Aerobic Versus Strength Training for Risk Factor Intervention in Middle-Aged

Men at High Risk for Coronary Heart Disease

M.A. Smutok, C. Reece, P.F. Kokkinos, C. Farmer, P. Dawson, R. Shulman, J. DeVane-Bell, J. Patterson,
C. Charabogos, A.P. Goldberg, and B.F. Hurley

To compare the effects of strength training (ST) to those of aerobic training (AT) for coronary heart disease (CHD) risk factor
intervention, we studied 37 previously untrained males (aged 50 -+ 9 years, mean 2 SD) before and after 20 weeks of either ST
(N = 14). AT (walk/jog, N = 13). or no exercise (inactive controls, N = 10). Lipoprotein and lipid profiles, blood pressure, and
glucose and insulin responses to an oral glucose tolerance test (OGlT) were assessed before and after the training period in all
three groups. The ST program produced significant reductions in plasma glucose levels at 60,90, and 120 minutes (P < .05) after
glucose ingestion, whereas the AT program resulted in significant reductions only at 90 and 120 minutes (P c .05). ST also
decreased insulin levels during fasting (P < .05) and at 90 and 120 minutes (P < .Ol) after glucose ingestion. AT decreased
insulin levels at 90 and 120 minutes (P < .Ol) after glucose ingestion. Both training programs reduced the total area under the
glucose tolerance curve for glucose (both P < .05) and insulin (both P < .05), but there were no significant differences in these
changes between the two groups. None of the glucose or insulin values were significantly altered in the control group. There
were no significant changes in lipoprotein and lipid profiles or blood pressure in any of the three groups. These results suggest
that ST and AT have comparable effects on risk factors for CHD. Both ST and AT improve glucose tolerance and reduce insulin
responses to oral glucose in middle-aged men with multiple risk factors for CHD.
Copyright 0 1993 by W.B. Saunders Company

R ECENT REPORTS SHOW a strong and consistent of the 37 subjects who completed all aspects of the study had an
relationship between physical inactivity and coronary abnormal lipid profile as defined by a total cholesterol level of at
heart disease (CHD),‘,’ and suggest that exercise training least 220 mg/dL or a total cholesterol high-density lipoprotein
cholesterol (HDL-C) ratio of at least 5.0 or a low-density lipopro-
should be emphasized in CHD prevention programs. How-
tein cholesterol ratio (LDL-C)/HDL-C ratio of at least 4.0. Twenty
ever, there is some uncertainty regarding the optimal
subjects had an abnormal blood glucose level, with fasting levels of
training modality that should be recommended, because
at least 140 mg/dL or a glucose value of at least 200 mg/dL and a
reductions in specific CHD risk factors have been reported 2-hour value of at least 140 mg/dL during an oral glucose tolerance
with both aerobic training (AT)3 and strength training (ST) test (OGTT),” and 11 were hypertensive as defined by a resting
programs.J Nevertheless, ST is not commonly endorsed to systolic blood pressure of at least 140 mm Hg and diastolic pressure
reduce risk factors for CHD due to its purported lack of of at least 90 mm Hg on at least two separate occasions.
effect on cardiovascular function5 and the potential for
acute elevations in blood pressure during exercise.h Measurement of Aerobic Capacity and Body Composition
Although numerous studies have examined the effects of Aerobic capacity ($Ozmax) was measured during a continuous
either ST or AT on lipid profiles separately, none have treadmill lest as described previously.s Oxygen uptake values were
compared these two training modalities with regard to recorded every 30 seconds using the Medical Graphics Cardiopul-
other risk factors in subjects at risk for CHD. Therefore, monary Exercise System 2001 (St Paul, MN) to analyze ventilation
and gas fractions. GOzmax was established when two of the
this study was designed to compare the effectiveness of ST
following three criteria were met: (1) maximal heart rate was
with that of AT for risk factor intervention in subjects at
within 10 beatsimin of age-predicted values; (2) respiratory ex-
risk for CHD as defined by the presence of two or more
change ratio values were greater than 1.10; and (3) a plateau in
major risk factors. oxygen uptake was achieved, as evidenced by no more than a

