Clinical Science (1993) 6,701-708 (Printed in Great Britain)
70I
Brisk walking and serum lipoprotein variables in formerly
sedentary men aged 42-59 years
D. J. STENSEL, A. E. HARDMAN, K. BROOKE-WAVELL*, D. VALLANCET, P. R. M. ]ONES*,
N. G. NORGAN* and A. F. WINDER7
Department of Sports Science and *Department of Human Sciences, loughborough University,
Loughborough, U.K., and ?Deportment of Chemical Pathology and Human Metabolism, Royal
Free Hospital Medical School, London, U.K.
(Received
30 AprillU) July 1993; accepted 29 July 1993)
1. The purpose of this study was to examine the
influence of brisk walking on serum lipoprotein
variables.
2. Seventy-two apparently healthy but physically inactive men (aged 42-59 years) were recruited. These
men were normotensive non-smokers without a
history of dyslipidaemia. Subjects were randomly
allocated on a 2 to 1 basis into either a walking
group (n=48) or a control group (n=24). Walkers
followed a self-monitored programme of brisk walking for 1 year, whereas control subjects maintained
their habitual lifestyle.
3. Treadmill walking tests were conducted to examine
changes in fitness. Concentrations of serum lipids and
lipoproteins were determined in fasting subjects. The
amount of body fat was measured by body density.
Circumferences at the waist and hip and skinfold
thicknesses were used to determine the distribution of
body fat. Dietary intakes were assessed by weighed
food inventories.
4. Seven subjects (six walkers and one control subject) dropped out during the study. Walkers did an
average of 28 (SEM 1.4; n=42) min of brisk
walking/day. This improved endurance fitness but did
not influence serum concentrations of total cholesterol, highdensity lipoprotein cholesterol, triacylglycerol, apolipoprotein A-1, apolipoprotein B or
lipoprotein (a). Neither body mass nor the amount of
body fat changed, relative to control subjects.
5. These data suggest that brisk walking does not
modify lipoprotein metabolism in normolipidaemic
middle-aged men.
INTRODUCTION
Epidemiological studies report a lower risk of
ischaemic heart disease (IHD) in phyically active
middle-aged men than among their less active
counterparts [l-31. The mechanism by which this
reduction in risk is conferred is unknown but may
include an effect on lipoprotein metabolism, as
endurance-trained men have more favourable lipid
and lipoprotein profiles than active men, most
noticeably high concentrations of serum highdensity lipoprotein (HDL) cholesterol [4]. In the
light of the strong evidence that a low serum
concentration of HDL cholesterol is an important
independent risk factor for IHD [5, 61, a number of
studies have investigated the potential of exercise to
influence lipoprotein metabolism in middle-aged
men [7-91. Most have employed jogging, a vigorous
form of exercise for these men, but, from the public
health point of view, it is important to investigate
the potential of less intense exercise to influence this
aspect of cardiovascular risk.
The purpose of the present study was therefore to
examine the effectiveness of brisk walking, a socially
acceptable form of exercise carrying a low risk of
injury, in modifying serum lipid and lipoprotein
variables.
METHODS
Experimental design
Men were studied in three cohorts, starting at 6
week intervals. Each cohort comprised walkers and
control subjects in a proportion of 2:1, and random
assignment took place on completion of baseline
tests. Control subjects undertook not to make any
changes to their exercise habits during the study,
whereas walkers followed a progressive, unsupervised programme of brisk walking. Compliance
with the programme was assessed by examination of
training diaries and by changes in the responses to a
standardized laboratory exercise test. All participants were asked not to make conscious changes to
their normal dietary habits and measurements were
repeated after 12 months.
Subjects
Seventy-two men aged between 42 and 59 years
were enrolled in the study, which had the approval
Kay words body cornpition, endurance fitness, exercise, lipoproteins, men.
Abbmvirtlw: HDL, highdensity lipoprotein; IHD. irhrcmic heart disease; Lp(r), lipoprotein (a), VLDL, very-lowdensity lipoprotein.
Corraspndanca Dr A. E. Hardman, Department of Spom Science, Loughborough University, Loughborough, Leicestershire LEI I 3TU, U.K.
D. J. Stensel
702
TabkI. Age and some physical characteristics of the subjects at
baselinu Values are mean (SEM).
Walkers
(n=42)
~
___
Age ( Y e 4
Height (m)
Body mas (kt)
Body mas index (kg/ml)
Predicted maximal 0,uptake (rnlmin-'kn-')
Controls
(n =U)
~
50.3 (0.8)
51.6 (1.0)
1.77 (0.01)
79.1 (1.5)
25.3 (0.4)
35.9 (4.5)
1.77 (0.01)
78.2 (25)
24.9 (0.6)
35.2 (5.3)
of the University's Ethical Advisory Committee.
