Association of Physical Activity and Visceral ... - Wiley Online Library

Association of Physical Activity and Visceral Adipose Tissue in Older Women and Men Steven E. Riechman,* Robert E. Schoen,*† Joel L. Weissfeld,* F. Leland Thaete,‡ and Andrea M. Kriska*

Abstract RIECHMAN, STEVEN E., ROBERT E. SCHOEN, JOEL L. WEISSFELD, F. LELAND THAETE, AND ANDREA M. KRISKA. Association of physical activity and visceral adipose tissue in older women and men. Obes Res. 2002;10: 1065–1073. Objective: Physical inactivity, abdominal fat, and age are known risk factors for diabetes, cardiovascular disease, and certain cancers. Previous evidence supports an inverse relationship between physical activity (PA) and abdominal fat estimated by waist circumference. However, few investigations used computed tomography (CAT) scanning for precise measures of abdominal fat. Research Methods and Procedures: Sixty-five female and 106 male (age, 64.5 ⫾ 5.2 years) participants in the Prostate, Lung, Colon and Ovarian Cancer Screening Trial underwent a cross-sectional L4 –L5 CAT scan to differentiate visceral adipose tissue (VAT). Subjects were also interviewed by phone to determine PA and physical difficulties (PD). Results: Women had lower VAT (170 ⫾ 84 vs. 205 ⫾ 95 cm2, p ⫽ 0.014), lower VAT/total fat (29.9 ⫾ 7.2% vs. 42.6 ⫾ 10.2%, p ⬍ 0.001), and higher total fat (596 ⫾ 385 vs. 482 ⫾ 183 cm2, p ⫽ 0.010) than men. PA was inversely correlated to VAT (r ⫽ ⫺0.164, p ⫽ 0.034) and total fat (r ⫽ ⫺0.231, p ⫽ 0.003) in men and women. Those who reported a PD had higher VAT (249 vs. 180 cm2, p ⬍ 0.001) and total fat (652 vs. 500 cm2, p ⫽ 0.008). Multiple regression analysis indicated total PA and PD were independently associated to VAT and total fat.

Received for review March 19, 2002. Accepted for publication in final form June 18, 2002. *Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania; and †Department of Medicine and ‡Radiology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania. Address correspondence to Steven E. Riechman, School of Exercise, Leisure, & Sport, Kent State University, PO Box 5190, Kent, OH 44242-0001. E-mail: [email protected] Copyright © 2002 NAASO

Discussion: This investigation suggests a beneficial effect of PA and a negative influence of PD on abdominal fat accumulation. Although the cross-sectional design limits cause-effect designations, these results are consistent with other studies showing PA/abdominal fat relation. Key words: disability, sex, computed tomography scan, questionnaire

Introduction The risk for cardiovascular disease (1–3), type II diabetes (3– 6), and certain cancers (7–10) is associated with abdominal fat, particularly visceral adipose tissue (VAT). Similarly, physical activity (PA) levels are inversely associated with these same diseases (11–17). Progression through the fifth to seventh decade of life is marked by a dramatic increase in these chronic diseases, with a simultaneous increase in obesity and decline in PA (18). Moreover, physical inactivity has been shown to be associated with VAT accumulation (19). The increase in VAT accumulation observed with aging and its important patho-physiological consequences may therefore be the result of decreased PA. The prevalence of obesity peaked about the sixth decade of life in men and women in the most recent (1988 to 1994) National Health and Nutrition Examination Survey study (20). The major health implications of obesity (i.e., type II diabetes, ischemic heart disease, hyperlipidemia, hypertension, and cardiovascular disease) are directly associated with central fat patterning (21). The association of central adiposity to poor health is related to VAT accumulation and associated hyperinsulinemia. Increased VAT is associated with elevated free fatty acids (22), which impair hepatic insulin clearance, resulting in hyperinsulinemia, increased gluconeogenesis (20), and elevation of very-low-density lipoprotein secretion (22). Very-low-density lipoprotein secretion and insulin are independent predictors of myocardial infarction (23). Several studies suggest that dietary intake is consistent across the adult age span, implying that the increasing prevalence of obesity with age may be at least partially OBESITY RESEARCH Vol. 10 No. 10 October 2002

1065

Physical Activity and Visceral Adipose, Riechman et al.