SUBJECTS AND METHODS

Subjects
Forty-four untrained men (50 2 10 years, mean 2 SD) volun-
teered for participation in this study after providing their written
informed consent according to guidelines approved by the Institu-
tional Review Board of the University of Maryland, College Park;
seven subjects dropped out of the study before completion.
Subjects volunteered for 20 weeks of either a ST program (n = 14),
an AT program consisting of walking and/or jogging (n = 13), or
an inactive control group (n = 10). Three of the seven dropouts
were in the AT group, two were in the ST group, and two were in
the control group. Subjects were screened for disease by medical
history, physical examination, and a graded exercise test on a
treadmill.’ Only those subjects who were nonsmokers, were not
taking insulin, and had no evidence of CHD were selected for the
study. All subjects were at risk for CHD based on having at least
two of the following three risk factors: (1) abnormal lipid profile;
(2) abnormal blood glucose level, either from impaired glucose
tolerance or type II diabetes; and (3) hypertension. Twenty-eight
From the Exercise Science Laboratory, Department of Kinesiology,
University of Matyland, College Park; and the Department of Medi-
cine, Division of Geriatrics, Francis Scott Key Medical Center, Johns
Hopkins University<Baltimore, MD.
Submitted July 1. 1991; accepted May 5, 1992.
Supported by a grant to B.F.H. and A.P.G. from the American
Heart Association, Maryland Afiliate; The Johns Hopkins Academic
Teaching Nursing Home Award (NIA) to A.P. G. (No. PO1 -AG04402-
6); a grant from the NIH to A.P.G. (No. ROI-AG07660); General
Clinical Research Center Grant (No. MOI-RR02719) to Johns Hop-
kins University School of Medicine; Intramural Funds from the
Metabolism Section, Laboratory of Clinical Phvsiologv Gerontology
Research Center, NIA; and the NIH grant to Washington University
Diabetes Research and Training Center Core Laboratory (No. AM
20579), St Louis, MO.
Address reprint requests to B.F. Hurley, PhD, Depatiment of
Kinesiology. Universityof Maryland. College Park, MD 20742.
Copyright 0 1993 by W.B. Saunders Company
0026049519314202-0007$03.0010

Metabolism, Vol 42, No 2 (February), 1993: pp 177-184 177

178 SMUTOK ET AL

150-mL min-’ or 2-mL kg-’ mm1 increase in oxygen uptake Training Programs
with an increase in workload.9 Subjects who did not meet these
ST. Subjects trained three times a week on nonconsecutive
criteria the first time were retested; all subjects eventually met the
days for approximately 20 weeks using Nautilus equipment (De-
criteria and were included in the data analysis.
land, FL). Before training, all subjects were shown the proper
Body composition was estimated from body density determined
technique for each exercise by a qualified instructor. The training
by hydrodensitometry.‘” Underwater weight was adjusted for
program consisted of two sets using the maximum amount of
residual lung volume using the nitrogen washout technique” on the
weight that could be lifted 12 to 15 times (repetitions maximum
Medical Graphics Pulmonary Function Test System 1070. Percent
[RM]) per set, with 11 different exercises and modified sit-ups.
body fat was calculated from body density values using the formula
Weights were adjusted throughout the training program as strength
of Brozek et al.‘O
levels increased. All subjects engaged in a brief warm-up period
consisting of static stretching and calisthenics before each training
Measurement of Blood Pressure
session. The following exercises were performed during each
Resting blood pressure was measured in the seated position workout: duo squat, leg extension, leg curl. hip and back, decline
before each training session in both training programs. The mean press, pullover. arm cross, behind-the-neck pullover. overhead
of six blood pressure measurements obtained during the first 2 press, lateral raise, rowing torso, and sit-ups. The resting interval
weeks of training was compared with six measurements taken between sets was limited to 90 seconds. Every training session was
during the last 2 weeks of training for both groups. carefully supervised and subjects were encouraged to exercise until
they failed to complete the last repetition attempted. The exercises
Dietary Records performed, amount of weight lifted (number of plates). and
number of sets and repetitions completed for each exercise were
All subjects were instructed by a registered dietitian regarding
recorded for every exercise session.
appropriate techniques for recording food records to ensure
Strength was measured as the maximum load that could be
accuracy. They were also reminded on a regular basis not to change
successfully lifted for one repetition (1-RM). The 1-RM test for
their dietary habits and alcohol intake throughout the program.
baseline upper- and lower-body strength was administered after
Subjects were asked to abstain from alcohol consumption for at
three to five workout sessions to allow subjects to become accus-
least 3 days before blood sampling. Food records were kept for 4
tomed to the exercise. The upper-body exercises included the
days before the initial blood sampling and OGTT, and the same
diet was replicated for the 4-day period before the final blood pullover, decline press, and overhead press; the lower-body exer-
sampling at the end of training. All food records were coded for cises included the duo squat. leg extension. and leg curl. The 1-RM
nutritive value according to US Department of Agriculture guide- was achieved by increasing the load by one plate after each
linesI and were analyzed for food composition and caloric content. successful lift until the maximum load was obtained for each
exercise. Subjects were given a 2-minute rest interval after every
three trials until the 1-RM was reached. These procedures were
Plasma Lipoprotein and Lipid Determinations
followed at both testing periods for both the ST and control groups.
Blood samples for lipid and lipoprotein measurements were AT. Following flexibility and warm-up exercises, subjects walked
obtained from each subject’s antecubital vein after a 12. to 14-hour and/or jogged on a motorized treadmill for 30 min/d 3 days each
overnight fast. Samples were placed into chilled tubes containing week for approximately 20 weeks. During the first week, treadmill
EDTA on two separate mornings within 1 week apart. If values for speed was adjusted to maintain an exercise intensity of 50% to 60%
either total cholesterol or HDL-C differed by more than 7%, a of each subject’s maximal heart rate reserve. This relative intensity
third blood sample was taken on another day; the baseline value was increased to 60% to 70% during the second week and 75% to
reported is the mean of the two determinations that did not differ 85% thereafter. Heart rates were periodically monitored through-
by more than 7%. After training, blood samples were drawn once out the training program. Gradual increases in speed and/or
in control subjects and 1 day (20 + 5 hours) and 2 days (46 2 7 elevation were administered throughout the training program to
hours) after the last exercise session in the training groups. accommodate conditioning.
Plasma total cholesterol and triglyceride concentrations were Control grollp. Control subjects completed the identical proce-
measured directly.‘” HDL-C level was measured after precipitation dures for testing at approximately the same two time periods as
of very-low-density lipoprotein (VLDL) and LDL by dextran both of the training groups. but did not participate in any regular
sulfate.‘” Since all subjects had triglyceride levels below 400 mg exercise for the entire 20.week period.
dL-I. LDL-C level was calculated using the formula LDL-C = total
cholesterol - (HDL-C + triglyceride/5).‘” All analyses were per-
formed in duplicate as previously described.15 StatisticalAnalysis
Age, height,, pretraining total body mass, percent body fat,
Plasma Glucose and Insulin Determinations
fat-free mass, VOzmax, and all blood variables were analyzed by
Glucose and insulin concentrations were measured in the fasting one-way ANOVA to determine whether differences existed be-
blood sample and in blood samples drawn 30, 60, 90, and 120 tween the groups before training.
minutes after ingestion of 40 g glucose/m* body surface area.r5 The chi-square test was used initially to determine whether there
Plasma glucose levels were measured by the glucose oxidase was an equal distribution of the three major risk factors among the
method16 with a glucose analyzer (Beckman Instruments, Fuller- groups. Within-group differences were analyzed collectively by
ton, CA). and insulin levels were determined by radioimmunoas- MANOVA. When MANOVA revealed significant differences.
say.” After training, OGTTs were performed again in the morning univariate ANOVA was performed with planned comparisons to
after a 12. to 14-hour fast, approximately 20 hours after the last determine the effects of each training modality. Significant changes
training session. from before to after training were compared with changes from
Areas under the OGTT curves for glucose and insulin were initial to final in the control group using the interactions derived
computed by the trapezoidal method using both fasting concentra- from ANOVA. The mean -C within-subject SEM are presented on
tions (incremental area) and zero (total areas) as the baseline. all figures to simplify interpretation of significant differences. All
EXERCISE TRAINING MODIFIES RISK FACTORS FOR CHD 179