Volunteers responded to advertisements placed in
the University and in the town. Graded by occupation [lo], six were upper middle-class, 34 middleclass, 20 lower middle-class and five skilled working
class. None was taking any medication known to
influence cardiovascular responses or lipid/
lipoprotein metabolism.
Subjects were appraised of the risks and demands
involved, and gave their informed consent. Criteria
for acceptance were (i) neither currently following a
programme of regular exercise nor employed in a
strenuous occupation, (ii) free of known cardiovascular disease, (iii) resting arterial blood pressure
below 160/90mmHg, (iv) plasma total cholesterol
concentration less than 6.7mmol/l, (v) non- or occasional smoker and (vi) willing to accept random
assignment. An undertaking was given that an
exercise programme would be offered to control
subjects at the end of the year. Seven subjects
dropped out over the year, two walkers because of
lack of time and one because of a recurrence of a
back injury. The remainder (three walkers, one
control subject) left the study because of ill-health.
Age and some of the physical characteristics of the
65 men who completed the study are shown in
Table 1.
Measurements
Changes in endurance fitness were assessed using
an incremental, uphill walking test on a motorized
treadmill. Each subject walked at a constant speed
(1.34 or 1.56m/s) for 4min up each of four increasingly steep gradients. Speed and grades were
selected to elicit approximately 50, 60, 70 and 80%
of each man's predicted maximal oxygen uptake.
Expired air samples were collected using Douglas
bags and were analysed using a paramagnetic
oxygen analyser (Taylor-Servomex) and an infra-red
carbon dioxide analyser (Lira MSA), calibrated
against reference gases of certified composition. Gas
volumes were measured using a dry gas meter and
were corrected to STPD. Heart rate was monitored
using electrocardiography (modified Lead 1).
Fingerprick samples (20 pl) of capillary blood were
obtained before exercise and at the end of each
et
al.
4 min test stage, immediately deproteinized and
stored at -20°C before analysis for lactate [ll].
Oxygen uptake at a reference blood lactate concentration of 2mmol/l was employed as a sensitive
index of endurance fitness which can be measured
directly during submaximal exercise.
At baseline and after 12 months, two 20ml blood
samples were obtained, not more than 5 days apart,
after an overnight fast. These were drawn from an
antecubital vein without stasis. Serum was separated
and stored at -70°C. With the exception of lipoprotein (a) [Lp(a)], which was assayed in one
sample only, both samples were analysed and the
mean value was adopted. Cholesterol concentration
was determined using a cholesterol oxidase method
(Boehringer, Mannheim) in serum and in the supernatant after heparin/manganese precipitation of low
and very-low-density (VLDL) lipoproteins [ 123.
Subfractions of HDL cholesterol are not reported
because of problems experienced in measuring
HDL, in frozen HDL-only-containing supernatant.
Triacylglycerol was determined by measurement of
glycerol after hydrolysis (Boehringer-Mannheim).
Imprecision was evaluated using standard sera
(Precinorm; Boehringer-Mannheim) and day-to-day
coefficients of variation were: < 3.2% and < 3.6% for
total and HDL cholesterol, respectively, and 4.5%
for triacylglycerol.
Apolipoproteins A-I and B were measured using
immunoturbidimetric methods using specific antisera (Incstar, Wokingham, Berks, U.K.). Assays
were monitored using freeze-dried control sera
(Incstar and Boehringer-Mannheim) and a mixture
of freeze-dried and fresh human sera provided
through a multi-centre national control scheme.
Coefficients of variation were below 2.4% and 1.8%
for apolipoproteins A-I and B, respectively. Lp(a)
was measured by an e.1.i.s.a. (Immuno Limited,
Sevenoaks, Kent) using a polyclonal first antibody
against apolipoprotein (a) and a second antiapolipoprotein (a) antibody conjugated to peroxidase. Assays were monitored using low and high
controls (Immuno) [coefficients of variation were
6.8% and 6.5% at mean Lp(a) concentratons of 20
and 45 mg/dl, respectively], human serum frozen at
- 70" C (coefficients of variation 6.2% and 5.8% at
20 and 55mg/dl, respectively) and a mixture of
freeze-dried and fresh human sera provided through
a multi-centre national control scheme.
With the exception of Lp(a), assays were performed after storage of samples for between 3 and
12 months. For Lp(a) all samples were assayed at
the end of the study, so that baseline samples had
been stored for between 13 and 20 months and
samples obtained after training had been stored for
between 1 and 7 months. All samples from each
cohort of men (i.e. walkers and control subjects in a
2:l ratio) were assayed at the same time to ensure
that any between-group differences were identified.
Body density was determined by hydrostatic
weighing [131, with simultaneous measurement of
Brisk walking and lipoprotein variables
lung residual volume using a three-breath nitrogen
dilution technique [14]. Percentage of body mass as
fat was estimated from density [l5]. Fat distribution
was assessed from measurement of body circumferences and skinfold thicknesses [16].