caused by a decline in PA (24). Several epidemiological studies suggest an association of physical inactivity to abdominal fat accumulation using indirect measures such as waist circumference or the ratio of waist-to-hip circumference. However, few studies have examined the association of PA on visceral adipose accumulation determined objectively by computerized imaging. Computerized imaging also offers the opportunity to differentiate subcutaneous adipose from visceral adipose. Evidence of the association of PA to VAT is particularly sparse in older men and women. A study by Hunter et al. (25) is one of the only studies to investigate such an association using computerized tomography (CT) scanning and to find a strong association between activity indices and VAT in middle-age men. Other studies suggest that there is a preferential reduction in VAT in response to exercise training (19,26 –29). The purpose of this investigation was to identify the association of past year total PA with VAT in older men and women. This was examined in those with and without physical limitations, such as difficulty getting out of bed or walking across a room. It was hypothesized that those individuals with higher PA levels will have lower VAT accumulation, and this relation will remain if the analysis is restricted to those without any reported physical difficulties (PDs).

Pittsburgh PLCO subjects who underwent a screening flexible sigmoidoscopy (n ⫽ 981) were invited by mail to participate in this ancillary study investigating the association of abdominal fat and colon polyps, and 480 agreed (48.9%). Participants in this ancillary study were asked to undergo a single-slice CT scan through the L4 –L5 interspace for quantification of visceral adipose tissue. For the purposes of the current effort, the first one-third of the total sample that completed the CT scan were also asked to complete a PA questionnaire. A total of 171 of the 172 subjects contacted (99.4%) agreed to complete this PA survey involving a phone interview. These 171 did not differ significantly from the remaining 309 of the ancillary study by race, gender, age, body mass index (BMI), or VAT. These 309 individuals were not contacted to participate. All the subjects who were approached for the current effort had polyps identified on screening. Subjects returned to clinical care for further evaluation and treatment of abnormalities found on flexible sigmoidoscopy screening. On further diagnostic procedures, polyps found on sigmoidoscopy may be adenomatous or preneoplastic, or may be nonneoplastic such as hyperplastic or lymphoid aggregates. Of the 171 participants, 97 (56.7%) had an adenomatous polyp on follow-up, and the remainder (74; 43.3%) had nonneoplastic findings (n ⫽ 62) or no findings (lost or undetermined, n ⫽ 12).

Research Methods and Procedures Subjects The subjects in this study were selected from enrollees in the Pittsburgh site of the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO), a multicenter randomized clinical trial evaluating the effect of cancer screening tests on site-specific cancer mortality (30). All subjects in the intervention arm of the trial undergo periodic cancer screening tests including chest X-ray and flexible sigmoidoscopy. Additionally, men undergo a digital rectal exam and prostate-specific antigen test and women undergo a CA-125 and a vaginal ultrasound. Subjects were recruited into PLCO through mass mailings. Participants in the PLCO trial met the following eligibility criteria: (1) age 55 to 74 years; (2) not currently undergoing treatment for cancer except basal cell or squamous cell skin cancer; (3) no known prior cancer of the colon, rectum, prostate, lung, or ovaries; (4) no surgical removal of the colon, one lung, ovary, or prostate (women randomized before 1998 (n ⫽ 151) had at least one ovary, whereas subjects randomized after 1998 (n ⫽ 20) were not subject to that eligibility criterion); (5) not participating in another cancer screening or cancer prevention trial; (6) men not taking Proscar in the past 6 months and women not taking Novaldex in the past 6 months; (7) ability to provide informed consent; (8) no more than one prostate-specific antigen test in the past 3 years; and (9) no colonoscopy, sigmoidoscopy, or barium enema in the past 3 years. 1066