data presented in tables are expressed as the mean f SD and the (46 2 7 hours) after the last training session. Only three
5% level of significance is used. subjects from each training group showed an improvement
in their lipid profiles as evidenced by a decrease in the ratio
RESULTS
of total cholesterol to HDL-C by more than 0.5 mg dL-i.
Group Charactetistics There was no consistent relationship between changes in
There were no significant differences among the groups VOZmax. weight, percent fat, or fat-free mass and changes
for age, height, total body mass, percent body fat, fat-free in lipid profiles in either training group. In addition, there
mass, VO:max, or any of the blood variables before was no consistent relationship between changes in lipopro-
training. Chi-square analysis demonstrated that the preva- tein and lipids and family history of heart disease.
lence of hyperlipidemia, impaired glucose tolerance, and
hypertension was the same among the groups. Blood Pressure
Percent, body fat decreased with AT (P < .05), but not
There were no significant changes in resting blood
with ST. VO:max increased by 19% with AT (P < .OOOl),
pressure in either of the training programs (Table 4);
but did not change with ST. There were no significant
however, a definite trend toward a reduced mean arterial
changes in total body mass or fat-free mass in either the AT
blood pressure response was observed among subjects with
or ST groups. Body composition, total body mass, and
mild or moderate hypertension as a result of each training
VOZmax also did not change in the control subjects (Table
program. For example, there was an average decrease of 5
1). mm Hg in diastolic blood pressure (93 ? 8 I’ 88 ? 5) in the
The ST program resulted in a 50% increase in upper-
AT group (n = 4) and a 6-mm Hg decrease in systolic blood
body strength (8 + 2 v 12 2 3 weight plates) and a 36%
pressure (152 + 17 v 146 ? 14) in the ST group (n = 4)
increase in lower-body strength when values for the I-RM
among those subjects with hypertension. Nevertheless,
test were averaged for upper- and lower-body exercises,
there were too few subjects in this category to make any
respectively. There were no significant changes in the I-RM
meaningful conclusions about the effects of training on
values for the control group.
blood pressure.
Dietag) Composition
Analysis of food records indicated that all subjects Glrlcose and Insulin Responses
abstained from alcohol and ingested at least 2.50 g carbohy- The AT program resulted in a lower plasma glucose
drates/d for 3 days immediately preceding each OGIT. concentration at 90 (P < .05) and 120 minutes (P < .Ol)
There were no significant differences in either group after glucose ingestion (Fig 1A). whereas the ST program
between pretraining and posttraining food records with resulted in decreased glucose concentrations at 60, 90, and
regard to total kilocalories consumed or percent of calories 120 minutes (P < .05; Fig 1B). The 90- and 120-minute
derived from carbohydrates, proteins, or fats. In addition, reductions in AT and ST groups were both significantly
there were no differences in the amount of cholesterol, different from changes in the control group. Although there
fiber, or fats derived from monunsaturated, polyunsatu- was more than a 20% reduction from both training pro-
rated, or saturated fatty acids between the two recording grams in total glucose incremental area (above basal) under
periods (Table 2). the curve, these changes did not reach statistical signifi-
cance in either the AT (P = .06) or ST (P = .09) groups
Lipoprotein and Lipid Profiles (Fig 2A). However, there was a significant reduction in total
The concentrations of triglycerides, total cholesterol, glucose area under the curve in both groups (P < .05) when
HDL-C, and LDL-C did not change with either training zero was used as the baseline (Fig 2B). These reductions
program (Table 3). These results were no different whether were also significantly different from changes in the control
values were measured 1 day (20 * 5 hours) or 2 days group. There were no significant differences between the

