The Effect of Body Fat Distribution

Pulmonary Function Tests*

on

Lynell C. Collins, MD, FCCP; Phillip D. Hoberty, EdD, RRT;

Jerome F. Walker, EdD, RRT; Eugene C. Fletcher, MD, FCCP;
and Alan N. Peiris, MD

Although the influence of obesity on pulmonary function tests has been examined, the role of body fat
distribution has received limited attention. Pulmonary
studies of patients severely affected by upper body obesity suggest they have more severely compromised lung
volumes than obese patients with lower body obesity.
We examined 42 healthy but normal or mildly obese
men to determine if body fat distribution influences
pulmonary function tests.
Multiple measures of adiposity showed a significant inverse relationship with both spirometry and static lung
volumes. However, the biceps skinfold thickness had the
strongest inverse relationship with total lung capacity
(TLC) compared to other anthropometric measures. The
waist-to-hip ratio (WHR) demonstrated a significant inverse relationship with static lung volumes only when
controlling for cigarette smoking. However, comparing
pulmonary function tests between patients with a WHR
less than 0.950 (lower body fat distribution) and subjects

with a WHR of 0.950 or greater (upper body fat distribution) revealed that FVC, FEVI, and TLC were significantly lower in the patients with upper body fat distribution. Stepwise multiple regression analysis was done
using all anthropometric variables and age which generated predictive equations that included the biceps
skinfold thickness for residual volume (RV) and TLC.
This suggests that upper body fat distribution may be
associated with a modest impairment of lung volumes in
normal and mildly obese men. Until the findings of this
study can be applied to a larger, ethnically and anthropometrically diverse population, and to women, we believe caution is warranted when standard equations are
used to predict pulmonary function tests in an anthropometrically diverse population.
(Chest 1995; 107:1298-1302)
Key words: pulmonary function tests; body fat distribution; obesity; waist-to-hip ratio; anthropometric measures

40292

though the influence of obesity on pulmonary function tests has been examined,4'5'8 the role of body fat
distribution has received limited attention. One
widely used measure of determining upper versus
lower body fat is the waist-to-hip ratio (WHR), with
a WHR of 0.950 or greater indicating upper body
obesity and a WHR of less than 0.950 indicating
lower body obesity.9 Preliminary studies of severely
obese persons with upper body obesity suggest that
they have more severely compromised lung volumes
than those with lower body obesity.10"'1 A decrease in
lung volumes appears to increase respiratory resistance contributing to exercise-induced dyspnea in
obese patients.8
The effect of body fat distribution on pulmonary
function tests in nonobese and/or mildly obese subjects has not been reported. Current lung function
prediction equations utilize only general measures of
body size (height and weight).12"13 If abdominal or
upper body deposition of fat decreases diaphragmatic excursion due to increased abdominal adiposity or increased weight on the chest wall, measured
lung volumes might be smaller than predicted using
current predictive equations. We examined healthy,
adult men to determine if distribution of body fat
significantly influences pulmonary function tests.

1298

Clinical Investigations

Approximately 20% of the US population are

overweight.' Obesity has been associated with
respiratory complications such as obstructive sleep
apnea2 and obesity hypoventilation syndrome,3 and
it is believed to reduce lung volumes. Obese persons
may have decreased expiratory reserve volumes
(ERV), particularly when in the recumbent position.4
However, excessive body weight may also indicate
robustness and muscularity, which may improve lung
volumes.5
A variety of anthropometric indices of body fatness
and body fat distribution have demonstrated that
upper body obesity carries a higher risk of cardiovascular and metabolic disease than does lower body
obesity in investigations of severe adiposity."6'7 Al*From Division of Respiratory and Environmental Medicine;
University of Louisville School of Medicine and Louisville Veterans Administration Medical Center (Drs. Collins and Fletcher); Division of Respiratory Therapy; University of Louisville
School of Allied Health (Drs. Hoberty and Walker) Louisville,
Kentucky; Division of Endocrinology, College of Medicine, East
Tennessee State University and Mountain Home Veterans Administration Medical Center (Dr. Peiris) Johnson City, Tennessee.
Manuscript received November 1, 1993; revision accepted September 22, 1994.
Reprint requests: Dr. Collins, Division of Pulmonary Medicine,
University of Louisville, Health Sciences Center, Louisville, KY

