255
Clinical Science ( 1998) 94,255-26 I (Printed in Great Britain)
Skin microcirculation in patients with Type I diabetes with and
without neuropathy after neurovascular stimulation
Thomas FORST, Andreas PFUTZNER, Thomas KUNT, Thomas POHLMANN, Ulrike SCHENK,
Rupert BAUERSACHS, Ernst KUSTNER and JiirgenBEYER
University Hospital Mainz, Department of Internal Medicine and Endocrinology, langenbeckstr. I ,
D-55 I 0 I Mainz, Germany
(Received 25 April123 October 1997; accepted 28 Oaober 1997)
1. Neurovascular inflammation is impaired in
patients suffering from diabetic neuropathy. The
aim of our study was to evaluate the distribution of
nutritive and total skin blood flow in diabetic
patients with and without neuropathy after neurovascular stimulation with acetylcholine.
2. ltventy patients with Q p e I diabetes, 10 with and
10 without neuropathy, and 10 age-matched nondiabetic control subjects, underwent microvascular
investigations before and after neurovascular stimulation by intracutaneous application of acetylcholine. The capillary blood cell velocity in the
nailfold of the hallux was measured by videophotometric capillaroscopy, and the total skin microcirculation in the same area by laser Doppler flowmetry.
3. The increase in total skin blood flow was significantly impaired in the group of neuropathic diabetic
patients compared with the non-neuropathic diabetic patients (17.5 f8 3 versus 51.0 f 16.2; P <0.05)
and the non-diabetic subjects (17.5 8.3 versus
67.8 f19.7; P <0.01). The increase in capillary
blood flow was not significantly impaired in Q p e I
diabetes patients with neuropathy.
4. The ratio between capillary blood flow and total
skin perfusion decreased significantly in the control
group (from 0.82 0.15 to 0.47 f0.11; P <0.005) and
in the Q p e I diabetes patients without neuropathy
(from 0.79 f0.12 to 0.43 f0.12; P <0.05), whereas
the decrease in the neuropathic group was statistically insignificant (from 1.05 f 0.19 to 0.72 fO.16).
5. Diminished total skin perfusion in the foot after
intracutaneous stimulation with acetylcholine in
Q p e I diabetes patients is associated with diabetic
neuropathy, indicating a disturbance in the neurovascular reflex arc. This impaired neurovascular
response is caused by a diminished total and subpapillary blood flow and not by a diminished nutritive capillary flow. There is no evidence of a
diminished nutritive capillary blood flow during
neurogenic inflammation in Q p e I diabetes patients
suffering from diabetic neuropathy.
INTRODUCTION
It is widely accepted that macroangiopathy, peripheral neuropathy, infection, haemorheological and
metabolical disturbances are important factors in
the pathogenesis of tissue breakdown and foot ulceration in diabetic patients. The concept of microvascular dysfunction, however, is still subject to
contention. This concept recognizes the fact that
numerous secondary, structural and functional disturbances in the microcirculation are present in diabetic patients. Peripheral diabetic neuropathy is
associated with microvascular overperfusion under
resting conditions and with an impaired hyperaemic
response in total skin blood flow [l-51. Excess
arteriovenous shuntflow due to a reduced peripheral
sympathetic tone has been reported [6-81. This pattern of blood flow seems to be controlled by local
reflexes whose function may be altered in diabetic
neuropathy [l, 91.
Neurogenic inflammation has been shown to be
reduced in diabetic patients suffering from peripheral neuropathy [10-131. Neurovascular malperfusion may be implicated in the development of
trophic foot ulceration [121. Intradermally applied
acetylcholine stimulates neurogenic inflammation by
an antidromic axon reflex and causes a spreading
vasodilatation and a flare response [lo, 12, 131.
Acetylcholine also elicits microvascular dilatation by
the release of the vasodilators nitric oxide, prostacyclin and the endothelium-derived hyperpolarizing
factor from endothelial cells [14-161.
Post-occlusive reactive hyperaemia has been
shown to be impaired in patients suffering from
peripheral vascular disease with and without diabetes mellitus [17, 181, which may be caused by a
low arterial pressure [19]. Otherwise, in diabetic
patients without macrovascular disease, total skin
blood flow during post-occlusive reactive hyperaemia seems to be unchanged [20-221. A maldistribution has been shown between nutritional capillary
and non-nutritional arteriovenous shuntflow, inde-
Key words: diabetic neuropathy, Type I diabetes, neurogenic inflammation, nutritive capillary blood Ilow. total skin blood flow.