Table 1. Characteristics of Training and Control Groups

AT (N = 1311 ST (N = 14)§ Control Group (N = 10)

Before Training After Training Before Training After Tralnrng Initial Final

Age W) 51 + a 4a+ 12 50 k a
Height (cm) 172 t 5 175 * 7 174 I! 6
Body mass (kg) 83.3 * 12.8 82.8 2 13.6 89.4 i il.9 90.7 2 12.2 89.3 t 14.8 90.2 + 14.2
% Body fat 27.6 ? 6.2 26.2 2 6.0* 26.0 ? 4.8 26.1 + 4.4 26.6 ? 4.9 26.7 ? 3.4
Fat-free mass (kg) 59.8 + 9.1 60.7 2 a.8 66.1 2 a.2 66.2 ? 9.1 66.2 + 6.9 66.6 ? 9.2
i/o,max (mL kg--’ min-‘) 31.0 2 6.8 36.5 ? 6.ltS 33.0 5 7.8 34.2 t 6.6 28.3 -t 5.9 29.5 r 6.3

NOTE. Values are means f SD.

*P < .05. significantly different from pretraining values.
tP i .OOOl, significantly different from pretraining values.
SP < .OOOl, change from before training to after training significantly different from initial change to final change in control group.
§N = 12 for % body fat, fat-free mass, and \io,max in AT; N = 13 for \io,max in ST.
180 SMUTOK ET AL

Table 2. Dietary Analyses of Training and Control Groups

AT(N = 13) ST (N = 14) ControlGroup (N = 10)

BeforeTraining After Training BeforeTraining After Training Initial Final

Total daily intake (kcal) 2,386 2 436 2,257 k 632 2,365 + 639 2,378 + 606 2,462 ? 648 2.448 ? 646
Carbohydrate (%) 52 k 12 53 k 10 46 + 8 47 + 9 50+_ 11 51 + 11
Protein (%) 18 + 4 17 5 4 19 + 3 19 r 6 17?3 18 t 3
Fat (%) 31 2 10 30-t 10 36 f 7 34 + 9 33 ? 8 32 + 9
Monounsaturated fats (g) 33-c 17 31 -+ 21 33+ 12 35 r 17 35* 14 34t 12
Polyunsaturated fats (g) 11 +6 1127 13 + 5 15 r 10 11 ?8 10 + 5
Saturated fats (g) 28+ 15 27 -c 16 35* 11 30 * 7 37 + 17 33+ 15
Cholesterol (mg) 323 5 182 277? 133 427 + 159 371 r 100 413 + 259 422 + 268
Fiber(g) 654 5*5 4*2 5r2 5+3 5+2

NOTE. Values are means 2 SD. There were no significant differences in dietary components between before training and after training within each
of the training groups or between initial and final within the control group