OBESITY RESEARCH Vol. 10 No. 10 October 2002

Quantification of VAT using Single-Slice CT Scanning After the initial lateral scout view using a 9800-CT scanner (General Electric, Milwaukee, WI), a single 10-mm cross-sectional abdominal scan was performed at 120 kVp and 170 mA with a scanning time of 2 s/scan (31). Using the trace function, a region of interest line was drawn at the junction of the subcutaneous adipose and the abdominal wall musculature, extending around the body to the back muscles. Using a pixel range of ⫺190 to ⫺30 Houndsfield units as the range for fat, the area of adipose tissue within the region of interest was determined using image analysis software. Using the trace function, the boundary separating the subcutaneous and visceral adipose was defined manually using a cursor, and the intra-abdominal adipose tissue area was recorded. Retroperitoneal adipose tissue was included in the visceral adipose tissue measurement. PA Questionnaire Past year leisure and occupational activity were assessed over the phone by a trained interviewer using the Modifiable Activity Questionnaire (14,32,33). The interviewer’s instructions and questionnaire calculations have been described previously (33). Hours per week (averaged over the past year) spent in occupational or leisure activities were converted to MET-hours per week by multiplying the hours per week by the relative intensity of the activity (MET). A value of 1 MET equals the energy requirements of an


Table 1. Subject characteristics stratified by gender

Age (years) Height (cm) Weight (kg) BMI (kg/m2) Visceral adipose tissue (cm3) Subcutaneous adipose tissue (cm3) Total adipose tissue (cm3) Visceral percent adipose tissue Physical Activity† Leisure (MET hours/week) Occupational (MET hours/week) Total (MET hours/week)

Women (n ⴝ 65; mean ⴞ SD)

Men (n ⴝ 106; mean ⴞ SD)

p Value*

64.5 ⫾ 4.9 162.6 ⫾ 6.7 74.2 ⫾ 15.6 28.1 ⫾ 5.6 170 ⫾ 84 381 ⫾ 127 551 ⫾ 193 29.9 ⫾ 7.2

64.5 ⫾ 5.6 177.5 ⫾ 7.7 88.4 ⫾ 14.7 28.0 ⫾ 4.2 205 ⫾ 95 277 ⫾ 118 482 ⫾ 183 42.6 ⫾ 10.2

0.919 ⬍0.001 ⬍0.001 0.945 0.014 ⬍0.001 0.022 ⬍0.001

10.0 (2.0 to 25.9) 60.0 (15.8 to 120.0) 65.1 (36.7 to 140.0)

20.0 (5.7 to 37.9) 44.0 (20.0 to 100.0) 65.2 (31.3 to 128.6)

0.005 0.980 0.302

* Student’s t test; difference between sex. † Median ⫾ 25th–75th percentile. BMI, body mass index.

individual at rest. Disability was assessed with three questions regarding difficulty getting in or out of bed, walking across a room, and/or walking for 10 minutes without resting. Walking for exercise has been excluded from all summary variables because past studies suggest this measure is relatively less reliable (33). Statistical Analysis Mean differences between genders for subject characteristics were analyzed using Student’s t test. These characteristics were normally distributed, satisfying assumptions of parametric tests; however, PA was not. Therefore, nonparametric Mann-Whitney U test was used to test for differences in PA between women and men. Nonparametric Spearman correlation coefficients were calculated for leisure, occupational, and total activity, and visceral, subcutaneous, total adipose, and visceral percent adipose tissue in women, men, and the total cohort. Corrections for non-normally distributed PA data were accomplished by categorizing the data into quintiles to perform further parametric analysis. Analysis of covariance was performed using visceral, subcutaneous, or total adipose or visceral percent adipose tissue as the dependent variable, quintiles of total activity or PDs as the independent variable, and age and gender as covariates in women, men, and the total cohort. Multiple regression analysis was performed to determine the independent association of gender, age, PDs, and quintiles of reported past year total activity to visceral, subcutaneous, and total adipose and visceral percent adipose tissue. The subcutaneous percent adipose tissue variable was omitted from analysis because it is redundant to visceral percent adipose tissue.

All analyses were conducted using SPSS software (version 10.1; SPSS Inc., Chicago, IL) and statistical significance was accepted at the p ⬍ 0.05 level.

Results Subjects The general characteristics of the 65 women and 106 men (97% white) who participated in the CT scan and PA questionnaire are presented in Table 1. Men were significantly taller and heavier than women, although BMI and age were similar. Men carried significantly greater VAT and visceral percent adipose than women, whereas women had increased subcutaneous (p ⬍ 0.001) and total abdominal adipose (p ⫽ 0.022) compared with men. Median leisure PA was significantly higher in men, but median occupational and total activity were similar. Leisure and/or Occupational Physical Activity and Abdominal Fat Components: Spearman Correlations Spearman correlation coefficients for past year leisure, occupational, and total physical activity and abdominal fat components for women and men are presented in Table 2. Leisure activity in women was significantly associated to visceral and total adipose, whereas occupational activity in men was significantly associated to visceral and visceral percent adipose tissue. Not surprisingly, the most consistent correlations to abdominal fat components were found with total PA. In these analyses, significant correlations were found for visceral and visceral percent adipose tissue in men, whereas correlations to visceral and total adipose were OBESITY RESEARCH Vol. 10 No. 10 October 2002