Downloaded From: http://journal.publications.chestnet.org/ by a Grantham & District Hosptial User on 12/04/2014

MATERIALS AND METHODS

Forty-four healthy male firefighters were referred for fitness
evaluations, none of which had any known medical problems.
Both current and exsmokers were included in the group. Evaluation included spirometry, lung volumes, and body fat determination, and a fasting serum sample was obtained to evaluate
triglycerides, cholesterol, and glucose levels. Our study was
reviewed and approved by the Human Studies Committee of the
University of Louisville (Kentucky) School of Medicine. Informed
consent from all subjects was obtained before they were tested.
Spirometry was performed by one tester using a calibrated
Gould Sentry System rolling-seal spirometer (Gould, Houston)
while subjects were seated. FVC, FEVI, FEF25-75, and FEVI/
FVC were measured and volumetric testing was performed in the
seated position utilizing the helium dilution technique on a calibrated Gould-Godart Pulmonet III (Gould-Godart, Bilthoven,
the Netherlands). Residual volume (RV) was measured, with total lung capacity calculated as the sum of the FVC and RV.
Spirometry and lung volume measurements were referenced
against predicted values using the equations of Crapo et al.12"13
The percentage of body fat was determined by the multiple
skinfold technique'4 performed by a single observer using a
Lange Skinfold Caliper (Cambridge Scientific Industries, Cambridge, Md). Skinfolds were measured at the biceps, triceps, and
at the subscapular, and suprailiac. Each skinfold was measured
three times and reported as the average of the three measurements. Waist and hip girth measurements were obtained using a
cloth measuring tape around the area of greatest girth of the abdomen and hips.9
Indices of obesity included the percent of body fat (% BF) calculated from the four skinfolds, body mass index (BMI), kg/m2,
and percent of ideal body weight (% IBW) (Metropolitan Life
Tables, 1983). Measures of body fat distribution were the waist to
hip ratio (WHR) and centrality index (CI) (ratio of subscapular
to triceps skinfolds).
Statistical analysis was performed using SPSS Release 4.1 software (SPSS, Chicago). Initial analysis consisted of preparing a
Pearson correlation matrix to assess relationships between the indices of obesity and the selected tests of pulmonary function.
Pearson correlations were compared to Spearman p values in order to account for the possibility of abnormally distributed variables. Partial correlations between significantly related variables
were then examined, controlling for the effect of pack-year
smoking history, which itself may influence pulmonary function
measurements. 15
Unpaired Student's t tests were used to compare the pulmonary
function results of subjects with mild obesity (percent of IBW
>120% but <150% or BMI >27.8 kg/m2 but <31.1 kg/M2)'6 to
normal subjects (percent of IBW '<120% or BMI <27.8 kg/m2).
The waist to hip ratio (WHR) has been used to define upper body
obesity (a WHR -0.950) (9). An unpaired Student's t test was
used to compare pulmonary function results of subjects with upper body fat distribution to those with lower body fat distribution,
as determined by the WHR. Lastly, stepwise multiple regression
analysis using all anthropometric variables and indices of obesity
was performed for each spirometry or lung volume test.
RESULTS

Demographic characteristics of the subjects and

the results of spirometric and lung volume testing are
shown in Table 1. Forty-four subjects underwent
spirometric testing. Two firefighters who did not
undergo hydrostatic weighing, also did not have lung
volumes measured. Therefore, 42 subjects had lung

