Physiology and Biochemistry Plasma Lipid Levels in Active and Sedentary Premenopausal Females A. C. Perry, M H. Shaw, L. Hsia, M. S. Nash, T Kaplan, J. F. Signorile, B. Appleyate Human Performance Laboratory, School of Education, University of Miami, Coral Gables, FL 33124 (U. S. A.)

A. C. Pei-ry, M H. Shaw, L. Hsia, M S. Nash, T Kaplan and .J. F Signorile, Plasma Lipid Levels in Active and Sedentary Premenopausal Females. Tnt J Sports Med, Vol l3,No3,pp2lO—2l5, 1992.

transported in the blood may be a more significant factor than the total amount of cholesterol circulating in the bloodstream. Epidemiological studies have shown that individuals having higher levels of high density lipoprotein cholesterol (J4DL-C)

have a lower incidence of coronary artery disease (CAD),

whereas individuals having high levels of low density cholesterol (LDL-C) are at greater risk for coronary artery disease (29).

Accepted: August 20, 1991

The majority of cross-sectional studies have Cross-sectional data on 19 long-distance runners (LD), 17 aerobic dancers (AD), 19 recreational joggers (RJ), and 15 inactive controls (IC) were examined for cardiovascular endurance and determination of plasma lipoproteins. Subjects included premenopausal eumenorrheic females who were non-smokers and presently not using oral contraceptives. Results indicated that all groups were similar in age, height and weight. Only the AD group had a signifi-

cantly lower percent body fat (p < .00 1) than the other groups. The LD, AD and RJ groups had a significantly higher VO2max than the IC group (p < .05), and the LD and AD groups had a significantly higher VO2max than the

RJ group (p < .05). Analysis of a one-day food log indicated that the only difference in diet among the groups was a significantly lower intake of total and monounsaturated fat in the AD group (p < .05) and a significantly lower intake of carbohydrates in the IC group than the AD group (p < .05). Analysis of plasma lipids revealed no significant differences in any lipid variables among the groups. These

findings indicate that healthy premenopausal eumenorrheic females with similar physical characteristics also have similar plasma lipid profiles regardless of their physical activity level.

been performed on active adult males and have shown a more favorable plasma lipid profile in comparison to sex and agematched inactive controls, i. e., increased HDL-C, decreased LDL-C, decreased TC, and decreased triglycerides (TG) (17, 33). Longitudinal studies of exercising males have attempted to correct for changes in body weight and dietary habits following training and have also shown a more favorable lipid

profile following training (5, 33). It has been reported that even when the data are statistically controlled, i. e., weight and

diet, a significant increase in the cardioprotective HDL-C levels is shown with exercise (11).

Research on adult females is limited and more controversial. Longitudinal studies of sedentary females undergoing an exercise program have shown no change in HDLC (5, 25), no change (5, 25) or a significant decrease (1) in LDL-

C, no change (25) or a significant decrease (5) in TC, and no change (5, 25) or a significant decrease (1) in TG levels. Crosssectional studies performed on female runners, however, have shown higher levels of HDL-C, lower LDL-C, and lower TC than inactive age- and sex-matched controls (24, 32). Most of the cross-sectional studies have been done on female distance runners and/or joggers who were lower in body weight and fat than their sedentary controls. These studies also included both pre- and postmenopausal females which may add confounding factors to the research design.

Key words

One distinct problem in assessing plasma Premenopausal, eumenorrheic, lipids, activity

lipids in a female population is the effect of endocrine changes that occur throughout the menstrual cycle. It has been shown that progestins are associated with lower HDL-C levels. Studies of oral contraceptive users have shown that women taking progestins with antiestrogenic activity have significantly lower

Introduction

HDL-C levels than women taking progestins having no antiestrogenic effect or women not taking progestins at all (4).

Total cholesterol (TC) is a primary risk factor in the development of coronary artery disease (27). Past research has indicated that the manner in which cholesterol is lnt.J.SportsMed. 13(1992)210—215 Georgmieme Verlag Stuttgart New York

Plasma lipids observed during the menstrual cycle have shown a significant decrease in TC during the luteal phase (14), a sig-

nificant decrease in the LDL-C during the luteal phase (14), and a significant increase in the HDL2 subfraction during midcycle (2). Therefore, it would be prudent in any lipid study

Downloaded by: National University of Singapore. Copyrighted material.