1067


Table 2. Spearman rank order correlations between physical activity and abdominal fat components Past year (MET hours/week) Leisure Women Men Total Occupational Women Men Total Total activity Women Men Total

Visceral fat

Subcutaneous fat

Total fat

Visceral percent

␳

p

␳

p

␳

p

␳

p

⫺0.302 ⫺0.102 ⫺0.123

0.015 0.299 0.111

⫺0.233 ⫺0.038 ⫺0.189

0.064 0.699 0.014

⫺0.288 ⫺0.079 ⫺0.198

0.021 0.424 0.010

⫺0.195 ⫺0.019 0.059

0.122 0.845 0.446

⫺0.084 ⫺0.208 ⫺0.140

0.510 0.033 0.070

⫺0.099 ⫺0.009 ⫺0.060

0.439 0.927 0.435

⫺0.110 ⫺0.144 ⫺0.155

0.388 0.144 0.040

⫺0.087 ⫺0.275 ⫺0.130

0.492 0.005 0.092

⫺0.201 ⫺0.202 ⫺0.164

0.111 0.039 0.034

⫺0.201 ⫺0.044 ⫺0.142

0.112 0.652 0.066

⫺0.228 ⫺0.159 ⫺0.202

0.070 0.105 0.008

⫺0.146 ⫺0.202 ⫺0.065

0.251 0.039 0.404

significant for the combined sample. For subcutaneous adipose, correlations were similar to visceral and total adipose in women; however, in men, these correlations were very low. Age-Adjusted Abdominal Fat Components by Quintile of Total Physical Activity To examine a potential threshold in the relation between abdominal fat and PA levels, the PA data were divided into quintiles (1 ⫽ low activity, 5 ⫽ high activity). Age-adjusted visceral, subcutaneous, and total adipose were significantly different across quintiles of total activity in men but not women (Table 3). Age- and gender-adjusted visceral and total adipose were significantly different across quintiles of total activity in the combined sample. Least significant difference post hoc analysis indicated that the lowest two quintiles of total activity had significantly higher visceral adipose than the third, fourth, and fifth quintile. The lowest two quintiles of total activity also had significantly higher total adipose than the fourth and fifth quintiles, whereas the first was also higher than the third.

Multivariate Analyses Multiple regression analysis was used to determine the independent association of reported past year PA levels and PDs to abdominal fat measures. These analyses are summarized in Table 5. Controlling for age and gender, higher total PA in the past year and the lack of PDs were significantly related to lower visceral, subcutaneous, and total adipose accumulation. The combined two-way interaction terms involving PA and other covariates did not significantly add to the model and were not included in the models presented. To examine the relation of PA and abdominal fat in those who do not have physical limitations to those who are active, data were reanalyzed, excluding those with reported PDs. The results of these analyses did not differ from the results presented (data not shown). Additionally, PDs and PA were still associated to abdominal fat measures when obese subjects (BMI ⬎ 30 kg/m2) were excluded from the analysis (data not shown). When data were analyzed separately for those with benign polyps and those with adenoma, the results were consistent with the findings of the total sample (data not shown).

Physical Difficulties and Abdominal Fat Components Age-adjusted abdominal fat means for women and men with and without any reported PDs are presented in Table 4. Mean visceral, subcutaneous, total adipose, and visceral percent adipose tissue were all higher in those reporting PD in women, men, and the total cohort. The trends for these mean differences were significant for all comparisons except for subcutaneous adipose in women and visceral percent adipose tissue in men and the total cohort. Those reporting PDs compared with those who did not had significantly lower occupational (44.1 vs. 68.4 MET 䡠 hours 䡠 week) and leisure (12.6 vs. 23.7 MET 䡠 hours 䡠 week) activity levels.