Dietary intakes were assessed from 7-day weighed
food inventories [17, 181 with subsequent analysis
using a computerized version (Microdiet, Salford
University, Salford, Lancs, U.K.) of food composition tables [19] and updates [20-231. Resting
arterial blood pressure was determined while subjects were seated, by one observer using a random
zero spygmomanometer. Vital capacity and forced
expiratory volume in 1 s were measured using
spirometry.
703
/_-
a
75th percentile
:
50th percentile
Zsth percentile
minimum
0
13
26
39
52
Time (weeks)
Fig. 1. Amount of brisk walking performed per day by the walking
group over each 4 week period of the study.
Exercise programme
Brisk walking was unsupervised, but each walker
received individual instruction throughout. Walkers
started with two or three 20 min brisk walks/week
and built up to an average of 20-25min/day and by
the end of 3 months and to an average of 4045min/day by the end of 6 months, maintaining
that level thereafter. The pace of walking was selfselected, the instructions being ‘Walk at a brisk
pace which is faster than you would usually walk
and which makes you feel slightly breathless’.
Brisk walking pace was assessed by a 1.61km
walk around an athletics track during which heart
rate was monitored by short range telemetry
(Sportstester; Polar Electro). Every 2 weeks walkers
submitted a training diary recording the number of
minutes of brisk walking.
Statistical analyses
The change between baseline and 12 months was
compared in walkers and control subjects using
Student’s t-test for independent means, adopting a
5% level of significance. Relationships were examined using Pearson’s product moment correlation
coefficient. Results are expressed as mean and SEM,
unless otherwise stated. Triacylglycerol and Lp(a)
data were not normally distributed and were transformed using natural logarithms before statistical
analysis.
RESULTS
Walkers completed an average of 28 (SEM 1.4;
n=42) min of brisk walking/day. The amount of
walking during each each 4 week period is shown in
Fig. 1, which presents median values and those for
men in the 25th and 75th centiles, as well as the
highest and lowest values. Brisk walking speed
increased in walkers but the change was not different from that shown by control subjects (Table 2).
There were, however, training-induced adaptations
in the walkers. Figs. 2 and 3 show that heart rate
and blood lactate concentration were reduced
T a b l e Z Changes in oxygen uptake at a reference blood lactate
concentration, brisk walking speed and heart rate during brisk
walking in walkers ( n 4 2 ) and control subjects ( n =U).Values are
mean (SEM). The asterisk indicates that a change between baseline and I2
months was significantly different between walkers and control subjects
(P <0.05).
O1 uptake at a [lactate] of Immol/l
(mlmin-l kg-I)
Walkers
Controls
Brisk walking speed (m/s)
Walkers
Controls
Heart rate during
track walk (beatstmin)
Walkers
Controls
Bveline
12 months
21.5 (0.7)
20.3 (0.6)
22.9 (0.7)*
18.8 (0.8)
I .90 (0.02)
1.88 (0.03)
I20 (3)
120 (3)
I .98 (0.03)
1.91 (0.03)
I 19 (3)
I20 (3)
during standardized treadmill exercise in walkers
and that these responses were different from those
of control subjects. The decrease in heart rate
during each test stage was related to the amount of
brisk walking done (r=0.31-0.37, degrees of freedom = 41). Oxygen uptake at a reference blood
lactate concentration of 2 mmol/l increased in
walkers, but decreased in control subjects, these
responses being significantly different (Table 2).
Despite the improvements in indices of endurance
fitness, brisk walking had no influence on the
concentrations of any of the lipid or lipoprotein
variables measured (Table 3). There were no
between-group differences in Lp(a) over time, but,
for the group as a whole, the mean concentration
was significantly higher at 12 months than at baseline. This increase was apparent whether samples
(24 of 140) with values on or below the detection
limit of the assay ( 5 mg/dl) were either omitted
from analysis or assigned a value of lmg/dl.
704
i”
f
I30
5
I20
D. J. Stensel et al.
8
I10
loo
0
I
2
3
0
I
2
Test stage
3
4
0
I
2
3
4
0
I
1
3
4
I
4
Fig. 2. Heart rate in ( a ) walkers and ( b ) control subjects at
baseline (0)
and at I2 months ( 0 )during each stage of the
incremental treadmill walking test. Values are means with bars
indicating SEM.
Test stage
Fig. 3. Blood lactate concentration in ( a ) walkers and ( b ) control
subjects at baseline (0)
and at I2 months ( 0 )during each stage of
the incremental treadmill walking test. Values are means with ban
indicating SEM.