Table 1-Demographic Characteristics of Subjects*

Parameter
Age, yrs
Height, cm
Weight, kg
% IBW. %
% BF, %
BMI, kg m2
WHR
CI
Biceps, mm
Triceps, mm
Subscapular, mm
Suprailiac, mm
Smokers, %
FVC, % of pred
FEVy, % of pred
FEVI/FVC, %
RV, % of pred
TLC, % of pred

Normal
n=28

Mildly Obese
n=14

Combined
n=42

35.91.8
176.5+1.2

33.7 1.4
174.5+1.6
105.33.9
132.3+2.6
28.30.9
33.3 0.8
0.976 0.01
1.5630.3
12.8 1.2
20.3 2.1
26.5 +2.3
35.92.9
35.7
93.72.5
95.12.8

35.21.3
175.90.94
90.8 2.4
110.62.9
24.7+0.8
28.4 0.7
0.929 0.01
1.30.1
9.40.7
16.5 0.9
19.3 1.4
24.3 1.9
26.2
95.21.6
95.41.8
80.21.9
106.4 +2.9
98.9 1.6

83.62.91
99.7 2.9
23.1+1.0
25.9 0.5f
0.906 + O.Olf
1.147+0.09
7.6 + 0.6f
14.6 + 0.8f
15.9 + 1.3f
18.5 + 1.7f
21.4
95.9+2.1
95.5 +2.4
78.8 2.8
108.4 3.4
100.2 + 2.0

83.2+0.9
101.7 +5.9
95.9 2.6

*Obesity defined as >120% of ideal body weight.tp<0.05 (comparing normal with mildly obese subjects); Demographic and physiologic characteristics of the total group of subjects and in the
subgroups of normal weight and mildly obese subjects. %
IBW=percent of ideal body weight; % BF=percent body fat;
BMI=body mass index; WHR=waist-to-hip ratio; Cl=centrality
index (subscapular skinfold/triceps skinfold); biceps=biceps skinfold thickness; triceps=triceps skinfold thickness; subscapular=
subscapular skinfold thickness; and, suprailiac=suprailiac skinfold
thickness; RV=residual volume; TLC=total lung capacity.

volumes measured in addition to their spirometry.

The two subjects who did not have lung volumes
measured were excluded from this study. None of the
subjects were severely obese (150% or greater of their
IBW), but 18 subjects were mildly obese, weighing
120% or more of their IBW. Table 1 includes the data
on all subjects, and on the subgroups composed of
normal weight and mildly obese individuals (mildly
obese being >120% IBW). Significant differences
identified by Student's t test between the mildly
obese and normal subjects are indicated with asterisks. No differences in pulmonary functions were
observed between these two groups. A comparison of
men divided into normal and obese groups based on
BMI was almost identical to Table 1, and therefore is
not shown.
Pearson correlations are shown in Table 2. Those
values listed as statistically significant were confirmed by Spearman p. Multiple measures of adiposity showed a significant inverse relationship with
both tests of spirometry, FVC, and FEV1, and lung
volume, TLC (p<0.05). However, the biceps skinfold
thickness had the strongest negative relationship with
TLC compared to other anthropometric measures.
The Pearson correlation matrix composed of variables from the normal weight men (<120% of their
IBW) showed stronger negative relationships between the anthropometric variables and pulmonary
CHEST / 107 / 5/ MAY, 1995

Downloaded From: http://journal.publications.chestnet.org/ by a Grantham & District Hosptial User on 12/04/2014

1299

Table 2-Pearson Correlation CoefficientsAll Subjects*

% IBW
% BF
BMI
WHR
CI
Biceps
Triceps
Subscapular
Suprailiac

Liters
8-

FVC

FEVy

RV

TLC

-0.28
-0.51

-0.15

-0.42f
-0.39f

-0.40f
-0.27
-0.14
-0.46
-0.30

-0.27
-0.22
-0.10
-0.40
-0.27

-0.38f

-0.3lf

-0.31

-0.18

-0.36t
0.05
-0.27
0.01
0.17
-0.26
-0.21
0.01
-0.30

-0.42f

7-

-0.48t
-0.23
-0.04

6-

-0.55f
-0.36f
-0.34f

-0.41t

*n=42; Pearson correlation coefficients for anthropometric measures

and pulmonary function tests for all of the subjects tested.