Abstract

mt. J. Sports Med. 13(1992) 211

Diet and Plasma Lipids

using premenopausal females to take blood samples at a uniform time during the menstrual cycle.

were given oral instructions on recording food intake and given food models to determine serving size. Food intake was

analyzed using a computerized data base (DINE System)

Methods

which contained 5600 food items. When nutrient values for a particular food could not be found, a similar food was substituted.

Following a 12-hour fast, blood was drawn from the antecubital vein of each subject at least 12 to 24 hours post exercise. All blood samples were taken between days five and seven of the menstrual cycle when estrogen levels were relatively low and stable.

A total of 70 premenopausal eumenorrheic

Serum lipids were measured according to a

subjects from South Florida volunteered to participate in the study. Subjects included 19 long-distance runners (LD), 19 recreational joggers (RJ), 1 7 aerobic dance instructors (AD), and 15 inactive controls (IC). All subjects were reportedly healthy following examination by their personal physician.

standardized protocol reported by the Lipid Research Clinics (16). Assays for this study were conducted by the Center for Disease Control (CDC) laboratory in Miami, Florida, which also assured calibration. Standards used for calibration were also developed by the CDC laboratory. Instruments utilized to

Exclusion criteria from this study included females who

perform the lipid assays at the CDC laboratory were con-

smoked, females taking oral contraceptives and/or receiving hormone replacement therapy which could influence their blood lipid levels and females taking medication.

stantly monitored for standardization in accordance with the guidelines set forth by the Lipid Research Clinics (16). The standard procedure for the CDC laboratory was to perform all assays in duplicate form and record the mean values. If there was a difference of more than two percent between any two samples, a third sample was performed.

In accordance with other studies using female athletes (24, 32), the LD group averaged at least 43.3 km (26

mi)/wk (mean =66 km (40 mi)/wk) and the RJ group averaged at least 10 km (6 mi)/wk (mean 26.6 km (16 mi)/wk). Aerobic dance teachers taught at least six aerobic dance classes per week including a minimum of 25 minutes of continuous aerobic dance routines (mean 9.4 hr/wk). All active participants had been either running or teaching aerobic dance at least two years prior to the study. The inactive control subjects exercised less than one time per week and not on a regular basis.

At least 24 hours post training all subjects underwent a graded exercise test (GXT) to determine maximum oxygen consumption (VO2max) using the SensorMedics System V metabolic cart. The stationary bicycle ergometer was chosen for the GXT to prevent the LD and RJ groups from having a training-specific advantage over the AD group. The

GXT used the McArdle Protocol (19), which increases the workload in increments of 30 watts every two minutes until one or more of the following criteria are met: (1) volitional ex-

The TC was measured spectrophotometrically after several enzymatic steps and read at a wavelength of 500

nanometers (nm). The cholesterol content of HDL-C was measured in the supernatant fraction of plasma after the removal of LDL-C and very low density lipoprotein cholesterol

(VLDL-C) by precipitation with heparin and manganese chloride (16). Another aliquot was separated by ultracentrifugation at a density of 1.006 grams per milliliter, with the bot-

tom fraction containing both HDL-C and LDL-C (16). The LDL-C was calculated by subtracting the HDL-C value from the TC in the bottom fraction (16). The VLDL-C was obtained by subtracting the sum of the LDL-C and HDL-C from the IC. The levels of HDL2 and HDL3 were determined according to

the procedures outlined by Gidez and co-workers (8) using dextran sulfate to precipitate HDL2 selectively. The TG levels were determined enzymatically following treatment with a lipase and the subsequent measure of glycerol release (20).

haustion, (2) a plateau in VO2max, and (3) a respiratory exchange ratio above unity.

A one-way analysis of variance was used to test for significant differences among the four groups. A Tukeys Multiple Comparison Test was used to compare means when a

Body fat was measured according to the procedures set forth by Durnin and Womersley (6). A Lange skinfold cliper was used to obtain the sum of four skinfold measurements: biceps, triceps, subscapular, and suprailiac. The reliability of the investigator using this formula was .94, which was higher than the reliability obtained using other formulas by Jackson and Pollack (12) and Sloan and Weir (30). Therefore, the formula by Durnin and Womersley (6) was selected.

significant F value was obtained. Correlation coefficients were calculated between selected physical characteristics and

serum lipids. The 0.05 level of probability was accepted as statistically significant. Results