Abbreviations: CBV, capillary blood cell velocity; LDF, laser Doppler flux.
Correspondence: Dr Thomas Fom.
256
T. Font et al.
pendent of diabetes mellitus, in patients with peripheral vascular disease [ 18, 221. Contradictory
results for the distribution of nutritional capillary
and subpapillary microvascular blood flow have been
published for diabetic patients without peripheral
vascular disease [23, 241.
Accordingly, the purpose of the present study was
to investigate total skin microvascular blood flow
and nutritional capillary blood flow in diabetic
patients with and without evidence of neuropathy
after neurovascular simulation with intracutaneously
applied acetylcholine.
METHODS
Subjects
Two groups of patients with Type I diabetes, one
with and one without features of diabetic neuropathy, were investigated and compared with a nondiabetic control group. Full clinical details of the
subjects investigated are given in Table 1. All
patients were clinically free from peripheral macrovascular disease, as evaluated by palpable foot
pulses and segmental blood pressure measurements.
Subjects were excluded from the study if they suffered from neuropathy of an origin other than diabetes mellitus, or were on vasoactive drugs or
medications known to influence the autonomic nervous system. No patient exhibited a blood glucose
level below 4 mmol/l at the start of the study, or suffered from hypoglycaemia before or during the
investigation. All subjects refrained from smoking
2 h before the study. The study has been carried out
in accordance with the Declaration of Helsinki
(1989) of the World Medical Association and all
subjects gave their informed consent.
Study conditions and investigation procedure
All measurements were performed in a room with
an ambient temperature of 21-24°C. All subjects
were allowed to lie down and to acclimatize for a
period of about 15 min. The skin microcirculation in
the nailfold of the hallux was investigated by laser
Doppler flowmetry [laser Doppler flux (LDF)] and
videophotometric capillaroscopy [capillary blood cell
velocity (CBV)] with the knee in a slightly flexed
position. After a steady signal had been obtained,
resting LDF and resting CBV were recorded for a
period of 3 min. Thereafter, 0.05 ml of acetylcholine
(acetylcholine chloride 1%; Ciba Vision, Wessling,
Germany) was injected intradermally just behind the
toe nailfold and again the LDF and CBV were
recorded at about 2-3cm away from the injection
site. The mean LDF and CBV before and 5min
after stimulation with acetylcholine were calculated.
Skin temperature at the dorsum of the hallux was
measured before LDF and CBV measurements
using an electronic thermometer (Digimed H l l S ,
Medizin-Messtechnik, Waldkirch, Germany).
Assessment of neuropathy
The autonomic nerve function was evaluated by
measurement of the heart rate variation with an
automated computer-based system (Pro Sci Card 11,
Linden, Germany) which is described in detail elsewhere [25]. Heart rate variation at rest, spectral
analysis of the low-, mid- and high-frequency range,
heart rate variation during deep breathing, the Valsalva manoeuvre and the lying-to-standing maximum/minimum ratio were measured and compared
with age-related normal ranges [25]. Blood pressure
response was tested in the supine position and 1 min
after standing. A decrease in systolic pressure of
more than 30mmHg was considered to be pathological. A score of three or more pathological tests
was defined as indicative of cardiovascular autonomic neuropathy.
Peripheral somatic neuropathy was assessed by
examining the following aspects: the ankle reflexes
using a standard tendon reflex hammer, the vibration perception threshold of the great toe by biothesiometry (Vibra Tester 100, PHYWE, Gottingen,
Germany) [26], and the thermal perception and heat
pain perception thresholds at the dorsum of the foot
using a marstock stimulator (Path-Tester, PHYWE).
Tabk I . Clinical chuacteristier of the investigated groups. Valws are means kS.E.M. Statistical significance, *P<O.OS.
Type I diabetes patients
with neuropathy (n = lo)
Type I diabetes patients
withoot neuropathy (n = 10)
36.5 f5. I
614
6
8.3 f3.9
14.6 f2.7
9.2 k0.I.
