas when apneas were spaced by at least 10 min
of rest (8). Th is contrasts to the findings by Bakovic and associates (5) where t he
recovery lasted at least 60 min after seria l apneas. Our studies using five apneas
spaced by 2 min, suggest an intermediate recovery time, of approximately 10 min
118
after the last apnea (23). However, when we used severa l such apnea series
spaced by 10 min, the recovery was still incomplete after 10 min between series
but complete 10 min after the last series (unpublished observations). Further
studies will reveal the physiological origin of spleen contraction, but it seems
clear th a t the mechanisms are different from th ose initia ting the diving
response.
Conclusions
These results suggest th a t the hematological spleen response is not neurally
linked with the human diving response, nor is it developing in the same time
span. The different initia tion and development indica te th a t the diving response
and the spleen response should be viewed as two different responses associated
with human apneic diving, each with specif ic effects on apneic performance.
References
1. Andersson J, Schagatay E. Arteria l oxygen desa turation during apnea in
humans. Undersea Hyperb Med 1998; 25: 21-25.
2. Andersson, JPA., Linér MH, Rünow E, Schagatay E. Diving response and
arteria l oxygen saturation during apnea and exercise in breath- hold divers. J
Appl Physiol 2002; 93: 882-886.
3. Andersson J, Linér M, Fredsted A, and Schagatay E. Cardiovascular and
respiratory responses to apneas with and without face immersion in exercising
humans. J Appl Physiol 2004; 96: 1005- 1010.
4. Andersson J, Schagatay E, Gislén A, Holm B. Cardiovascular responses to cold
water immersions of the forearm and face, and the ir rela tionship to apnoea. Eur J
Appl Physiol 2000; 83: 566-572.
5. Bakovic D, Va lic Z, Eterovic D, et al. Spleen volume and blood flow response
to repeated breath- hold apneas. J Appl Physiol 2003; 95: 1460-1466.
6. DeBruijn R, Richardson M, Schagatay E. The function of the human diving
response in the immersed diver. Abstract. Eur Underwater and Baromed Soc
(EUBS) meeting Copenhagen, Denmark August 2003: 27-31.
7. Elsner R, Gooden B. Diving and asphyxia: A comparative study of animals and
man. Physiologica l Society Monograph 40. Cambridge Univ Press, Cambridge,
1983; 175 pp.
8. Espersen K, Frandsen H, Lorentzen T, et a l. The human spleen as an
erythrocyte reservoir in diving related interventions. J Appl Physiol 2002; 92:
2071-2079.
9. Heath JR, Irvin CJ. An increase in breath- hold time appearing after brea t h
holding. Resp Physiol 1968; 4:73-77.
119
10. Hurford WE, Hong SK, Park YS, et a l. Splenic contraction during breath- hold
diving in the Korean ama. J Appl Physiol 1990; 69: 932-936.
11. Hurford WE, Hochachka PW, Schneider RC, et al. Splenic contraction,
catecholamine release, and blood volume redistribution during diving in the
Weddell seal. J Appl Physiol 1996; 80:298-306.
12. Irving L. Bradycardia in human divers. J Appl Physiol 1963; 18: 489-491.
120
Errata
A number of incorrect in-text cita tions were discovered after the thesis was
published, in the appendices for articles IV through VII. It appears th a t
EndNote was a bit overzealous and either replaced or added severa l incorrect
cita tions, in [block parentheses] only, without th is coming to my attention. I
apologize for th is inconvenience. Please note th at apart from these article
appendices, all cita tions and references in the thesis are to my knowledge correct.
The correct cita tions for Papers IV through VII are listed below, in the order in
which they appear in the respective appendices.
Paper IV was published online on October 8, 2008 and can be found in its entirety
a t the following link: http://dx.doi.org/10.1007/s00421-008-0873-9.
S incerely
Matt Rich ardson
PAPER IV: Hypoxia augments
concentration and hematocrit
apnea-induced
increase
in
hemoglobin
[82]: Hurford et a l 1996; [155-157]: Bakovic et a l 2005, Hurford et a l 1996,
Schagatay et a l 2001; [77]: Espersen et a l 2002; [50, 76]: Bakovic et a l 2005,
Schagatay et al 2001; [78]: Richardson et a l 2003; [158]: Schagatay et a l 2005; [35]:
Bakovic et al 2003; [81, 159]: Donald and Aarhus 1974, Greenway and Stark 1968;
[68]: Hoka et a l 1989; [160]: Qvist et a l 1993; [12, 161]: Engel et a l 1946, Klocke and
Ra hn 1959; [162]: Corkey et al 1998; [163]: Hopkins 2000; [50]: Schagatay et a l
2001, 2005; [164]: Lin et al 1983.
PAPER V: Hypercapnia moderates hemoglobin increases during apnea.
[165]: Andersson et a l 2002; [48]: Hurford et al 1990; [50]: Schagatay et a l 2001;
[77]: Espersen et a l 2002; [50, 76]: Bakovic et a l 2005, Sch agatay et a l 2001; [48, 50]:
Hurford et al 1990, Schagatay et al 2001; [50, 79]: Schagatay et al 2001, 2005;
[166]: Rich ardson et a l 2006; [50, 167]: Richardson et al 2007; Schagatay et a l
2001; [168]: Schagatay et a l 2007; [165]: Andersson et al 2002; [169]: Richardson et
a l 2005; [170]: Lin et a l 1974; [13]: Schagatay and Andersson 1999; [171-173]:
Meadows et a l 2004, Poulin et al 1996, Vovk et a l 2002; [174, 175]: Lloyd et a l 1958,
Mohan and Duffin 1997; [6]: Schagatay 1996; [176]: Jung and Stolle 1981; [163];
Hopkins 2000; [177]: Ahmad and Loeschcke 1982; [85, 88, 178]; Ayers et al 1972,
Davies et a l 1978, Herman et a l 1982; [179]: Kara et a l 2003; [180]: Biesold et a l
1989; [181]: Fukuda et a l 1989; [182]: Hoka et al 1992.