All subjects completed a one-day food diary

Physical characteristics of all subjects are presented in Table 1. There were no significant differences among the four groups in age, height, or weight. The AD group, how-

that was representative of their typical diet. Since the majority of subjects in this study refused to complete a two- or three-day food log, the investigators felt compelled to secure at least a one-day food log to obtain a general idea of total calories and

ever, showed a significantly lower percent body fat than the LD, RJ, and IC groups (p < .001). As shown in Table 1 the highest VO2max was observed in the LD group followed by the AD, RJ and IC groups respectively. The IC group had a signifi-

nutrients consumed by subjects in the four groups. Subjects

cantly lower VO2max than the other three groups (p < .05)

Downloaded by: National University of Singapore. Copyrighted material.

The purpose of this study was to (1) compare the levels of cardiovascular fitness in premenopausal females engaging in long-distance running, recreational jogging and aerobic dance, (2) compare the plasma lipid profile of active and inactive females at a uniform time in the menstrual cycle, and (3) compare nutrient intake among active and inactive premenopausal females.

mt. J. Sports Med. 13 (1992)

A. C. Perry, M. H. Shaw, L. Hsia, M. S. .Nash, T. Kaplan, J. F. Signorile, B. Appleyate

Table 1 Physical characteristics of female long-distance runners, aerobic dance instructors, recreational joggers and inactive controls (means SD). Physical Characteristics

Long-Distance Runners (LD)

Aerobic Dance Instructors (AD)

Recreational Joggers (RJ)

Inactive Controls (C)

Age (yrs)

37.2±9.0

32.2±6.0

32.6±8.3

33.3±7.3

Height (in)

64.3±2.7

63.9±2.7

66.1±2.3

64.7±3.0

Weight (kg)

57.8±7.3

54.4±5.1

57.6± 12.0

60.1

Bodyfat (%)

26.8±4.8

21.8±4.0***

28.1

30.8±5.7

VO2max

49.0±7.2

48.0±7.2

42.1

31 .3±6.1*

)mlskg1min) *LD AD, RJ< IC (p RJ (p < .05) * * Significantly different than the other three groups (p < .001).

Table2 Plasma lipid and lipoprotein characteristics among the groups (means±SD). Long-Distance Runners

Aerobic Dance Instructors

(n=19)

(n17)

182.2±30.9

HDL-C

Recreational Joggers

Controls

(n=19)

(n=15)

179.8±23.6

199.9±58.7

187.0±64.8

73.8± 13.3

74.8± 13.6

72.3± 14.6

73.5±12.8

LDL-C

92.7±25.2

81.3±21.5

107.3±43.2

97.5±53.6

VLDL-C

16.0±9.0

23.7±23.4

21.4±8.5

16.1 10.0

Triglycerides

65.3±21.6

57.2± 14.6

65.3±24.9

58.7± 17.6

HDL2

9.2± 10.3

11.2±8.4

9.8±5.2

HDL3

64.5±14.1

65.5±11.0

62.4±13.3

64.5±13.8

HDL/LDL

0.9±0.3

1.0±0.5

0.8±0.3

0.9±0.3

TC/HDL

2.5±0.4

2.4±0.4

2.8±0.8

2.6±0.8

Lipid Variables

Total Cholesterol

Inactive

9.1

ANOVA yielded no significant difference among groups in any lipid variables.

and the RJ group had a lower mean VO2max than either the LDC or AD groups (p < .05). Plasma lipid and lipoprotein values of the four groups are presented in Table 2. There were no significant dif-

ferences in any lipid variables among the four groups. The standard deviation in the IC and RJ groups were quite large for

certain lipid variables. This was attributed to an abnormally elevated TC and LDL-C level of 415 mg/dl and 282 mg/dl in one IC subject and 404 mg/dl and 205 mg/dl respectively in one RJ subject. When the analysis was done without these data, there was no difference in the statistical results. In the LD group the standard deviation was larger than the mean HDL2 due to the fact that two subjects had an extremely high HDL2 of 46 mg/dl and 38 mg/dl respectively. There were no signifi-

cant differences in the cardioprotective HDL2 subfraction in any of the groups. All four groups had total cholesterol levels

under 200 mg/dl and a TC/HDL cardiac risk ratio below three, which is below the average value associated with cardiac risk.