two training groups for any of the individual time points or DISCUSSION
for glucose areas under the curve. The results of this study demonstrate that ST has effects
Fasting insulin levels were significantly reduced by ST on risk factors for CHD similar to those of AT in middle-
(Fig 3B; P < .05), but not by AT (Fig 3A). Plasma insulin
aged men. Both ST and AT reduced glucose and insulin
concentration was significantly reduced at 90 and 120
responses to glucose ingestion during an OGTT, suggesting
minutes after glucose ingestion by both training programs
that both programs are equally effective for improving
(Fig 3A and B; P < .Ol for both time points and both
glucose tolerance and insulin response. Such changes in
training groups). These changes differed significantly from
regulation of glucose metabolism should protect against the
those of the control group. Peak insulin levels were attained
development of non-insulin-dependent diabetes mellitus
at 90 minutes before training and 60 minutes after training
(NIDDM)‘s and possibly atherosclerosis.‘”
in the AT group (Fig 3A), and 120 minutes before training
Previous investigations show that AT decreases both
and 60 minutes after training in the ST group (Fig 3B).
plasma glucose and insulin concentrations in response to
There was a significant reduction in the incremental insulin
0GTT.20 ST also reduces the insulin response to glucose
area under the curve for the AT group, but not for the ST
group (Fig 4A). These changes were not significantly ingestion, but glucose tolerance did not change in these
different in either training group from those of controls. studies4,*l,**; however, only normoglycemic subjects were
When using zero as baseline, a 24% reduction (P < .Ol) in studied. In the present study, all subjects either closely
the total insulin area under the curve resulted from the AT approached or met criteria for impaired glucose tolerance
program, and a 21% reduction (P < .05) resulted from the or NIDDM. In this regard, although none of the NIDDM
ST program (Fig 4B). Only the results from the AT patients changed their disease status, in three of four
program were significantly different from controls. There subjects in the ST group with impaired glucose tolerance,
were no differences between the two training groups for glucose tolerance normalized following training. To our
insulin responses at any specific time points or for total knowledge, this is the first report that demonstrates an
insulin areas under the curve (Fig 4A and B). In addition, improved glucose tolerance from ST with responses similar
there were no differences between initial and final glucose to those resulting from AT.
or insulin values at any specific time points or total areas There appears to be an empirical basis for the finding
under the curve in the control group. that ST is just as effective as AT in improving the regulation
Hematocrit levels were unchanged following both train- of glucose metabolism. Studies in isolated muscle indicate
ing modalities, suggesting that changes in glucose and that muscle contraction produces an insulin-like effect on
insulin levels were not influenced by training-induced glucose uptake, regardless of whether the contractions are
hemoconcentration or hemodilution. performed isometrically or isotonically.23 Although no infor-

Table 3. Plasma Lipid and Lipoprotein Levels in Training and Control Groups

AT IN = 13)’ ST (N = 14) ControlGroup IN = lo)*

BeforeTraining After Training BeforeTraining After Training Initial Final

Triglycerides 187 t 163 172 +- 153 168 2 110 154 2 114 199 2 91 181 ? 97
Total cholesterol 212 t 40 210 2 38 200 ? 26 201 ? 26 204 -e 33 202 * 32
LDL-C 135 2 29 138? 29 130 * 22 131 -+ 23 123 2 33 131 2 29
HDL-C 41 + 9 43 2 7 37 + 7 37 2 10 382 11 395 12

NOTE. Values are means ? SD in mg/dL. There were no significant differences in lipid levels between before training and after training within each
of the training groups or between initial and final within the control groups.
*N = 12 for LDL-C and HDL-C in AT; N = 9 for triglycerides in control groups. There were no significant differences between groups.
EXERCISE TRAINING MODIFIES RISK FACTORS FOR CHD 181

Table 4. Blood Pressure in Training and Control Groups A. INCREMENTAL GLUCOSE AREA UNDER OGll CURVE

AT IN = 13) ST (N = 14) (N = 10)

.g
Control Group
BefOre After Before After :: 8000
Training Training Training Training Initial Final
z

Systolic 133+8 132-+9 137~ 17 135k 15 12829 129-t8 s 8000

Diastolic 86 -t 6 84 -t 7 85 2 6 83 + 7 84 + 7 84 2 9
w
Mean 9 4caO
arterial* 102-t5 10023 102k9 100+10 99k7 99+9
2
NOTE. Values are means -+ SD in mm Hg. s 2000
*Mean arterial blood pressure = diastolic + (systolic ~ diastolicji3;
none of the values are significantly different.

mation was obtained in the present study to explain the TFlAlNlNG GROUP
mechanism by which either AT or ST reduced glucose or
insulin responses after glucose ingestion, other investiga-
tors suggest that AT programs might reduce insulin secre-
tiorP by increasing the number of glucose transporters,25.zh B. TOTAL GLUCOSE AREA UNDER OGTT CURVE
specifically the GLUT 4 transporter.?’
It has been suggested that improvements in glucose
regulation from exercise training are secondary to reduc-