0.97 f0.07
305.3 f 16.8
2.5 f I.8
218
40.3 k 4 . 6
515
4
6.8 f2. I
12.3 f2.4
7.3 0.04
1.01 k0.13
3 10.5 f26.9
0.5 I k0.34
416
Non-diabetic control
group (n = lo)
~
Age (r-1
Sex (MIF)
Smoker
Mean blood glucose at time of inwstigation (mmolll)
Duration of diabetes (years)
Hawnoglobin Alc (%)
Plasma creatinine (mgldl)
Plasma fibrinogen (mg/dl)
Albuminuria (mg/24 h)
Retinopathy (YeslNo )
~
38.8k 2.0
416
5
*
0.95 f0.04
329.4 f26.6
0/10
Skin blood tbw in diabetic neuropathy
Age-adjusted values in an individual patient greater
than the 95th percentile, compared with values from
a healthy population, were regarded as abnormal
P61.
Diabetic patients were classified as suffering from
diabetic neuropathy if the total score of the abovedescribed neurological investigations (including
cardiovascular function) exhibited three or more
pathological test results.
Assessment of retinopathy
Diabetic retinopathy was assessed using a nonmydriatic fundus camera (Canon CR 4-45-NM,
Tokyo, Japan) together with an examination by an
ophthalmologist. Retinopathy was graded as present
or absent, regardless of the severity of retinal
changes.
Blood flow measurements
Laser Doppler flowmetry. The technique of laser
Doppler flowmetry (MBF 3D, Moor Instruments,
Devon, UK) was used to measure total skin microcirculation at the toe nailfold [27, 281. Laser Doppler flowmetry is a continuous and non-invasive
technique of cutaneous blood flow measurement
that has been validated using synchronous dynamic
capillaroscopy [29]. The capillary component of
superficial skin blood flow is physiologically minimal
as compared with the subpapillary blood flow.
Therefore, the LDF signal is largely indicative of
thermoregulatory and subpapillary blood flow
[30-321. Resting LDF (rLDF) and peak LDF
(pLDF) were each measured over a period of 3 min.
The absolute increase in LDF (ALDF) from baseline after axon reflex stimulation was calculated.
Vieophotometnc capillaroscopy. Using capillary
microscopy it is possible to study the nutritional
capillaries in the skin directly. This is a technique
for measuring blood flow that allows a direct estimate of nutritive capillary blood flow alone [31, 321.
Nailfold capillaries of the hallux were displayed on a
TV-monitor using a television microscope (Capi
Scope, Moritex Europe Ltd, Cambridge, U.K.) at a
magnification of 600. The image was stored on videotape for subsequent analysis. One-minute flow
was analysed in each of four previous recorded
capillaries with a good contrast, clear focus and
visible signals for at least 1 min. Mean CBV was calculated from 10 single measurements out of each
capillary. CBV was determined using the ‘line shift
diagram method’, a computerized videophotometric
technique (Cap Image, Dr. Zeintl Software
Engineering, Heidelberg, Germany). Using this
method a scale is drawn adjacent to the blood vessel
and the movement of plasma gaps are displayed in
the form of vertical lines in a line shift diagram. The
gradient of the lines in the diagram is determined
257
and the velocity is calculated. The advantage of this
method is that zero velocities and negative velocities
can also be measured and calculated. Resting capillary blood cell velocity (rCBV) and peak capillary
blood cell velocity (pCBV) after stimulation with
acetylcholine were measured. The absolute increase
in CBV (ACBV) from baseline after axon reflex
stimulation was calculated.
In addition, the perfused capillaries (capillary
density) were projected on to a defined area and
calculated as number of capillaries per area. The
erythrocyte column width was measured at the
arterial side of the capillary loop.
Blood tests
Venous blood samples were taken for the determination of HbAlc, creatinine and plasma fibrinogen. HbAlc was only measured in diabetic patients.
Urinary albumin was measured in diabetic patients
by 24 h urine sampling, using an immuno-diffusion
technique.
Statistical analysis
Results are expressed in terms of the arithmetic
mean f S.E.M. The results obtained from measurements of skin blood flow were not normally distributed. Non-parametric methods of statistical
analysis have been used. The Mann-Whitney U-test
and the Wilcoxon rank sum test were used in the
statistical analysis. Multiple regression analysis and
the product moment correlation coefficient were
used to compare LDF and CBV with skin temperature, age, duration of diabetes, HbAlc and albuminuria.
RESULTS
As shown in Table 1, all groups were comparable
with regard to age, sex, creatinine, plasma fibrinogen
blood glucose levels or smoking habits. HbA1, was
significantly increased in the group of diabetic
patients with neuropathy compared with the group
without neuropathy. No statistically significant difference was found with regard to duration of diabetes, presence of retinopathy and albuminuria
between the diabetic subgroups.