PAPER VI: Spleen and lung volumes correla te with performance in elite apnea
divers
[48]: Hurford et a l 1990; [50]: Schagatay et a l 2001 [55]: Qvist et a l 1986; [183]:
Barcroft and Stevens 1927; [50]: Schagatay et a l 2001; [36, 49]: Laub et a l 1993,
Stewart and McKenzie 2002; [35]: Bakovic et a l 2003; [50, 184]: Schagatay et a l
2001, Prommer et a l 2007; [185]: Prassopoulos et al 1997; [79]: Schagatay et a l 2005;
[186]: Ornhagen et al 1998; [187]: Andersson et a l 1998; [188]: Andersson and
Schagatay 1998b ;[189]: Andersson and Schagatay 1998a; [190]: Carey et a l 1956;
[191]: Butler 2001; [79]: Schagatay et al 2005; [36, 49]: Laub et a l 1993, Stewart and
McKenzie 2002; [192]: Byrne et a l 2005; [185]: Prassopoulos et a l 1997; [193]:
Geraghty et a l 2004; [79]: Schagatay et al 2005; [35]: Bakovic et a l 2003; [102]:
Rich ardson et al 2005; [48]: Hurford et a l 1990; [35]: Bakovic et a l 2003; [184]:
Prommer et al 2007; [194]: van Rensburg et a l 1991; [195]: Hong et a l 1963; [196]:
Kotton and Fine 2008; [49, 197]: Chiba et a l 1984; Laub et al 1993; [198]: Andersson
et al 2002; [199]: Andersson et al 2004; [200]: Ferretti and Costa 2003; [35]: Bakovic
et a l 2003; [190]: Carey et a l 1956; [201]: Gaultier and Crapo 1997; [196]: Kotton
and Fine 2008; [202]: Nakajima et a l 1998; [203]: Sch agatay et al 2000; [204]:
Strœmme and Ingjer 1978; [203]: Schagatay et al 2000; [35, 50]: Bakovic et a l 2003,
Schagatay et a l 2001; [35, 48, 79, 205]: Bakovic et al 2003, Hurford et a l 1990,
Rich ardson et a l 2006, Schagatay et a l 2005; [50, 79]: Schagatay et al 2001, 2005;
[192]: Byrne et a l 2005; [206]: Andersson et al 2008; [191]; Butler 2001; [195]: Hong et
a l 1963; [186]: Ornhagen et a l 1998; [207]: Overgaard et a l 2006; [208]: Schaefer et
a l 1968; [189, 209]: Andersson and Schagatay 1998a; [102]: Rich ardson et a l 2005;
[147]: de Bruijn et al 2008; [210]: Ferrigno et al 1987; [211]: Ridgway and
McFarland 2006.
PAPER VII: Cardiovascular and haematological adjustments to apneic diving in
humans – is the ‘spleen response’ part of the diving response?
The in-text cita tions in th is article are correct. A page from the list of references
was left out, however, making the reference list incomplete. The missing
references are listed below:
13. Kawakami Y, Na telson BH, DuBois AB. Cardiovascular effects of face
immersion and factors affecting the diving reflex in man. J Appl Physiol 1967; 23:
964-970.
14. Laub M, Hvid-Jacobsen K, Hovind P, Kanstrup IL, Christensen NJ, and
Nie lsen SL. Spleen emptying and venous hematocrit in humans during exercise. J
Appl Physiol 1993; 74: 1024-1026.
15. Lin YC. Breath- hold diving in terrestria l mamma ls. Exercise and Sports Sci
Rev 1982; 10, 270-307.
16. Lindholm P and Linnarsson D. Pulmonary gas exchange during apnoea in
exercising men. Eur J Appl Physiol 2002; 86: 487-491.
17. Qvist J, Hill RD, Schneider RC, Falke KJ, Liggins GC, Guppy M, Elliot RL,
Hochachka PW, Zapol WM. Hemoglobin concentrations and blood gas tensions of
free-diving Weddell seals. J Appl Physiol 1986; 61: 1560-1569.
18. Rich ardsson M, deBruijn R, Schagatay E. Hypoxia – a trigger for spleen
contraction? Abstract at the 31st annual meeting of EUBS and XVmeeting of
ICHM, Barcelona, Spa in. September 2005: 7-10.
19. Schagatay E, Andersson J. Diving response and apneic time in humans.
Undersea & Hyperbaric Med 1998; 25: 13-19.
20. Schagatay E, Holm B. The effects of water and ambient a ir temperatures on
human diving bradycardia. Eur J Appl Occup Physiol 1996; 73: 1-6.
21. Schagatay E, van Kampen M, and Andersson J. Effects of repeated apneas on
apneic time and diving response in non-divers. Undersea Hyperb Med 1999; 26:
143-149.
22. Schagatay E, Andersson J, Ha llén M, På lsson B. Selected contribution: role of
spleen emptying in prolonging apneas in humans. J Appl Physiol 2001; 90: 16231629.
23. Schagatay E, Haughey H, Reimers J. Speed of spleen volume changes evoked
by seria l apneas. Eur J Appl Physiol 2005; 93: 447-452.
24. Schuitema E, Holm B. The role of different facia l areas in eliciting human
diving bradycardia. Acta Physiol Scand 1988; 132: 119-120.
25. Stewart IB, Warburton DE, Hodges AN, Lyster DM, McKenzie DC.
Cardiovascular and splenic responses to exercise in humans. J Appl Physiol 2003;
94: 1619-1626.