Analysis of dietary intake is shown in Table 3. The majority of subjects refused to complete a two- or threeday food log and some subjects were unwilling to complete a one-day food log. It is unclear why subjects were so reluctant to complete a detailed food log. Reasons cited included insufficient time, pressing schedules and too time-consuming to complete. A total of 16 LD, 17 AD, 16 RJ and 14 IC subjects or 63 out of 70 subjects returned a 24-hour food log. According to

the one-day food log, there were no significant differences among the groups in any nutrient category except for total and monounsaturated fat and carbohydrate intake. The AD group had a significantly lower intake of total fat (p < .05) and mon-

ounsaturated fat (p < .05) than the other groups, and the IC group had a significantly lower intake of complex carbohydrates than the AD group (p < .05). The IC group had a consistently higher intake of cholesterol, total fat and saturated fat

and a consistently lower intake of fiber although the above mentioned values were not statistically different than the other

groups. Interestingly, the LD group, which did the highest volume of aerobic work, also showed the lowest total caloric intake.

Downloaded by: National University of Singapore. Copyrighted material.

212

Diet and Plasma Lipids

mt. J. Sports Med. 13(1992) 213

______

Table 3 Nutrient intake among groups (means SD. LongDistance Runners

(n=16)

Recreational

Joggers (n=16) 1730.0 545.0

1690.0

29.4

28.9

Absolute Energy lntake

1600.0

(Kcals/day)

487.0

1700.0 536.0

23.3

31.5

Relative Energy Intake

Inactive Controls

Dance

Instructors (n=17)

(n=14) 342.0

10.0

(Kcals/kg bw/day) 15.4

16.2

15.1

15.4

1.0 .4

1.2 .6

1.1

1.1

Carbohydrate Intake (% Keats/day)

48.2 10.5

52.9

50.4

10.6

12.3

Fat Intake (% Kcals/day)

32.7

27.6**

31.1 12.1

37.6

10.2

Saturated Fat

12.2

11.6

13.4

14.4

7.9

5.5

6.1

7.9

Protein Intake (% Kcals/day) Protein Intake (gm/kg bw/day)

.4

.3

41.4*

(%totalfat) Polyunsaturated Fat

(%totalfat) 12.6

10.5"

11.6

15.1

(% total fat)

4.0

4.5

4.5

4.5

Alcohol

2.8

1,1

1.8

4.7

17.4

15.4

17.0

11.6+

224.0 139.0

273.0

216.0 136.0

Monounsaturated Fat

(D/o Keals/day)

Fiber (gm/day)

Cholesteroi (gm/day)

196.0

362.0 240.0

+ Rounded off to nearest whole number. Significantly different than the AD group (p < .05). • *Significantly different than all groups (p < .05).

Discussion

This is in contrast to previous research showing that percent body fat and distance run are the strongest predictors of HDL-

The present study is in agreement with past research showing a higher level of cardiovascular endurance, as indicated by VO2max, in aerobically trained individuals. It is

C, with percent body fat being inversely related to HDL-C (24, 28). Although the AD group had a significantly lower percent body fat (Table 1) than the other groups, their lipid levels were

important to note that the aerobic dance instructors had

comparable to the other subjects (Table 2). The IC group in this study had the highest percent body fat and a mean HDL-C

VO2max levels very similar to the long-distance runners when tested on a piece of equipment which did not provide a train-

ing- specific advantage, i. e., stationary bicycle ergometer. Milburn and Butts (22) concluded from their study that aerobic dancing can be very effective in improving cardiovascular endurance. The present study supports those findings. There was an inverse correlation of — .299 between VO2max and LDL-C and an inverse correlation of- .274

between VO2max and the TC/HDL risk ratio. This correlation, although weak, was significant (p < .05) and may be related to the sample size studied which was large in comparison to other lipid studies performed in females.

In the present study, there was no significant correlation between percent body fat and any lipid variable.

level of 73.5 mg/dl. This is greater than the norm but similar to active subjects in previous studies (24, 32) and similar to the ac tive groups in the present study. The LD group also had a 26.8 percent body fat, which is higher than the 21 percent body fat reported for the LD group in the Wood study (32) and the 22.5 percent body fat reported in the Moore study (24). The mean HDL-C for the LD group, however, was 73.8 mg/dl, which is

comparable to 75 mg/dl and 78 mgJdl respectively reported in the aforementioned studies. The RJ group had a 28.0 percent body fat and an HDL-C level of 72.3 mg/dI, which is also comparable to active groups in previous studies (24, 32). While Wood (32) used hydrostatic weighing to determine lean body mass and percent body fat, Moore (24) used the formula by Durnin and Womersley (6) to determine percent body fat, which was also used in the present study.