Blood flow measurements
The reproducibility of the laser Doppler flowmetry readings was tested by performing duplicate
measurements of neurovascular stimulation on two
different days in eight patients. The intraindividual
coefficient of variation was 17.5%. LDF was also
measured after intradermally injecting 0.05 ml of
T. Fom et al.
150
0.9% NaCl solution in five control subjects. The
mean LDF signal decreased from 97.7 f67.4 to
72.5 f40.1 after intracutaneous NaCl injection (not
significant).
The reproducibility of the CBV measurements
was also tested on eight patients. An intraindividual
coefficient of variation of 11.0 % was found. Again
the test procedure was performed using 0.05ml of
NaCl solution instead of acetylcholine on five control subjects. The CBV increased from 165 f12 pm/s
to 170 f13 pmls after intracutaneous NaCl injection
(not significant).
The skin temperatures on the dorsum of the hallux were comparable between the investigated
groups (Table 2).
Total skin microcirculation
No significant differences were observed in rLDF
at baseline between the investigated groups. After
neurovascular stimulation there was a strong
increase in LDF in the non-diabetic control group
(P<0.005) and in the non-neuropathic diabetic
group (P<0.005), whereas only a slight, albeit significant, increase was observed in the neuropathic diabetic group (P< 0.05). The absolute increase from
baseline LDF (ALDF) was significantly diminished
in the neuropathic group compared with the nonneuropathic diabetic group or the non-diabetic control group (Table 2). No relationship was found
between total skin microcirculation after axon reflex
stimulation and age (r = 0.113; P = 0.27), duration
of diabetes (r = 0.068; P = 0.37), HbA1, (r = 0.084;
P = 0.36), plasma creatinine (r = 0.28; P = 0.07),
plasma fibrinogen (r = 0.125; P = 0.26) or albuminuria (r = 0.052; P = 0.42). No significant difference
was found in LDF between male and female subjects or between patients with or without retinopathy.
Capillary blood flow
No significant difference was found between the
investigated groups with respect to capillary density
or basal erythrocyte column width. After neurovascular stimulation, erythrocyte column width was significantly higher in the group of diabetic patients
without neuropathy compared with the control
group (Table 2). After neurovascular stimulation,
capillary blood flow increased significantly in the
control group and in the group of diabetic patients
without neuropathy. In the group of diabetic
patients with neuropathy, however, the increase in
CBV was not found to be statistically significant. No
significant difference between the investigated
groups could be observed regarding rCBV, pCBV or
ACBV (Table 2). A slight association was found
between plasma fibrinogen levels and capillary blood
flow in response to axon reflex stimulation
(r = -0.34, P = 0.03). No relationship between
capillary blood flow and age (r = 0.14; P = 0.23),
duration of diabetes (r = 0.24; P = O.ll), HbAlc
(r = 0.26; P = 0.13), plasma creatinine (r = 0.07;
P = 0.36) or albuminuria (r = 0.34; P = 0.09) could
be observed. The capillary blood flow response to
acetylcholine (ACBV) was more pronounced in
female than in male subjects (76 f17 pm/s versus
38+_12 pm/s; P<0.05). The response (ACBV) was
significantly higher in Type I diabetes patients suffering from retinopathy (n = 6) than in patients
without retinopathy (91 f27 pm/s versus 44 10
pm/s; P < 0.05).
The ratio between rCBV and rLDF, representing
the proportion of nutritive capillary blood flow to
total skin microcirculation [20], was not significantly
different between the investigated groups when
comparing the basal readings. A significant decrease
in the CBVLDF ratio was observed after neurovascular stimulation with acetylcholine in the control
group and in the diabetic group without neuropathy
(P<0.005 and P <0.05 respectively). Only an insig-
Table Z Mircuhtory muwFements in the investigatedgroups. Valws are means +S.E.M. Statistical signifme *P <0.05
compared with diabetic patients without neuropathy: tP<0.05, ttP<O.OI compared with control group.