DISCUSSlON
Data relating to the amount and distribution of
body fat are shown in Table 4. Body weight
increased slightly in control subjects and decreased
slightly in walkers, and a somewhat greater decrease
in body fatness was observed in walkers than in
control subjects, but between group differences were
not significant. Neither body circumference, waist/
hip ratio nor the sum of four skinfold thicknesses
changed in a different way in walkers and control
subjects.
Results of the analysis of food intake inventories
are shown in Table 5. Mean energy intake was
unchanged for both groups. Similarly, intakes of
carbohydrate, total fat and cholesterol showed no
significant change, although the ratio of polyunsaturated to saturated fat increased when all
subjects were considered together. The change in
protein intake differed significantly between groups,
decreasing in walkers and increasing in control
subjects.
There was no significant change in systolic blood
pressure [walkers mean 115 (SEM 2; n=42)mmHg
versus 114 (SEM 2; n=42)mmHg at baseline and
12 months, respectively] or diastolic blood pressure
[walkers mean 79 (SEM 2; n=42)mmHg versus 80
(SEM 2; n=42)mmHg] or in lung function
measures.
The subjects were middle-aged sedentary males, at
the upper limit of the desirable range of weight for
height, with more favourable blood lipid profiles
than the typical British male. Mean serum total
cholesterol concentration (5.7 mmol/l) was lower
than values reported for asymptomatic British men
of comparable age, i.e. 6.0-6.3mmol/l [26, 271.
Nevertheless, the average total cholesterol concentration was higher than desirable and thus a majority of these men would be advised to make dietary
change in an attempt to decrease their risk of IHD
[28]. Serum triacylglycerol concentrations were also
lower than values (not log-transformed) reported for
a more representative sample than the present
group and mean HDL cholesterol concentration,
measured with the same precipitation technique,
was 14% higher [27]. Furthermore, because of the
criteria established for entry into the study, the men
were normotensive and non-smokers. The subjects,
therefore, were physically inactive but otherwise not
at high risk of IHD. Random allocation of subjects
into walkers and control subjects resulted in two
groups which were similar with regard to the lipid
and lipoprotein variables of interest as well as
indices of body fatness and endurance fitness
(Tables 2, 3 and 4).
So that the effect of a socially acceptable change
Brisk walking and lipoprotein variables
Tabla 3. Lipid and lipoprotein variables In walkers ( n =a) and
control subjects (n323). Values are mean (SEM).
Baseline
Triacylglycerol (mmol/l)*
Walkers
Controls
1.3 (1.1-1.5)
1.4 (1.1-1.8)
5.7 (0.2)
5.7 (0.3)
5.8 (0.2)
5.6 (0.2)
1.4 (0.1)
1.4 (0.1)
1.3 (0.1)
1.3 (0.1)
T d cholesterol (mmol/l)
Walkers
Controls
HDL cholesterol (mmol/l)
Walkers
Controls
Lowdensity lipoprotein
cholesterol (mmol/l)
Walkers
Controls
VLDL cholesterol (mmol/l)t
Walkers
Controls
Tabla5. Average daily intake of energy and m a nutrients for
walkers ( n 139) and control wbjcds ( n 121). Values are mean (SEM).
12 months
1.3 (1.1-1.6)
1.4 (1.1-1.8)
705
Baseline
12 months
10.6 (0.2)
10.8 (0.3)
10.6 (0.3)
10.9 (0.4)
103 (4)
107 (4)
I05 (4)
109 (6)
290 (10)
299 (12)
286 (13)
293 (15)
297 (I8)
323 (19)
Protein (g)
Walkers
Controls
(I)
Walkers
Controls
fit
3.6 (0.1)
3.6 (0.3)
0.7 (0.1)
0.7 (0.1)
3.8 (0.2)
3.6 (0.2)
0.7 (0.1)
0.8 (0.1)
Grbohydrate (g)
Walkers
Controls
Cholesterol (mg)
Walkers
Controls
306
293 (I5)
Polyunraturated/saturated fat ntio
Apolipoprotein A-I (g/l)
Walkers
Controls
Apolipoprotein ElW (g/l)
Walkers
Controls
I.42 (0.04)
1.42 (0.07)
I .4l (0.03)
1.40 (0.04)
on (0.03)
0.73 (0.04)
0.79 (0.03)
0.78 (0.04)
13.2 (8.5-20.6)
7.1 (3.6-14.1)
16.8 (I1.1-25.5)
13.3 (8.1-21.8)
Tabla4 Body composition and rnthropomatric measurnants of
mlhn (n =a)and control subjects ( n 123). Values are mean (SEM).