Downloaded by: National University of Singapore. Copyrighted material.

Nutrient

Aerobic

A. C. Perry, M. H. Shaw, L. Hsia, M S. Nash, T Kaplan, .J. F. Signorile, B. Appleyate

The average distance run by the LD group in

Hormonal fluctuations are also known to af-

this study was 66 km (40 mi)/wk, which is considerably higher than the weekly average of5l .6 km(3 1.0 mi) and 52.5 km(31 .5 mi) reported previously. It is also higher than the 25.8 km (16

fect plasma lipids (2, 14). Females have higher HDL-C levels than males of the same age range and this may be due, in part, to their higher levels of circulating estrogens (4). Since all subjects were eumenorrheic females who were not using oral contraceptives and sampled at a uniform time in their menstrual cycle, the variations in circulating estrogen and progesterone should have been accounted for. It would be necessary, however, to measure circulating levels of estrogens, particularly 17 Il-estradiol, in order to confirm similar hormone levels in all subjects.

mi)/wk reported by the RJ group in this study. The plasma lipid profile of the LD group, however, was similar to the active and inactive groups in this study and comparable to the lipid profiles of active groups in previous studies (24, 32).

Past research has indicated that diet may have an effect upon the plasma lipid profile. Modest increases in alcohol tend to increase HDL-C levels while increased levels of carbohydrates are associated with a decrease in HDL-C (7). It

is important to note the inherent limitations of interpreting a one-day food log. According to Basiotis and colleagues (3), day-to-day variability in food energy and nutrient intake limits the precision and accuracy of a one-day food log. In the Basiotis study (3) a three-day food log was found to be accurate for examining nutrient intake in a group of 16 females. Therefore, caution is recommended in interpreting the dietary intake data.

The only significant difference among groups reported for a one-day food log was a lower total fat and mono-

unsaturated fat intake in the AD group and a lower carbohydrate intake for the IC group than the AD group (Table 3). Reduction in total fat may be beneficial in lowering TC and LDLC thereby reducing coronary risk (9). Studies also show that diets containing higher amounts of monounsaturated fats may also reduce TC and LDL-C levels (18, 21). Although the AD group reported a lower intake of total fat which would tend to reduce TC and LDL-C, they also reported a lower intake of monounsaturated fat which would tend to have the opposite effect possibly negating any net changes in TC and LDL-C in this group. The AD group also revealed a significantly higher

intake of carbohydrates than the IC group, which had the lowest daily intake of carbohydrates (Table 3). Since carbohydrates are a preferred fuel source during exercise, it would be

expected that all active groups would have a significantly higher intake of carbohydrates. The RJ group, however, had the highest standard deviation and the LD group had a lower intake of carbohydrates than both the AD and RJ groups. Therefore, it is not surprising that the only significant difference in carbohydrate intake was between the AD and IC groups (Table 3).

Although total caloric intake was equivalent among the groups, LD runners who averaged 66 km (40

The present study showed that there were no significant differences in the plasma lipid profile of aerobically trained and untrained euinenorrheic females (Table 2). This is primarily due to the high HDL-C and low cardiac risk level of the IC group used in this study. One reason why our data may be different than previous studies is that the IC group was similar in age, height, weight and percent body fat to the active subjects (Table 1). Previous studies revealing a significant relationship between physical activity and plasma lipids have also shown significant differences in body composition among their subjects. These same studies have also shown a significant inverse relationship between HDL-C and percent body fat (24) and a positive relationship between HDL-C and lean body mass (28). Past research has confirmed a significant inverse relationship between percent body fat and HDL-C (10, 15, 31) and a positive relationship between percent body fat and LDL-C (3). In the present study, there were no significant differences in percent body fat among the active and inactive groups with the exception of the AD group, which had a lower percent body fat than all the groups; therefore, differences in percent body fat were not a confounding factor in the present study.

All subjects in this study were recruited from Miami, Florida, which is a year-round warm weather climate.