Toe skin tempetature (“C)
Resting LDF
Peak LDF
ALDF
Capillary density
(capillarieskm’)
Resting erythrocyte
cdumn width (am)
Peak erythrocyte
column width (pm)
Resting CBV (pmls)
Peak CBV (pmls)
ACBV hmls)
Type I diabetes patients
with neurcpathy (n = 10)
Type I diabetes patients
without neuropathy (n = lo)
Non-diabetic coned
group (n = 10)
27.0f 1.1
38.5k 18.1
55.5 f I7.2t
17.5 f8.3*.tt
49.9 3.0
28.7 f0.9
56.8 k 26.2
124.5 f43.8
67.8 19.7
53.3 f3.8
27.9 +0.9
37.4 f8.0
88.4 f 19.4
51.Of 16.2
52.6 3.8
9.4+ 1.1
11.2*1.1
8.6 & I.O
9.7+ 1.1
13.5 I .ot
+
9.2 k I.O
201 f 13
240 f28
43+ I8
241 13
301 f 2 0
61 18
227+ IS
283 25
56f17
+
+
Skin blood tlow in diabetic neuropathy
Diabetic patients
withneumpathy
contml
Diabtkpatkmts
withoutneuropethy
subiects
Fe. I . Ratio between CBV and LDF in drbaic patients with and
without neuropathy and the nondrbaic control group Mom and
after axon reRex stimulation. m. Basal CBV/LDF ratio;
CBV/LDF ratio. Values are meanrfS.E.M.
0,
stimulated
nificant decrease was found in the diabetic group
with neuropathy (Fig. 1).
Conclusions
Diabetes mellitus may lead to a number of neurovascular complications which may be involved in
the development of foot ulceration and loss of digits. In recent years it has been postulated that autonomic C-fibre neuropathy and sympathetic denervation of skin vessels may lead to an increased
arteriovenous shunt flow and a maldistribution of
nutrient capillary blood flow [4, 6, 24, 331. The neurogenic inflammatory response to injury can also be
impaired in diabetes mellitus, which may be an
important factor contributing to tissue ulceration
and diabetic neuroarthropathy [ l l , 121. Neurogenic
inflammation is mediated by small nociceptor
C-fibres that also subserve pain sensation. An antidromic axon reflex causes a spreading vasodilatation, probably through the action of substance P on
microvessels and mast cells [34]. Non-neurogenic
total skin blood flow responses appear to be little
affected by diabetes in the absence of major-vessel
disease because a substantial increase in cutaneous
blood flow above resting levels was observed in the
following cases: during post-ischaemic hyperaemia
[20, 21, 24, 281, vasodilatation during local heating
[l], the first part of the Lewis triple response [ l , 3,
101 or direct vasodilatation after stimulation with
methacholine or sodium nitroprusside [21, 35, 361.
In our investigation intracutaneous injection of
acetylcholine was used to stimulate cutaneous blood
flow. Acetylcholine elicits vasodilatation through distinct mechanisms; firstly by direct stimulation of the
endothelium and the release of vasodilatory substances [13-161; secondly by the activation of an
axon reflex mechanism and the initiation of neurogenic inflammation [lo, 12, 131. In our investigation
we have shown that the total vascular response to
259
intradermally applied acetylcholine is significantly
impaired in Type I diabetes patients suffering from
neuropathic complications but was not affected in
patients suffering from microvascular complications
such as retinopathy or albuminuria. This finding suggests that the attenuated vascular response in Type I
diabetes patients with neuropathy after intracutaneous stimulation with acetylcholine is mainly due to a
loss of nociceptive C-fibre function and is less influenced by a disturbed endothelial function. It is also
unlikely that the impaired vascular response may be
the result of a disturbed integrity of vascular wall
since the response was unaffected in diabetic
patients without neuropathy. This is in accordance
with an unaffected hyperaemic response after
ischaemia in diabetic patients without macrovascular
disease [20-221 and with an intact endotheliumdependent vasodilatation after stimulation with
methacholine in Type I diabetes patients without
complications [21]. It could also be conceived that
acetylcholine has an endothelium-dependent vasodilatory effect when applied directly into the bloodstream or to the endothelial cells, but mainly acts
via a nociceptive axon reflex arc when applied intracutaneously.
In our study the neurovascular vasodilator
response could have been induced via two separate
pathways. One is via the activation of a neurogenic
axon reflex by acetylcholine and the other is through
the stimulation of nociceptor fibres by skin trauma
[ l , 31. The second pathway seems to have had less
influence on our measurements, since intracutaneous injection of NaCl did not significantly alter skin
blood flow in our investigation. Acetylcholine stimulates the afferent branch of the axon reflex arc and
the flare response can be eliminated by using local
anaesthetics and is also absent in denervated skin
[lo, 131. It has been shown that acetylcholine has a
direct excitatory effect on non-myelinated C-fibres
in the saphenous nerve of the cat, which can be
eliminated by using hexamethonium but not by atropine [37]. Pilocarpine produces direct vasodilatation
identical to that produced by acetylcholine but no
flare response [101. Therefore, the microvascular
response of intradermally applied acetylcholine
appears to be mediated via the nicotinic action of
acetylcholine. In addition, intradermally applied
acetylcholine may act via the prejunctional inhibition of the sympathetic neurotransmission by muscarine receptors [38].