A. AEROBIC TRAINING
c
a 200 -
a,
g 180-

%
8 160-

3
” 140-
TRAlMNGGRO&
I
3 120 - Fig 2. Incremental (above basal) and total (above zero) plasma
glucose response to glucose ingestion before (0) and after (W)
a loo- training; values are means f SE. lf < .05 for differences between
values obtained before and after training; tP < .05 for change from
I 1 I I 1 4
before training to after training compared with change in controls.
fasting 30 60 90 120
I_ Minutes after -I
glucose ingestion
tions in body fat.‘s The findings from the ST program refute
this conclusion, since percent body fat did not change. An
increase in muscle mass is postulated as one mechanism by
B. STRENGTH TRAINING
which ST might improve glucose regulation2’Jz; however,
there were no changes in fat-free mass determined by
g 200 - hydrodensitometry in the present study. Nevertheless. this
‘a, does not necessarily mean that muscle mass did not change,
E 180 - because we have recently observed increases in regional
z muscle mass with ST despite changes in total fat-free mass
s 160 -
measured by hydrodensitometry.2y
3
W 140 - The possibility that improvements in glucose tolerance
and reductions in insulin responses in the present study
$
120 - were due to the short-term effects of the last exercise
session cannot be excluded. The residual effects of the last
exercise session may play a role in the increase in insulin
1 I I I I I sensitivity observed following training? Short-term exer-
fasting 30 60 90 120 cise can increase both glucose disposal” and insulin respons-
I Minutes after 1 es.32 However, the elapsed time between the last exercise
glucose ingestion
session and the OGTT was the same for all subjects in the
present study.
Fig 1. Plasma glucose concentration before (fasting) and 30,60,90,
and 120 minutes after glucose ingestion, before (0) and after (W) AT
The finding that neither training modality affected lipo-
(A) and ST (8); values are means f SE. lfJ < .05, l*P < .Ol for protein and lipid profiles was unexpected. Although several
differences between before- and after-training values, investigators report no improvements in lipid levels with
182 SMUTOK ET AL

A. AEROBIC TRAINING A. INCREMENTAL INSULIN AREA UNDER OGlT CURVE

100 - 12000
P
^
E 10000
3 80- ;
2 z
E 8000
z 53
80 -
zi z 8000
Y 5
3 40- 4 4000

$ 20- f 2000

0
I I I I I AEROBK: .5mmGTH mama_
0
fasting 30 80 90 120 TRAINING GROUP
I- Minutes after 1
glucose ingestion

B. TOTAL INSULIN AREA UNDER OGlT CURVE

B. STRENGTH TRAINING
14000
P
k 12000

H 10000
E
3
2 8000
z
6WO
s
P 4000
s
!j 2000
0
AERoelc

fasting 30
)_
Fig 3. Plasma insulin concentration before (fasting) and 30,6O,gO,
and 120 minutes after glucose ingestion, before (0) and after (m) AT
(A) and ST (6); values are means f SE. lP c .05, **P < .Ol for
differences between before- and after-training values.
either AT33,34 or ST programs,35 others studies do show
significant changes in lipid profiles with both training
modalities.36 An insufficient training intensity or duration3’
the initial VOzmax being too high,38 HDL-C level being too
10w,~~no weight 10ss,~Oor the absence of a decrease in body
fat are some of the most common reasons for the lack of
change in lipid profiles. After careful examination for the we do not have an explanation
effects of each of these factors, it does not appear that any
of them could explain the discrepancy between the findings
in the present study and those of studies that found
improved lipid profiles after either AT or ST. Although the
ST group did have relatively low initial HDL-C levels and
neither group changed their weight with training, other
studies have reported improvements in lipid profiles with
similar initial HDL-C levels4 and without weight 10~s.~~~‘~~~ CHD, and there were also alterations
Furthermore, there was a significant decrease in body fat in
the AT group, a factor that appears to be important to the
improvement in lipid profiles with AT.43 It is conceivable
that the initial level of adiposity was a factor inhibiting a
TRAINING GROUP
Fig 4. incremental (above basal) and total (above zero) plasma
insulin response to glucose ingestion before (0) and after(U) training;
values are means 1?:SE. lP c .05, **P < .Ol for differences between
values obtained before and after training; tP c .05 for change from
before training to after training compared with change in controls.
significant improvement in lipid profiles. In this regard,
Coon et ali5 also reported no change in lipoprotein and
lipid profiles in older obese men after AT. Most of the
subjects in our study were obese; however, after examina-
tion of the results in responders (n = 3 for each group)
versus nonresponders, we were unable to identify any
factors that could explain the lack of improvement in
lipoprotein and lipid profiles with either AT or ST. Thus,
for the conflicting results
between the present study and those that have found
favorable alterations in lipid profiles with either AT or ST.
However, in some of the previous studies that reported
improvements in lipid profiles with ST, there was a lack of
control for normal variations of lipid and lipoprotein levels
by not having a control group, drawing only one blood
sample, and using subjects who had no risk factors for
in body weight or
composition.44 The results of the present study show that
when these limitations are considered, neither ST nor AT
improves lipoprotein and lipid profiles in middle-aged men.
The finding of no significant changes in resting blood
EXERCISE TRAINING MODIFIES RISK FACTORS FOR CHD 183