Baseline
I2 months
Body
(kd
Walkers
Controls
79. I (I.5)
78.2 (2.5)
78.9 (I.5)
78.9 (2.7)
Body density (kg/m')
Walkers
Controk
I034 (2)
1032 (3)
1036 (I)
1032 (3)
28.8 (0.8)
29.9 (I.5)
27.7 (0.8)
29.6 (I.6)
51.4 (2.1)
51.2 (4.1)
50.5 (2.3)
49.9 (3.7)
(%I
Walkers
Controls
Sum d four skinfdd
thicknow (mm)
Walkers
Controls
Waisthip circumference ratio
Walkers
Controls
0.41 (0.04)
0.45 (0.04)
0.46 (0.03)
0.46 (0.05)
The asterisk indicates that I change between baseline and 12 months was
significantly different between walkers and control subicco (P<0.05).
'Antilogs of the mean and 95% confidence intervals of I q transformed data [24].
tWculated using the Friedwald formula P I , LDL by subtraction.
ht
Walkers
Controls
0.946 (0.010)
0.953 (0.014)
0.948 (0.0lO)
0.947 (0.015)
in exercise behaviour could be evaluated, the exercise programme was deliberately unsupervised and
flexible. It was important, therefore, to provide
evidence of the amount of exercise done and of the
results of that exercise. Inspection of walking diaries
showed that, although adherence to the exercise
programme was variable (Fig. l), walkers reported
an average of some 30min of exercise/day. This was
equivalent to around 14 miles of brisk walkindweek
and is greater than the level achieved in many other
training studies. In support of the diaries, laboratory
measures showed clear evidence of a training effect:
decreases in heart rate and blood lactate concentration were particularly clear at higher exercise
intensities (Figs. 2 and 3) and, in the outdoor
walking test, walkers were able to walk at a faster
speed for an unchanged heart rate (Table 2). There
was little change in the response to exercise in
control subjects, suggesting that they had complied
with the request not to alter their physical activity
levels.
Despite the improvements in endurance fitness
there was no change in any of the lipid or lipoprotein variables measured as a result of brisk
walking. The lack of change in total cholsterol is
probably not surprising as total cholesterol does not
differ consistently between trained and sedentary
subjects [4] and a majority of longitudinal studies
report no change in this variable, (e.g. [8, 293).
HDL cholesterol concentration has, however, frequently (but not invariably) been found to increase
706
D. j. Stensel et al.
with training in middle-aged men [7, 30, 311, in
conflict with the present observations. The lack of
change in HDL cholesterol concentration in our
men is consistent with unchanged values for
apolipoprotein A-I, the main apolipoprotein of
HDL. Furthermore, there is no evidence that coexisting changes in dietary patterns or body fatness
levels have confounded interpretation of these data.
We must conclude that this exercise programme,
selected for its widespread applicability, does not
result in increases in HDL cholesterol concentration
in middle-aged normolipidaemic men,
It is possible that there is a threshold of exercise
energy expenditure below which HDL cholesterol
concentration does not increase. Eight to 10 miles of
running/week [31] and the expenditure of more
than about 5 M J in exercise/week [4] have been
suggested as minimum ‘doses’ of exercise likely to
influence lipoprotein metabolism. We used measurements of oxygen uptake and carbon dioxide production during treadmill walking to estimate the energy
expenditure of walking. Our conservative estimate
of the average weekly energy expenditure by the
men in brisk walking is 5.5MJ, above the threshold
suggested. Moreover, we examined the HDL
cholesterol data in subgroups of men whose average
exercise time was equivalent to fewer than 10 miles/
week, between 10 and 15 miles/week and over 20
miles/week. This failed to reveal any effect which
was present only in those walkers who did more
exercise. Therefore, in this group of men, too low an
exercise energy expenditure is probably not an
explanation for the lack of change in HDL
cholesterol concentration. Indeed the concept of a
minimum energy expenditure may not be appropriate as some studies using training programmes
likely to involve similar or lower levels of energy
expenditure than achieved in the present study have
reported increases in HDL cholesterol concentration
[8, 321.
An alternative explanation is that brisk walking
did not constitute sufficiently vigorous exercise to
influence lipoprotein metabolism. Previous studies
reporting an increase in HDL cholesterol concentration have employed jogging, running or stationary cycling [7, 8, 301. A lack of consistency in
the reporting of training intensity, if at all, makes
comparison difficult, but in the present study exercise intensity averaged 68% of predicted maximal
heart rate, corresponding to about 57% of maximal
oxygen uptake [33]. Most studies finding increases
in HDL cholesterol concentration have employed
intensities of 70% or more of maximal heart rate
(e.g. [9, 31, 343). We therefore examined our data
for evidence of a change in those subjects walking at
higher heart rates by dividing the exercise group
into tertiles according to heart rate during brisk
walking, but, even in the subgroup walking at
speeds eliciting heart rates equivalent to 80% of
maximum, there was no change in HDL cholesterol
concentration. No relationship between changes in
oxygen uptake at a reference blood lactate concentration, an index reflecting adaptations of skeletal
muscle, and changes in HDL cholesterol concentration was apparent.