Our observations confirm previous investigations
showing a diminished increase in total skin blood
flow as measured by laser Doppler flowmetry after
axon reflex vasodilatation in diabetic patients suffering from diabetic neuropathy [ l l , 12, 35, 361. The
reduced hyperaemic response in total skin microcirculation to axon reflex stimulation in patients suffering from diabetic neuropathy underlines the need
for an intact nerve supply for regulating the thermoregulatory arteriovenous shuntflow. An increased
capillary blood pressure, which has been demon-
260
T. Forst e-tal.
strated in toe and finger nailfolds of diabetic
patients [14, 39, 401, may be a consequence of the
opening of arteriovenous shunt vessels. The interaction of sympathetic neuropathy and microvascular
blood flow may lead to an increased transmural
pressure, thus promoting the formation of neuropathic oedema [33, 411, and may accelerate longterm capillary structural damage [40, 421 and
increase vascular leakage [43j.
On the other hand, it has been postulated that the
sympathetic denervation of precapillary vessels may
lead to a maldistribution of capillary blood flow with
a striking reduction of nutritive capillary flow. The
impaired nutritive capillary blood flow may lead to
an impaired exchange of nutrients, oxygen and waste
products in the skin. Whereas an increased skin
capillary and arteriovenous shunt flow has been
observed in diabetic patients under resting conditions [23], a recent investigation demonstrated a significantly impaired nutritional capillary flow in Type
I diabetes patients during post-occlusive hyperaemia, indicating a capillary steal phenomenon under
ischaemic conditions [24].
In our investigation, videophotometric capillaroscopy revealed patent capillaries and a similar distribution of terminal row capillaries in each of the
investigated groups. Our data contradict the hypothesis of advanced morphological microangiopathy
in patients suffering from diabetes mellitus. In a
recent study, Morris et al. [14] reported a significant
increase in finger capillary blood flow after the iontophoresis of acetylcholine, which was accompanied
by an increase in capillary pressure in healthy subjects. In accordance with this observation, our data
confirm an increase in foot capillary blood flow after
intracutaneous injection of acetylcholine. Our study
was able to demonstrate in addition that the
increase in capillary blood flow is not affected in
diabetic patients without neuropathy and is only
slightly influenced by diabetic neuropathy.
The ratio between CBV and LDF, representing
the distribution of blood flow between total skin
microcirculation and nutritive capillary flow, was not
significantly different between the investigated
groups under basal conditions. After neurovascular
stimulation with acetylcholine, a significant decrease
in the ratio between nutritive capillary blood flow
and total skin blood flow was found in control subjects and diabetic patients without neuropathy but
no significant change could be observed in neuropathic diabetic patients. These findings suggest a
deterioration in the regulation of precapillary resistance in patients with autonomic C-fibre neuropathy
and a maldistribution of skin perfusion with a reduction of total skin blood flow during neurogenic
inflammation. Whether this neurovascular malfunction has any effect on the development of trophic
tissue ulceration needs further investigation. In contrast to the finding of a decreased nutritive capillary
blood flow after post-occlusive ischaemia in diabetic
patients [24], our findings do not support a
diminished capillary blood flow after neurovascular
stimulation with acetylcholine in Type I diabetes
patients with or without neuropathy.
In conclusion the present findings support an
impaired neurogenic inflammatory response to
injury in total skin perfusion in Type I diabetes
patients suffering from diabetic neuropathy. In addition the present study demonstrates a maldistribution between total skin blood flow and nutritive
capillary blood flow, which is conditioned by a
diminished increase in non-nutritional blood flow
and is not accompanied by a diminished capillary
blood flow in Type I diabetes patients with neuropathy during neurogenic inflammation. There is no
evidence for a capillary steal phenomenon during
neurovascular stimulation in Type I diabetes
patients with or without neuropathy.
ACKNOWLEDGMENTS
We thank the nurses and staff of the Department
of Endocrinology of the University Hospital Mainz
for their assistance. This paper includes data from
the theses of Mrs Ulrike Schenk and Mr Thomas
Pohlmann. This study was supported by the
‘German Diabetes Association’ and the ‘Burger
Busing Foundation’.