This investigation examined the association of PA levels and abdominal fat in older women and men. Using the gold standard of CT quantification for adipose tissue compartments, PA and PDs were found to be independently related to visceral, subcutaneous, and total adipose. Age and gender also contributed to the variance in abdominal adipose tissue compartments but did not eliminate the significant associations involving PA and PDs. Other investigations have demonstrated an inverse association of PA to central fat distribution (18). Few studies

1068


Discussion


Table 3. Abdominal fat components by quintiles of past year total (leisure and occupational) physical activity

Women (quintile: total activity)* 1 2 3 4 5 p (ANCOVA) Men (quintile: total activity)* 1 2 3 4 5 p (ANCOVA) Women ⫹ Men (quintile: total activity)† 1 2 3 4 5 p (ANCOVA)

Visceral

Subcutaneous

Total

Visceral %

187 198 158 146 158 0.474

396 417 402 311 377 0.231

583 615 560 457 535 0.289

31.1 31.2 27.9 30.3 28.6 0.742

275‡ 231§ 174 186 171 0.001

348¶ 274 236 279 248 0.022

623¶ 505 410 465 419 0.001

43.6 45.2 43.1 40.5 40.4 0.454

240‡ 219‡ 168 171 164 0.001

363 329 301 288 295 0.073

603‡ 547** 469 459 459 0.003

39.0 39.9 37.3 36.7 35.9 0.348

* Age adjusted. † Age and gender adjusted. ‡ p ⬍ 0.05 vs. 3, 4, 5 quintile (LSD post hoc analysis). § p ⬍ 0.05 vs. 3, 5 quintile (LSD post hoc analysis). ¶ p ⬍ 0.05 vs. 2, 3, 4, 5 quintile (LSD post hoc analysis). ** p ⬍ 0.05 vs. 4, 5 quintile (LSD post hoc analysis). ANCOVA, analysis of covariance; LSD, least-significant difference.

have addressed this question using CT quantification. In a study of 137 men, 30 to 71 years of age (25), a strong association between activity indices and VAT (measured by CT) was found and remained after adjusting for peripheral skinfolds. In another cross-sectional study, VAT was inversely related to PA in 220 white women (17 to 77 years of age) after adjusting for age, menopausal status, subcutaneous abdominal adipose, and percentage of total body fat (34). Prospective studies of fat loss suggest that the association of PA to visceral adipose is because of a preferential reduction in VAT in response to exercise training (19,26 –29). Women had lower visceral and increased subcutaneous adipose compared with men in this predominately white cohort. Similar differences between gender were reported in the Health ABC cohort of 1439 women and 1391 men between 70 and 79 years of age of white and African-

American decent (35). However, the white men and women of the Health, Aging and Body Composition (ABC) cohort had generally lower visceral adipose (131.8 cm2, women; 169.3 cm2, men) than that observed with the present population. These differences may be explained by age because the Health ABC cohort is an average of 10 years older than the current population. Although the prevalence of obesity defined by BMI increases from the third to the seventh decade, the prevalence of obesity declines from the seventh to the eighth decade as shown in National Health and Nutrition Examination Survey III (20). Similar to the age-related increase that occurs with BMI, relative visceral adipose volume increased with age, as shown in a cross-sectional study of 66 men and 96 women using CT scans (36). In that study, the age-related increase in the relative visceral adipose was greater in men than in premenopausal women but similar to postmenopausal OBESITY RESEARCH Vol. 10 No. 10 October 2002

1069


Table 4. Abdominal fat components (cm3) stratified by physical disability

Women* None (n ⫽ 56) Any (n ⫽ 9) p (ANCOVA) Men* None (n ⫽ 85) Any (n ⫽ 21) p (ANCOVA) Women ⫹ Men† None (n ⫽ 141) Any (n ⫽ 30) p (ANCOVA)

Visceral

SubQ

Total

Visceral %

BMI

Weight

160 236 0.015

371 452 0.096

531 688 0.033

29.2 34.7 0.037

27.2 31.5 0.072

161.5 179.6 0.172

193 259 0.005

258 356 0.001

451 615 ⬍0.001

42.4 43.5 0.657

27.6 31.4 ⬍0.001

189.1 219.8 ⬍0.001

180 249 ⬍0.001

301 395 ⬍0.001

481 644 ⬍0.001

37.4 39.8 0.197

27.3 31.5 ⬍0.001

178.6 205.8 ⬍0.001

* Age adjusted. † Age and gender adjusted. ANCOVA, analysis of covariance.