Changes in HDL cholesterol concentration may
be enhanced when an exercise programme is accompanied by decreases in body fatness [35] and the
lack of change in fatness in the walkers may be
important. This lack of change is surprising in the
light of the considerable amount of energy expended
in walking. The food intake data provide no support for the view that energy intake increased to
offset the increased energy expenditure. It should be
noted, however, that this type of information has
limited precision and subjects are known to alter
their dietary behaviour when keeping food inventories [18].
As might be expected, some subjects (23 of 42
walkers) did show a decrease in body fat during the
study, but, even among these men, HDL cholesterol
concentration remained unchanged. Moreover, in a
previous study of women we found that brisk
walking increased HDL cholesterol concentration in
the absence of decreases in body fatness [36, 371.
Hormonal influences on lipoprotein metabolism
may explain these conflicting findings: oestrogen
increases HDL cholesterol concentration, possibly
because of enhanced turnover of VLDL [38], and it
is possible that women are more sensitive than men
to exercise-induced changes in HDL cholesterol
concentration.
Increases in HDL cholesterol concentration with
training may accrue from peripheral changes in
trained muscle, independent of changes of body
fatness: enhanced capillarization is accompanied by
increased lipoprotein lipase activity and hence accelerated degradation of triacylglycerol-rich lipoproteins, increasing the transfer of surface materials
to nascent HDL [39]. In the present study walkers
showed a marked decrease in blood lactate concentration, clear evidence of an adaptive increase in
muscle oxidative capacity, but no increase in HDL
cholesterol concentration. This may mean that brisk
walking improves muscle oxidative capacity by
mechanisms other than improvements in the microcirculation, possibly increases in mitochondria1
protein [40]. Certainly studies reporting improved
capillarization after exercise training have involved
more strenuous exercise than the present regimen
[41, 421.
The lack of influence of brisk walking on apolipoprotein A-I and apolipoprotein B conflicts with
some earlier reports [9, 301 but confirms others
[8, 31, 43, 441, while the lack of change in triacylglycerol supports the suggestion that exercise training only lowers plasma triacylglycerol when pretraining values are elevated above 1.35-1.69 mmol/l
~41.
There are few data on the effects of environmental
factors on Lp(a) concentration and the present
study presents new information in this regard. Diet-
Brisk walking and lipoprotein variables
ary changes and drug regimens which are effective
in ameliorating other lipid risk factors have little or
no effect on Lp(a) [45, 461, although weight loss by
dieting reduced Lp(a) concentration by 19% in
obese patients [47]. Only one report has been found
of the effects of exercise, an average decrease of
22% in 16 men after 8 days of cross-country skiing
[48]. In the present study Lp(a) concentration increased similarly in walkers and control subjects.
Concentrations of this macromolecular complex
may be under rather strict genetic control [49] and
therefore relatively insensitive to environmental
influences.
The reason for the increase in Lp(a) concentration
in the present study is unknown. It was not attributable to changes in a few atypical individuals. The
apparent small increases in Lp(a) concentrations at
the lower end of the distribution may be real, but
levels less than the lower limit of detection for this
assay (5 mg/dl) were for statistical purposes recorded
as lmg/dl. These apparent changes in Lp(a) concentrations are not, in our view, of proven significance. An effect of storage before analysis is possible, although values from re-analysis of a second
aliquot of 12 samples after storage at -70°C for 1
additional month showed no evidence of such an
effect. Furthermore, one recent report suggested
that levels obtained by e.l.i.s.a., the method
employed here, may be less affected by sample
storage than those obtained by the radial immunodiffusion method [SO].
Epidemiological evidence has associated both the
amount and pace of habitual walking with the risk
of heart attack in middle-aged men. Amongst
Harvard alumni there was a 21% decrease in allcause mortality (overwhelmingly from IHD) as the
amount of walking increased from fewer than 3
miles/week to more than 9 miles/week [Sl]. Among
British civil servants, men who rate their usual
pace of walking as fast (>4m.p.h.) showed a lower
attack rate than other men [2]. In the British
Regional Heart Study, men who walked fast for
more than a mile each way to and from work
exhibited a lower risk than other men [3]. The men
in the present study showed marked improvements
in endurance fitness as a result of walking briskly
for an average of about 30min/day over a long
period and yet lipid and lipoprotein profiles were
unchanged. Our findings do not support the suggestion that regular walking modifies lipoprotein
metabolism in middle-aged men and strengthen the
argument for research into alternative mechanisms
which might mediate the effect of exercise on the
risk of IHD.
ACKNOWLEDGMENTS
This work was supported by the British Heart
Foundation.
707
REFERENCES
I. Menbarger RS, Wing AL, Hyde RT. Physical activity as an index of heart
attack risk in college alumni. Am J Epidemiol 1978; IQI: 161-75.