REFERENCES
I. Rayman G, Williams SA Spencer PD, Smaje LH, Wise PH, Tooke JE. Impaired
microvascularresponse to minor skin trauma in type I diabetes. Br Med J 1986;
292: 1235-8.
2. Rayman G, Haaan A, Tooke JE. Blood flow in the skin of the foot related to
posture in diabetes mellitus. Br J Med 1986; 292: 87-90.
3. Walmsley D, Wales JK, Wiles PG. Reduced hypetaemiafdlowing skin trauma:
evidence for an impaired microvascularresponse to injury in the diabetic foot.
Diabetologia I989 3 2 736-9.
4. Stevens MJ, Edmonds ME, Foster A W , Douglas SLE. Watkins PJ. Paradoxical
blood flow responses in the diabetic neuropathic foot: an assessment of the
contribution of vascular denervation and microandopathy. Diabetic Med I992 9:
49-54.
5. Shore AC, Price 4,Sandema DD, Tripp JH. Tooke JE. Postural~induced
Msoconmiction in diabetes mellis. AKh Dis Child 1994,70 22-6.
6. Boultm AIM. Scarpello JHB, Ward JD. Venous orlgenation in the diabetic
1982 2 2
neurcpathic foot: evidence of arteriovenous shunting. Diabet-a
6-8.
7. Corbin DOC, Young Rj, Morrison DC, et al. Blood flow in the foot.
polyneuropathyand foot ulcetation in diabetes mellitus. Diabetologia I987 3 0
48-73,
8. Uccidi L, Mancini L, Gordano A, Solini A, Magnani P, Manto A Lower limb,
arteriovencus shunts, autrmomic neuropathy and diabetic foot. Diabetes Res Clin
Pact 1992; 1 6 123-30.
9. Henrikson 0, Paaske WP. Local regulation of Mood flow in the periphetal tissue.
Acta Chir Scand 1980; 502 63-74.
10. ParkhouseN, Le Qwsne PM. Quantitative objeaive aSSeSunent of peripheral
Mxiceptive C fibre function. J Neurol Neurosug Psychiatry I988,51: 28-34.
I I. ParkhouseN, Le Qwsne PM. Impaired neurogenicvascular response in patients
with diabetes and neuropathic foot lesions. N Engl J Med 1988; 318: 1306-9.
12. Walmsley D,wiles PG. Early loss of neurogenic inflammation in the human
diabetic foot Clin Sci 1991; 80:605- 10.
13. BenamKh EE. Low PA The acetyidrdine-induced flare response in evaluation of
~mdlfibre dyrfunction. Ann N w d 1991; 2 9 590-5.
14. Monis SJ, ffinzek S. shore AC. The effect of acetykholine on finger capillary
pressure and capillary flow in hedthy vdunteen. J P h y d 1996; 494: 307- 13.
Skin blood flow in diabetic neuropathy
15. Furchgott RF, Zawadzki ]v. The obligatay role of endothelialcells in the
relaxationof arterial smooth muscle by acetykhdine. Nature (London) 1988; 27:
373-6.
16. PanzaJA, Q w m i A4 Brush JE, Epstein SE. Abnormal endotheliumdependent
vascular relaxation in patients with essential hypertension. N Engl J Med 1990;
3U: 22-7.
17. Wahlberg E. JomeskogG, Olofrron P, SwedenborgJ, Fagrell 6. The influence of
resaive hypetaemia and leg dependency on skin microcirculationin patients with
peripheral arterial occlusive disease (PAOD), with and without diabetes. VASA
1990; 1 9 301-6.
18. Bongard 0, Fagrell 6. Discrepanciesbetween total and nutritional skin
microcirculationin patients wid, peripheral arterial occlusive disease. VASA 1990,
1 9 105-11.
19. Henrich H, Johnsen PC. Influence of arterial pressure on reactive hyperaemia in
skeletal muscle capillaries. Am J Physid I978 u4:352-7.
20. McVeigh GE. Brennan GM, JohnstonGD, et al. Impairedendotheliumdependent
vasodilatation in patientswith type 2 (non-insulin-dependent) diabetes mellis.
Diabetologia I992 3 5 77 I -6.
21. Smits P, Kapma JA,Jacobs MC, LuttermannJ, Thien T. Endothelium-dependent
vascular relaxation in patients with type I diabetes. Diabetes 1993; 4 2 148-53.