pressure with either training modality also was unexpected. In summary, 20 weeks of ST had the same effects as 20
Several investigators have found significant reductions in weeks of AT (walking/jogging) on risk factors for CHD.
blood pressure with either AT45 or ST.4,46 In the present Although both training modalities reduced glucose and
study, there was a trend toward a reduction in diastolic insulin responses to an OGTT, neither program altered
blood pressure with AT and a decrease in systolic blood lipoprotein and lipid profiles or resting blood pressure.
pressure with ST in the hypertensive men. It is conceivable
that the lack of effect from either training modality in
normotensive subjects prevented a statistically significant ACKNOWLEDGMENT

decrease that may have occurred if all subjects were We thank Ellen Rogus, Marilyn Lumpkin. Elizabeth Bannon,
hypertensive. This possibility is currently being investigated and Howard Baldwin for their laboratory assistance, and Dorothy
in hypertensive subjects. O’Donnell for her assistance in typing the manuscript.

REFERENCES

I. Powell KE. Thompson PD. Caspersen CJ, et al: Physical and lipoprotein lipids in healthy older men: A cross-sectional and
activity and the incidence of coronary heart disease. Annu Rev longitudinal intervention study. Metabolism 38:1201- 1209, 1989
Public Health 8:253-287, 1987 16. Alpert NL: Instrument series: Report no. 14, glucose ana-
7. Blair SN, Kohl HW, Paffenbarger RS, et al: Physical fitness lyzer and BUN analyzer. Lab World 24:40,1973
and all-cause mortality: A prospective study of healthy men and 17. Herbert V, Law KS, Gottlieb GW, et al: Coated charcoal
women. JAMA 262:2395-2401.1989 immunoassay of insulin. J Clin Endocrinol25:1375-1384, 1965
3. Lampman RM. Santinga PJ, Savage PJ, et al: Effect of 18. Lillioja S, Mott DM, Howard BV, et al: Impaired glucose
exercise training on glucose tolerance, in vivo insulin sensitivity, tolerance as a disorder of insulin action. N Engl J Med 318:1217-
lipid and lipoprotein concentrations in middle-aged men with mild 1225, 1988
hypertriglyceridemia. Metabolism 34:205-211, 1985 19. Stout RW: Overview of the association between insulin and
4. Hurley BF. Hagberg JM, Goldberg AP, et al: Resistive atherosclerosis. Metabolism 34:7-12, 1985 (suppl)
training can reduce coronary risk factors without altering VOzmax 20. Reitman JS, Vasquez B. Klimes L, et al: Improvement of
or percent body fat. Med Sci Sports Exert 20:150-154,1988 glucose homeostasis after exercise training in NIDDM. Diabetes
5. Hurley BF, Seals DR, Ehsani AA, et al: Effects of high Care 7:434-441, 1984
intensity strength training on cardiovascular function. Med Sci 21. Miller WJ. Sherman WM, Ivy JL: Effect of strength training
Sports Exert 16:483-48X. 1984 on glucose tolerance and post-glucose insulin response. Med Sci
Sports Exert 16:539-543, 1984
6. Palatini P, Mos L, Munari L, et al: Blood pressure changes
during heavy-resistance exercise. J Hypertens 7:S72-S73, 1989 22. Craig BW. Everhart J, Brown R: The influence of high-
resistance training on glucose tolerance in young and elderly
(suppl 6)
subjects. Mech Ageing Dev 49:147-157. 1984
7. Bruce RA. Hornstein TR: Exercise stress testing in evalua-
23. Holloszy JO, Narahara HT: Studies of tissue permeability.
tion of patients with ischemic heart disease. Prog Cardiovasc Dis
X. Changes in permeability to 3-methylglucose associated with
ll:371-390, 1969
contraction of isolated frog muscle. J Biol Chem 240:3493-3500,
H. National Diabetes Data Group: Classification and diagnosis
1965
of diabetes mellitus and other categories of glucose intolerance.
24. Kahn SE, Larson VG, Beard JC, et al: Effect of exercise on
Diabetes 2X:1039-1057. 1979
insulin action, glucose tolerance, and insulin secretion in aging. Am
9. Taylor HL, Buskirk E. Henschel A: Maximal oxygen intake as
J Physiol258:E937-E943. 1990
an objective measure of cardiorespiratory performance. J Appl
25. Hirshman MF, Wallberg-Henriksson t-I. Wardzala LJ, et al:
Physiol8:73-80, 1955
Acute exercise increases the number of plasma membrane glucose
10. Brozek JF. Grande F. Anderson J, et al: Densitometric transporters in rat skeletal muscle. FEBS Lett 238:235-239. 1988
analysis of body composition: Revision of some quantitative assump-
26. Hirshman MF, Wardzala LJ, Goodyear LJ. et al: Exercise
tions. Ann NY Acad Sci 110:113-140.1963
training increases the number of glucose transporters in rat
I 1. Jalowski AA, Dawson A: Measurement of lung volume: The adipose cells. Am J Physiol257:E520-E530. 1989
multiple breath nitrogen method, in Clausen JL (ed): Pulmonary 27. Douen AG, Ramlal T, Rastogi S, et al: Exercise induces
Function Testing Guidelines and Controversies. New York, NY, recruitment of the “insulin-responsive” glucose transporter. J Biol
Academic. 1982, pp 116-127 Chem 265:13427-13430,199O
12. Adams CF: Nutritive value of American foods in common 28. National Institutes of Health: Diet and exercise in noninsulin-
units, in Agriculture Handbook 456. Washington, DC, USDA, dependent diabetes mellitus. Nat Inst Health Consensus Dev Conf
1975 Consensus Statement 6:1-7. 1986
1.7. Lipid Research Clinics Program: Manual of Laboratory 29. Hurley BF, Redmond BA. Koffler KH. et al: Assessment of
Operations. Lipid and Lipoprotein Analysis. Bethesda, MD, Na- strength training effects on leg composition in older men using
tional Heart and Lung Institute, US Department of Health, magnetic resonance imaging. Med Sci Sports Exert 23:SlOB. 1991
Education. and Welfare publication NIH, 1974, pp 75-628 30. Heath GW, Gavin JR. Hinderliter JM. et al: Effects of
14. Warnick R. Benderson JM, Albers JJ: Quantification of high exercise and lack of exercise on glucose tolerance and insulin
density lipoprotein subclasses after separation by dextran sulfate sensitivity. J Appl Physiol55:512-517. 1983
and Mg?’ 714 precipitation. Clin Chem 28:1379-1388, 1982 31. Bourey RE, Coggan AR. Kohrt WM. et al: Effect of exercise
15. Coon PJ, Bleecker ER, Drinkwater DT. et al: Effects of body on glucose disposal: Response to a maximal insulin stimulus. J
composition and exercise capacity on glucose tolerance. insulin, Appl Physiol69:1689-1694, 1990
184 SMUTOK ET AL