women. In the present study, age was only weakly associated to increased visceral and decreased subcutaneous adipose, resulting in a significant association of age to visceral percent adipose tissue. The association of abdominal fat to age may have been diminished in the present study by the narrow age range of the participants and inclusion of subjects in their eighth decade, but it remained consistent with other work in which visceral adipose, particularly the relative proportion, increases with age. In our study, there was a threshold of PA beyond which further increases in activity were not associated with further decreases in abdominal fat measures. This was particularly apparent for visceral adipose. Such an activity threshold may be important in designing public health recommendations for activity, but translation of these estimated activity levels are likely to be questionnaire-, interviewer-, and subject-specific. A previous study had also suggested a PA threshold using crude measures of obesity (37). Individuals who carried the Gln27Glu polymorphism in the ␤-2 adrenoreceptor gene and had low activity had increased obesity, as measured by weight and BMI, compared with noncarriers or carriers with comparatively high activity. The association of PA, catecholamine receptor, and central adiposity in a thresholddependent manner is biologically plausible for several reasons. The visceral component of abdominal fat is more metabolically active and responsive to catecholamines than subcutaneous or other fat depots (29), and catecholamines increase with exercise (38). A review of investigations by Ross et al. (19) indicate there is a preferential reduction in VAT in response to exercise training possibly through the 1070


effect of catecholamines (29) because other forms of negative energy balance (i.e., caloric restriction) seem to reduce abdominal fat components equally (26 –28). Our results indicate that although all abdominal fat components are lower in those with higher activity, the magnitude of the visceral adipose difference is greater. Thus, catecholamines may partially explain the association of PA and the relative amount of visceral adipose. Those individuals reporting PDs were, not surprisingly, found to have significantly lower PA levels as determined by questionnaire. Therefore, inclusion of subjects who reported PDs or who were obese may have limited ability to be active, which might foster the association of PA and abdominal fat. However, the association of PA to abdominal fat remained after exclusion of those with PDs or those who were obese, confirming this robust relation. Likewise, PDs were significantly associated with higher abdominal fat levels, even after accounting for differences in PA levels. One reason for the independent contribution of the existence of PDs is that it may be secondary to subclinical disease, which in turn, can result in alteration in fat deposition or metabolic regulation. A secondary reason may be that PDs measure a different portion of the PA spectrum than that assessed by our PA measure. PA levels in this study were obtained by questionnaire, which assesses activities more intense than the activities of daily living. Because the greatest amount of time in a day is spent in lower intensity activities, a decline in the amount of time spent in activities of daily living in those with PDs may result in the accumulation


Table 5. Multiple regression analysis to determine the independent association of age, gender, indicators of physical disability, and the quintiles of total physical activity to abdominal fat components Variable DV ⫽ visceral adipose Gender Age (years) Disability (yes/no) Total physical activity* DV ⫽ subcutaneous adipose Gender Age (years) Disability (yes/no) Total physical activity* DV ⫽ total adipose Gender Age (years) Disability (yes/no) Total physical activity* DV ⫽ visceral percentage Gender Age (years) Disability (yes/no) Total physical activity*

Nonstandardized regression coefficient

SE

p Value

⫺33.4 2 57.8 ⫺56.3

13.5 1.2 18.1 13.4

0.014 0.108 0.002 ⬍0.001

106.9 ⫺2.6 77.6 ⫺44.5

18.8 1.7 25.2 18.6

⬍0.001 0.135 0.002 0.018

73.4 ⫺5.7 135.1 ⫺100.8

28.0 2.6 37.6 27.8

0.010 0.821 ⬍0.001 ⬍0.001

⫺0.124 0.005 0.013 ⫺0.027

0.014 0.001 0.019 0.014

⬍0.001 ⬍0.001 0.483 0.061

* Dichotomized to high/low activity (1, 2 vs. 3, 4, 5 of the quintiles): gender, men ⫽ 1 and women ⫽ 2; disability: none ⫽ 1 and any ⫽ 2; quintile: lowest ⫽ 1 and highest ⫽ 5. DV, dichotomized dependent variable.