22. JiimerkogG, B r i m K, Fagrell 6. Skin capillary circulation is more impaired in
the toes of diabetic than nondiabetic patientswith peripheral vascular disease.
Diabetic Med 1995; I2 36-4 I
23. Flynn MD, Edmondr ME, Tooke JE, Watkins PJ. Direct measurementof capillary
blood flow in the diabetic neuropathicfoot. Diabetdogia 1988; 31: 652-6.
24. JorneskogG, Brismar K, Fagrell 6. Skin capillary circulation severely impaired in
toes of patients with IDDM. with and without late diabetic complications.
Diabetdogia 1995; 38: 474-80.
25. W e r D, Laux G, Dannehl K, et al. Asrertment of cardiovascular autonomic
function: age-related normal ranges and reprcducibili of spectral an*s,
vector analys~s.and standard tests of heart rate variation and blood pressure
response. Diabetic Med 1992; 9 166-75.
26. Claus D, Mustafa C, Vogel W, Hen M, Nwndiirfer 6. Asrertment of diabetic
neuropathy: definition of norm and discrimination of abnormal newe function.
Muscle N m 1993; 1 6 757-68.
27. Bircher A, de Boer EM, Agner T, Wahlberg JE, Serup J. Guidelines for
measurement of cutaneous blood flow by laser doppler flowmetry. Contact
Dermatitis 1994; 3 0 65-72.
26 I
28. Forst T, Wiianer 4 Arin M, et al. Effect of tmnscutaneouselearid newe
stimulation on total skin microcirculationin the diabetic foot. Diabetdogia 1995;
38: A79.
29. Tooke JE. hrgreen J. Fagrell 8. Synchronous assesunent of human skin
microcirculationby laser Doppler flowmetry and dynamic capillaroxopy. Int
J Microcirc Clin Exp 1983; 2 277-84.
30. Rendell M, Bergmann T, ODonnell G. Drobny ED, BorgosJ, Bonner RF.
Miawaxular Mood flow, vdume, and velocity measured by laser doppler
techniques in IDDM. Diabetes 1989; 38: 819-24.
3 I. Shami SK, Chittenden 51. Microangiopathy in diabetes mellis: II.Features
complications and investigation. Diabetes Res I99I; 17: 157-68.
32. Tooke JE. Methoddogies used in the study of the microcirculationin diabetes
m e l l i . Diabetes Metab Rev 1993; 9 57-70.
33. Watkins PJ, Edmonds ME. Sympathetic nerve failure in diabetes. Diabetdogia
1983; 2 5 73-7.
34. L e m k k F. Sir Thomas Lewis nocifensor system, histamine and substanceP-containing primary afferent nerws. Trends Neuroxi 1983; 6: 106-8.
35. Wertermann RA, Widdop RE, Hogan C, Zimmet P. Noninvasivetests of
nwrOvaxular responses in diabetes m e l l i . Neuroxi Lett 1987; 81: 177-82.
36. Westermann R, Widdop R, Low A, Handord J, Kozak W, Zimmet P.
NoninMsivetem of neurovascularfunction: reduced axon reflex respwes in
diabetes mellis of man and meptozotocindiabetes of the tat. D
b Res
Clin Pmct 1988; 5 49-54.
37. Vanhoutte PM. Levy MN. Prejunctid chdinergic modulation of adrenergic
neurotransmission in the cardiovaxular +em. Am J Physid 1980; 238:
H275-81.
38. Douglas WW, k h i e JM. The excitatay action of acetylchdine on cutaneous
non-myelinatedfibres. J Physiol 1960; I 5 0 501-14.
39. Rayman G, Williams S, Hassan A, Gamble J, T d e JE. Capillary hypertensionand
owrperfusion in the feet of young diabetics. Dabet Med 1985; 2: 304
40.Sandemann DD, Shore AC, Tooke JE. Relation of skin capillary p u r e in
patients with insulin dependent diabetes d l i s to complications and metabdic
control. N Engl J Med 1992; 327 760-4.
4 I. Flynn MD, Tooke JE. Aetiology of diabetic foot uketation: a role for the
microcirculation. Diabetic Med 1992; 8: 320-9.
42. Tooke JE. Microvascular function in human diabetes. Diabetes 1995; 44:721-6.
43. Bollinger A Frey J,Jiger K, Furrer J, SegliasJ, Siegenthaler W. Patterns of
diffusion through skin capillaries in patients with long term diabetes. N Engl
J Med 1982; 307 1305- 10.