32. Devlin JT, Hirshman M, Horton ED, et al: Enhanced serum lipids in young men with low high density lipoprotein
peripheral and splanchnic insulin sensitivity in NIDDM men after cholesterol. Arteriosclerosis 8:245-251, 1988
single bout of exercise. Diabetes 36:434-439, 1987 40. Williams PT, Krauss RM, Vranizan KM, et al: Changes in
33. Lipson LC, Bonow RO, Schaefer EJ, et al: Effect of exercise lipoprotein subfractions during diet-induced and exercise-induced
conditioning on plasma high density lipoproteins and other lipopro- weight loss in moderately overweight men. Circulation 81:1293-
teins. Atherosclerosis 37:529-538, 1980 1304,199o
34. Allison TG, Iammarino RM, Metz KF, et al: Failure of 41. Kiens B, Jorgensen I, Lewis S, et al: Increased plasma
exercise to increase high density lipoprotein cholesterol. J Cardiol HDL-cholesterol and apo A-l in sedentary middle-aged men after
Rehab 1:257-265, 1981 physical conditioning. Eur J Clin Invest 10:203-209.1980
35. Kokkinos PF, Hurley BF, Smutok MA, et al: Strength 42. Thompson PD, Cullinane EM, Sady SP, et al: Modest
training does not improve lipoprotein-lipid profiles in men at risk changes in high density lipoprotein concentration and metabolism
for CHD. Med Sci Sports Exert 23:1134-1139, 1991 with prolonged exercise training. Circulation 78:25-34, 1988
36. Blessing D, Stone M, Byrd R, et al: Blood lipid and 43. Nikkila EA, Taskinen MR, Rehunen S, et al: Lipoprotein
hormonal changes from jogging and weight training of middle-aged lipase activity in adipose tissue and skeletal muscle of runners.
men. J Appl Sports Sci Res 1:25-29, 1987 Relation to serum lipoproteins. Metabolism 27:1661-1671,1978
37. Wood PD, Haskell WL, Blair SN, et al: Increased exercise 44. Hurley BF: Effects of resistive training on lipoprotein-lipid
level and plasma lipoprotein concentrations: A one-year random- profiles: A comparison to aerobic exercise training. Med Sci Sports
ized controlled study in sedentary middle-aged men. Metabolism Exert 21:689-693, 1989
32:31-39. 1983 45. Martin JE, Dubbert PM, Cushman WC: Controlled trial of
38. Tran ZV, Weltman A, Glazes G. et al: The effects of exercise aerobic exercise in hypertension. Circulation 81:1560-1567, 1990
on blood lipids and lipoproteins: A meta analysis of studies. Med 46. Harris KA. Holly RG: Physiological response to circuit
Sci Sports Exert 15:393-402, 1983 weight training in borderline hypertensive subjects. Med Sci Sports
39. Raz L, Rosenblit H, Kark JD: Effect of moderate exercise on Exert 19:246-252, 1987