of fat over time and would be independent of the contribution of the higher intensity “PA levels.” One limitation of this study is the inclusion of individuals with colon polyps from a larger study, the PLCO. However, controlling for histological diagnosis of the polyps did not alter the association of PA and PDs to abdominal fat components. Another limitation is the subjective assessment of PA. Also, the cross-sectional design of this study limits determination of a cause-effect relation. The finding of this study, however, is supported by several prospective studies that have demonstrated a cause-effect relation between exercise and visceral adipose loss (19). The current investigation sought to determine if physical inactivity and PDs were important determinants of abdominal fat components in older women and men. Physical inactivity and PDs were independently associated to increased abdominal fat, and there seems to be a threshold from which increases in PA no longer reduce visceral adipose. These results add support to public health recom-

mendations to increase PA for the prevention of the health implications of increased central obesity in older women and men.

Acknowledgments This study was supported by an American Cancer Society Institutional Research Grant and by the National Cancer Institute (K07-CA72561). The authors thank Amy Borner and Karen Foley for their assistance on this project. References 1. Folsom AR, Kaye SA, Sellers TA, et al. Body fat distribution and 5-year risk of death in older women. JAMA. 1993;269: 483–7. 2. Lapidus L, Bengtsson C, Larsson B, Pennert K, Rybo E, Sjo¨stro¨m L. Distribution of adipose tissue and risk of cardiovascular disease and death: a 12 year follow-up of participants in the population study of women in Gothenburg, Sweden. BMJ. 1984;289:1257– 61. OBESITY RESEARCH Vol. 10 No. 10 October 2002

1071


3. Bjo¨ rntorp P. “Portal” adipose tissue as a generator of risk factors for cardiovascular disease and diabetes. Arteriosclerosis. 1990;10:493– 6. 4. Kaye SA, Folsom AR, Sprafka JM, Prineas RJ, Wallace RB. Increased incidence of diabetes mellitus in relation to abdominal adiposity in older women. J Clin Epidemiol. 1991; 44:329 –34. 5. Ohlson LO, Larsson B, Svardsudd K, et al. The influence of body fat distribution on the incidence of diabetes mellitus:13.5 years of follow-up of the participants in the study of men born in 1913. Diabetes. 1985;34:1055– 8. 6. Pouliot MC, Despre´ s JP, Nadeau A, et al. Visceral obesity in men: associations with glucose tolerance, plasma insulin, and lipoprotein levels. Diabetes. 1992;41:826 –34. 7. Sellers TA, Kushi LH, Potter JD, et al. Effect of family history, body fat distribution, and reproductive factors on the risk of postmenopausal breast cancer. N Engl J Med. 1992; 326:1323–9. 8. Schapira DV, Kumar NB, Lyman GH, Cavanagh D, Roberts WS, LaPolla J. Upper-body fat distribution and endometrial cancer risk. JAMA. 1991;266:1808 –11. 9. Lapidus L, Helgesson O, Merck C, et al. Adipose tissue distribution and female carcinomas: a 12-year follow-up of participants in the population study of women in Gothenburg, Sweden. Int J Obes. 1988;12:361– 8. 10. Schoen RE, Tangen CM, Kuller LH, et al. Increased blood glucose and insulin, body size, and incident colorectal cancer. J Natl Cancer Inst. 1999;91:1147–54. 11. Ivy JL. Role of exercise training in the prevention and treatment of insulin resistance and non-insulin-dependent diabetes mellitus. Sports Med. 1997;24:321–336. 12. Powell KE, Thompson PD, Caspersen CJ, Kendrick JS. Physical activity and the incidence of coronary heart disease. Annu Rev Publ Health. 1987;8:253– 87. 13. Kriska AM, Blair SN, Pereira MA. The potential role of physical activity in the prevention of non-insulin-dependent diabetes mellitus: the epidemiological evidence. Exercise Sport Sci R. 1994;22:121– 43. 14. Kriska AM, Bennett PH. An epidemiological perspective of the relationship between physical activity and NIDDM: from activity assessment to intervention. Diabetes-Metab Rev. 1992;8:355–72. 15. Pereira MA, Kriska AM, Day RD, Cauley JA, Laporte RE, Kuller LH. A randomized walking trial in postmenopausal women: effects on physical activity and health 10 years later. Arch Intern Med. 1998;158:1695–701. 16. Kriska AM. Physical activity and coronary heart disease in a biracial population. Clin J Sport Med. 1998;8:71. 17. Shephard RJ, Shek PN. Association between physical activity and susceptibility to cancer: possible mechanisms. Sports Med. 1998;26:293–315. 18. Toth MJ, Beckett T, Poehlman ET. Physical activity and the progressive change in body composition with aging: current evidence and research issues. Med Sci Sport Exerc. 1999;31: S590 –S6. 19. Ross R. Freeman JA. Janssen I. Exercise alone is an effective strategy for reducing obesity and related comorbidities. Exerc Sport Sci Res. 2000;28:165–70. 1072