2. Morris IN, Clayton DG. Everitt MG. Semmence AM, Burgar EH. Exercise in
leisure time: coronary attack and death rates. Br Heart J 1990; 0: 325-34.
3. Shaper AG, Wannamethee G. Physical activity and ischrmic heart diswe in
middle-aged British men. Br H u r t J 1991; 6& 384-94.
4. Haskell WL. The influence of exercise training on plasma lipids and
lipoproteins in health and diswe. A m Med Sand 1986 711 (Suppl.): ?S-37.
5. Multiple Risk Factor Intervention Trial Research Group. Relationship h e e n
baseline risk factors and coronary heart disease and total mortality in the
Multiple Risk Factor Intervention Trial. Prev Med 1986; IS: 254-63.
6. Wilson PWF, Abbott RD, Castelli WP. High density lipoprotein cholesterol
and mortality. The Framingham Heart Study. Arteriosclerosis 1988; k 73741.
7. Wood PD, Stefanick ML. Dreon DM, et al. Changes in plasma lipids and
lipoproteins in overweight men during weight lorr through dieting as
compared with exercise. N Engl J Med 1988, 319: 1172-9.
8. Marti B, Suter E, Risen WF, Tschopp A, Wanner KU , Gutmiller F. Effects
of long-term, selfmonitored exercise on the serum lipoprotein and
apolipoprotein profile in m i d d l q e d men. Atherosclerosis 1990; 81: 19-31.
9. Stubbe I, Hanwn P. Gustafson A, Nillson-Ehle P. Plasma lipoproteins and
lipolytic enzyme activities during endurance training in sedentary men:
changes in high-density lipoprotein subfractions and composition. Metab Clin
Exp 1983; 32 1120-8.
10. Abrams M. Education, social class and readership of newspapers and
magazines. London: Joint Industry Committee for National Readership
Surveys, 1968.
II . Maughan RJ. A simple, npid method for the determination of glucose, lactate,
pyruvate. alanine, Mydroxybutyrate and acetoacetate on a single 2 0 ~ blood
1
ample. Clin Chim Acta 1982 Iz1: 23140.
12. Gi d u LE, Miller GJ, Burstein M. Slagle 5, Eder HA. Separation and
quantitation of subclasses of human plasma high density lipoproteins by a
simple precipitation method. J Lipid RM 1981; U: 1206-23.
13. Jones PRM, Norgan NG. A simple system for determination of human body
density by underwater weighing. J Physiol (London) 1974 u(: 71-3P.
14. Rahn H. Fenn WO, Otis AB. Daily variations of viul capacity, residual air,
and expiratory reserve including a study of the residual air method. J Appl
Physiol 1949 I: 725-36.
IS. Siri WE. Gross composition of the body. Adv Biol Med Phys 1961; I: 239-80.
16. Weiner JS, Lourie ]A. Practical human biology. London: Academic Press, 1981:
87-97.
17. Marr JW. Individual dietary surveys: purposes and methods. World Rev Nutr
Diet 1971; 13: 105-64.
18. Bingham SA. The dietary assessment of individuals: methods, accuracy, new
techniques, and recommendations. Nutri Abstr Rev 1987; 51: 7 0 9 2 .
19. Paul AA. Southgate DAT. McCance and Widdown’s ‘The composition of
foods’. London: HMSO. 1978.
20. Paul AA, Southgate DAT, Russell J. First supplement to McCance and
Widdowson’s ‘The composition of foods’, London: HMSO, 1980.
21. Tan SP, Wenlock RW. Buss DH. Immigrant foods. Second supplement to
McCance and Widdowson’s ‘The composition of foods’, London: HMSO, 1985.
22. Holland ID, Unwin ID. Buss DH. Cereals and cereal products. Third
supplement to McCance and Widdomon’s ‘The composition of foods’.
Nottingham: Royal Society of Chemistry and Ministry of Agriculture, Fisheries
and Food, 1988.
23. Holland ID, Unwin ID, Burr DH. Milk products and eggs. Fourth supplement
to McCance and Widdowson’s ‘The composition of foods’. Cambridge: Royal
Society of Chemistry and Ministry of Agriculture, Fisheries and Food. 1989.
24. Gardner MJ, Altman DC. Statistics with confidenctconfidence intervals and
statistical guidelines. London: British Medical h i a t i o n , 1989: 24-7.
25. Friedwald WT, Levy RI, Fredricksan DS. Estimation of the concentration of
low density lipoprotein cholesterol in plasma, without the use of the
preparative ultracentrifuge. Clin Chem 1972; Ik 49e502.
26. Pocock SJ, Shaper AG. Philips AN, Walker M, Whitehead TP. High density
lipoprotein cholesterol is not a mljor risk factor for ischaemic heart disease
in British men. Br Med J 1986. 2n: 515-9.