It is possible that control subjects in this area are outdoors more often and normally more active than control subjects in other regions. The VO2max of the IC group averaged 31.3 mis/kg/mm, which is typical of untrained females in this age range (13). If activity level in the IC group was slightly increased, it was not reflected in their VO2max levels. It may also

be possible that the IC group represented a more health conscious segment of the population since recruitment was organized from a university-based setting, and oral contraceptive users and smokers were excluded from the study. There is also

a possibility that hormonal changes that normally occur

mi)/wk had a similar mean weight and percent body fat to the other groups (Table 1). Theoretically at similar caloric intakes, the increased mileage should have produced a decrement in weight and body fat in the LD group. Recent research has suggested that individuals incurring a substantial caloric deficit may have a suppression of their metabolic rate although exer-

throughout the menstrual cycle may mask the beneficial effects of training upon plasma lipids. Therefore, if the plasma lipid profile of eumenorrheic premenopausal females reveals elevated HDL-C, reduced LDL-C, and reduced cardiac risk, it may be possible that the impact of exercise upon plasma lipids

cise of sufficient intensity and duration should reverse that process (23). This phenomenon has been observed more

that premenopausal eumenorrheic, non-obese females,

frequently in exercising females and may have occurred in the LD group in our study. It would be necessary to measure basal

metabolic rate in all groups in order to confirm this finding. This can be done using indirect calorimetry or the more recent doubly labelled water technique that relies upon two stable isotopes of water to measure daily energy expenditure in a free-living environment (26).

is substantially diminished. It is implied from the present study

trained or untrained, have similar lipid profiles which are associated with low levels of cardiac risk. These results are in contrast to previous cross-sectional research completed on females and in sharp contrast to those studies completed on active versus inactive males.

Downloaded by: National University of Singapore. Copyrighted material.

214 mt. J. Sports Med. 13(1992)

list. I Sports Med. 13 (1992) 215

Diet and Plasma Lipids

Ballantyne D., Clark A., Dyker G. S., Gilli C. R., Hawthorne V. M.,

Henry D. A., Hale D. S., Murdock R. M., Semple T., Stewart G. M.: Prescribing exercise for the healthy: Assessment of compliance and effects on plasma lipids. Health Bull 36 (4): 169—176, 1978.

2 Barclay M., Barclay R. K., Skipski V. P., Temebus-Kekish 0.,

17 Martin R. P., Haskell W. L., Wood P. D.: Blood chemistry and lipid profiles of elite distance runners. Ann NYAcad Sci 301: 346— 360, 977. 18 Mattson F. H., Grundy S. M.: Comparison of effects of dietary sat-

urated, monounsaturated and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. JLipidRes 5; 26: 194—202, 1989. 19 McArdle W. D., Katch F. I., Pechar G. S.: Comparison of continu-

Mueller C. H., Shah E., Ilkins W. L.: Fluctuations in human serum

ous and discontinuous treadmill and bicycle tests for VO2max.

lipoproteins during the normal menstrual cycle. Biochem J 96:

Med Sci SportsExerc5 (3): 157, Fall 1973. 20 McGowen M. W., Artiss J. D., Strandbergh J. R., Zak B.: A peroxidase-coupled method for the calorimetric determination of serum triglycerides. Clin Chem 29:538—542, 1983, 21 Mensink R. P., Katan M. B.: Effect of a diet enriched with monounsaturated or polyunsaturated fatty acids on levels of low density

205—209, 1965. Basiotis P. P., Welsh S., Cronin F. J., Kelsay J. L., Mertz W.: Num-

bers of days of food intake records required to estimate individual and group nutrient intakes with defined confidence. JNutr 117: 0, 1638—1641, 1987. Bradley D. D., Wingerd J., Pettitti D. B., Krause R. M., Rancharan S.: Serum high density lipoprotein cholesterol in women using oral contraceptives, estrogens and progestins. N Engi J Med 299: 17— 20, 1978. Brownell K. D., Bachorik P. S., Ayerle R. S.: Changes in plasma lipid and lipoprotein levels in men and women after a program of moderate exercise. Circulation 65: 577—484, 1982. 8 Durnin G. A., Womersley J. V.: Body fat assessed from total body density and its estimation from skinfold thickness: Measurement on 481 men and women aged 16—22 years. Br J Nutr 32: 77—97, 1974.

and high density lipoprotein cholesterol in healthy women and men. NEnglJMed32l: 7,436—441, 1989. 22 Milburn S., Butts N.: A comparison of the training responses to aerobic dance and jogging in college females. Med Sci Sports Exerc 15: 510—513, 1983. 23 Mole P. A.: Impact of energy intake and exercise on resting metabolic rate. Sports Med 10:2,72—87(1990).