20. Kotz CM, Billington CJ, Levine AS. Obesity and aging. Advances in the care of older people with diabetes. Clin Geriatr Med. 1999;15:391– 412. 21. National Institutes of Health, National Heart Lung and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. Rockville, MD: National Institutes of Health, National Heart, Lung, and Blood Institute; 1998. 22. Matsuzawa Y, Shimomura I, Nakamura T, et al. Pathophysiology and pathogenesis of visceral fat obesity. Obes Res. 1995;3:187S–194S. 23. Despre´ s JP, Lamarche B, Mauriege P, et al. Hyperinsulinemia as an independent risk factor for ischemic heart disease. New Engl J Med. 1996;334:952–7. 24. Jebb SA, Moore MS. Contribution of a sedentary lifestyle and inactivity to the etiology of overweight and obesity: current evidence and research issues. Med Sci Sport Exerc. 1999;31:S534 –S41. 25. Hunter GR, Kekes-Szabo T, Snyder SW, Nicholson C, Nyikos I, Berland L. Fat distribution, physical activity and cardiovascular risk factors. Med Sci Sports Exerc. 1997;29: 362–9. 26. Janssen I, Ross R. Effects of sex on the change in visceral, subcutaneous adipose tissue and skeletal muscle in response to weight loss. Int J Obes. 1999;23:1035– 46. 27. Despre´ s JP, Nadeau A, Bouchard C. Physical training and changes in regional adipose tissue distribution. Acta Med Scand Suppl. 1988;723:205–12. 28. Ross R, Janssen I. Is abdominal fat preferentially reduced in response to exercise-induced weight loss? Med Sci Sports Exerc. 1999;31:S568 –S72. 29. Jensen MD, Cryer PE, Johnson CM, Murrey MJ. Effects of epinephrine on regional free fatty acid and energy metabolism in men and women. Am J Physiol. 1996;270: E259 –E64. 30. Prorok PC, Andriole GL, Bresalier, et al. Prostate, lung, colorectal, and ovarian cancer screening trial project team. Design of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Control Clin Trials. 2000;21(Suppl): 273–309S. 31. Schoen RE, Thaete FL, Sankey SS, Weisfield JL, Kuller LH. Sagittal diameter in comparison with single slice CT as a predictor of total visceral adipose tissue volume. Int J Obes. 1998;22:338 – 42. 32. Kriska AM, Caspersen CJ. A collection of physical activity questionnaires for health related research. Med Sci Sports Exerc. 1997;29:S73–S78. 33. Kriska AM, Knowler WC, LaPorte RE, et al. Development of a questionnaire to examine the relationship of physical activity and diabetes in Pima Indians. Diabetes Care. 1990; 13:401–11. 34. Hunter GR, Kekes-Szabo T, Treuth MS, Williams MJ, Goran M, Pichon C. Intra-abdominal adipose tissue, physical activity and cardiovascular risk in pre- and post-menopausal women. Int J Obes. 1996;20:860 –5. 35. Harris TB, Visser M, Everhart J, et al. Waist circumference and sagittal diameter reflect total body fat better than visceral


fat in older men and women: the Health, Aging and Body Composition Study. Ann NY Acad Sci. 2000;904:462–73. 36. Kotani K, Tokunaga K, Fujioka S, et al. Sexual dimorphism of age-related changes in whole-body fat distribution in the obese. Int J Obes. 1994;18:207–12.

37. Meirhaeghe A, Helbecque N, Cottel D, Amouyel P. ␤-2 adrenoreceptor gene polymorphism, body weight, and physical activity. Lancet. 1999;353:896. 38. Galbo H. Hormonal and Metabolic Adaptation to Exercise. New York: Thieme-Stratton; 1983.


1073