708
D. J. Stensel
27. Mann )I. Lewis B. Shepherd J, e t al. Blood lipid concentrations and other
cardiovascular risk factors: distribution. prevalence and detection in Britain.
Br Med J 1988; 2%: 1702-6.
28. Shepherd J. Beaeridge DJ. Durrington P, et al. Strategies for reducing
coronary heart dieae and desirable limits for blood lipid concentrations:
guidelines of the British Hyperlipidaemia Association. Br Med J 1987; m:
124%.
29. Sasaki J. Shindo M, Tanaka H. Ando M. Arakawa K. A long term aerobic
exercise programme decreases the obesity index and increases the high
density lipoprotein cholesterol concentration in obese children. Int J Obes
1987; II: 33945.
30. Kiens 8. Jorgensen I( Lewis S, et al. Increased plasma HDL-cholesterol and
apo A-I in redimenary middle-aged men after physical conditioning. Eur J
Clin Invest 1980; 10: 203-9.
31. Wood PD. Haskell WL, Blair SN. et al. Increased exercise level and plasma
lipoprotein concentrations: a one-year randomized, controlled study in
sedentary, middle-aged men. Metab Clin Exp 1983; 3k 31-9.
32. Stein RA. Michielli DW, Glantz MD. et al. Effects of different exercise
training intensities on lipoprotein cholesterol fractions in healthy middle-aged
men. Am Heart J 1990; 1 1 9 277-83.
33. American College of Sports Medicine. Guidelines for exercise testing and
prescription. 4th ed. Philadelphia: Lea and Febiger. 1991: 100.
34. h r i a IE. Faria EW. Effect of exercise on blood lipid constituents and aerobic
capacity of fire fighters. J Sports Med Phys Fitness 1991: 31: 75-81,
35. Tran ZV. Weltman A. Differential effects of exercise on serum lipid and
lipoprotein levels seen with changes in body weight. J Am Med Assoc 1985;
W: 919-24.
36. Hardman AE. Hudson A, Jones PRM, Norgan NG. Brisk walking and plasma
high density lipoprotein cholesterol concentration in previously sedentary
women, Br Med J 1989; 199: 1204-5.
37. Hardman AE, Jones PRM, Norgan NG. Hudson A. Brisk walking improves
endurance fitness without changing body fatness in previously sedentary
women. Eur J Appl Physiol 1992; a5: 354-9.
et al.
38. Miller VT. LaRosa JC. Sex steroids and lipoproteins. In: Redmond GP, ed.
Lipids and women’s health. New York: Springer-Verlq, 1991: 48-65.
39. Kiens B. Lithell H. Lipoprotein meubolism influenced by training-induced
changes in human skeletal muscle. J Clin Invest 1989; 8& 558-64.
40. Holloszy JO, Coyle EF. Adaptations of skeletal muscle to endurance exercise
41,
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
and their metabolic consequences. J Appl Physiol: Respir, Environ Exer
Physiol 1984; 5& 831-83.
lngier F. Effects of endurance training on muscle fibre ATP-ase activity.
capillary supply and mitochondria1 content in man. J Physiol (London) 1979
294 419-32.
Andersen P, Henriksson J. Capillary supply of the qudriceps femoris muscle
of man: adaptive response t o exercise. J Physiol (London) 1977; 110:
677-90.
Weintraub MS, Rosen Y. Otto R. Eisenberg 5, Breslow JL. Physical exercise
conditioning in the absence of weight 101s reduces fasting and postprandial
100744.
triglyceride-rich lipoprotein levels. Circulation 1989;
Despres JP, Tremblay A, Moorjani S, et al. Long term exercise training with
constant energy intake. 3. Effects on lipoprotein levels. Int J Oba 1990; II:
85-94.
Lawn RM. Lipoprotein (a) in heart diswe. Sci Am 1992 26& 26-32.
Loscalzo J. Lipoprotein (a), a unique risk factor for atherothrombotic disease.
Arteriosclerosis 1990; 10: 672-9.
Sonnichsen AC, Richter WO. Schwandt P. Reduction of lipoprotein (a) by
weight IOU. Int J Obes 1990; 14 487-94.
Hellsten G, Boman K, Hallmans G, Dahlen G. Lipids and endurance physical
activity. Atherosclerosis 1989; 7 5 934.
Seed M. Hoppichler F, Reaveley D. et al. Relation of serum lipoprotein (a)
concentration and apolipoprotein (a) phenotype t o coronary heart disease in
patients with familial hypercholesterolemia. N Eng J Med 1990; 311: 1494-9.
Craig WY. Ledue TB. The effects of long term storage on serum Lp(a) levels.
Atherosclerosis 1992; 93: 261.
Paffenbarger RS, Hyde RT. Wing AL. Hsieh C C . Physical activity. alkause
mortality, and longevity of college alumni. N Engl J Med 1986; 314 605-13.