24 Moore C. E., Hartung G. H., Mitchell R. E., Kappus C. M., Hinderletter J.: The relationship of exercise and diet on high density lipoprotein cholesterol levels in women. Metabolism 32: 189— 196, 1983.

Ernst N., Fisher M., Smith W., Gordon T., Rificind B. M., Little J. A., Mishkel M. A., Williams 0. D.: The association of plasma high

25 Perry A., Trapp J., Weeks L.: The effects of interval aerobic training in plasma lipid fractions of male and postmenopausal seden-

density lipoprotein cholesterol with dietary intake and alcohol consumption. The Lipid Research Clinics Program Prevalence

tary faculty. JSports Med Phys Fitness 26(2): 186—193, 1986. 26 Poehlman E. T., Melby L. L., Goran M. L.: The impact of exercise and diet restriction on daily energy expenditure. Sports Med 11(2): 78—101, 1991. 27 Ross R.: The pathogenesis of atherosclerosis — an update. NEngIJ Med 314: 488—500, 1986. 28 Rotkis T., Boyden T. W., Pamenter R. W., Stanforth P., Wilmore J. A.: High density lipoprotein cholesterol and body composition of female runners. Metabolism 30: 994—995, 1981. 29 Schonfeld G.: Lipoproteins in atherogenesis. Artery 5: 305—329,

Study in Circulation 62 (supplement IV): 41—52, 1980.

8 Gidez I., Miller G., Burnstein M.: Separation and quantification of subclasses of human plasma high density lipoproteins by simple precipitative procedure. JLipidRes 23: 1206—1223, 1982. Ginsberg H. H., Barr S. C., Gilbert B. A., Karmelly W., Deckelbaum R., Kaplan K., Ramukrishman R., Halleran S., Dell R. B.: Reduction of plasma cholesterol levels in normal men on an American Heart Association step 1 diet or a step 1 diet with added monounsaturated fat. NEnglJMed 322: 574—579, 1990. '° Gordon T., Castelli W. P., Hjorland M. C., Kannel W. B., Dawber T. R.: High density lipoprotein as a protective factor against coronary heart disease. The Framingham Study, Am JMed 62: 707— 714, 1977. Hartung G. H., Foreyt J. P., Mitchell R. E., Viasek I., Gotto A. M.:

Relationship of diet to high density lipoprotein cholesterol in middle-aged marathon runners, joggers, and inactive men. NEngi JMed 102: 357—361, 1980. 12 Jackson A. S., Pollack M. L.: Practical assessments of body composition. Physician Sports Med 13: 5,76—90, 1985. 3 Kattus A. A.: Chairman of the Committee on Exercise. Exercise Testing and Training of Apparently Healthy Individuals: A Handbook for Physicians. Dallas, TX: American Heart Association, p

1979.

30 Sloan A., Weir J.: Nomograms for prediction of body density and total body fat from skinfold measurements. J Appl Physiol 28: 2, 220—221, 1970.

31 Tran V. T., Weltman A., Glass G. V., Mood D. P.: The effects of exercise on blood lipids and lipoproteins: A meta-analysis of studies. Med Sci Sports Exerc 15 (5): 393—402, 1983.

32 Wood P. D., Haskell W. L., Stern M. P., Perry C.: Plasma lipoprotein distributions in male and female runners. Ann of NYAcad Sci 301: 748—761, 1977.

Wood P. D., Haskell W. L. Klein H., Lewis S., Stern M. P., Farguhar J. W.: The distribution of plasma lipoproteins in middleagedrunners.Metabolism 11:1249—1257,1977.

15, 1975.

" Kim H., Kalkhoff R.: Changes in lipoprotein composition during the menstrual cycle. Metabolism 28:663—668, 1979.

Arlette Perry, Ph. D.

Lopez S., Vial R., Balart L., Arroyave G.: Effect of exercise nad physical fitness on serum lipids and lipoproteins. Atherosclerosis

Human Performance Laboratory University of Miami P0 Box 248065 Coral Gables, FL 33124

29: 1—9, 1974. Manual of Laboratory Operations. Lipid Research Clinic Program,

Vol. 1 Lipid and Lipoprotein Analysis, DHEW publication 75— 628. Government Printing Office: Washington, DC, 1974.

Downloaded by: National University of Singapore. Copyrighted material.

References