tp<0.05.

function tests than seen in the entire group (see Table 3). Biceps skinfold thickness showed strongest
correlation to all spirometric values and static lung
volumes except to RV. Partial correlation analysis of
all the subjects, controlling for pack-year smoking
history revealed a statistically significant inverse relationship between WHR and biceps skinfold thickness with static lung volumes.
We compared pulmonary function tests between
the subjects with a WHR <0.950 (n=25) and those
with a WHR 20.950 (n=17). FVC, FEV1 and TLC
were lower in subjects with a WHR >0.950, p<0.05

(Fig 1).

Stepwise multiple regression analysis for all subjects using all anthropometric variables and age was
performed for each spirometric and volumetric
measurement, generating the coefficients in Table 4.
Age, height, % BF, BMI, and the biceps skinfold
thickness all contributed to the predictive equations.
Stepwise multiple regression analysis for the normal
group was done, and revealed the biceps skinfold to
be a predictor for all pulmonary function tests except
for RV.
DISCUSSION
To our knowledge this is the first study examining
the effect of body fat distribution on pulmonary
Table 3-Pearson Correlation Coefficients-Normal

Subjects*
FVC

% IBW
% BF
BMI
WHR
CI
Biceps
Triceps

-0.25
-0.55t
-0.18
-0.18
-0.01

-0.60f
-0.52f
-0.40ft

FEVI

RV

TLC

-0.12

-0.46f

-0.46f

0.18
-0.26
0.24
-0.18
-0.14
-0.26
-0.01
-0.18

-0.4O3
-0.37

-0.27
-0.16
-0.01

-0.55t

-0.40f

-0.43f
-0.30
-0.08

-0.56t

-0.55f

Subscapular
-0.33
-0.34
-0.32
-0.15
Suprailiac
-0.35
*n=28; Pearson correlation coefficients for anthropometric measures
and pulmonary function tests for the normal weight subjects.
tp<o.os.

FVC

FEV1

EI1WHR

TLC

> 0.950 LIWHR < 0.950

FIGURE 1. Spirometry and lung volumes in subjects with a

waist-to-hip ratio (WHR) >0.95 (n=17) compared with those
with a WHR <0.95 (n=25). *p<0.05.

function tests in lean and mildly obese men. In this

investigation, we found that a simple measure of
body fat distribution, primarily the biceps skinfold,
improved prediction of normal pulmonary measurements in nonobese adults compared with predictions
based on general measurements.12'13 This extends the
preliminary findings in morbidly obese patients,
suggesting that body fat distribution may affect pulmonary function tests.10'11 The biceps skinfold thickness is a marker for fat distribution to the upper
body,17 therefore our findings suggest that upper
body fat distribution may be associated with a mild
decrease in spirometry and lung volumes.
In the preliminary work of Enzi et al and Muls et
al,10"'1 which assessed pulmonary function in morbidly obese patients, the WHR was clearly differentiated in subjects having the most severe lung volume
compromise (those with a WHR 20.95) and in those
with less respiratory impairment (those with a WHR
<0.95). In these obese patients, the intra-abdominal
adipose pressing upward on the diaphragm prevents
full downward excursion during deep inspiration,
which decreases TLC, FVC, and FEV1'8 as shown in
our study. However, the lack of correlation between
WHR and confidence intervals in pulmonary function tests suggests that truncal or intra-abdominal fat
does not contribute significantly to these measurements in normal or mildly obese men.
A shortcoming of all the anthropometric measures
used in skinfold analysis is the inability to accurately
assess visceral adiposity.'7 9 The weaker relationship
of the biceps to pulmonary function tests as observed
in the mildly obese men, may reflect the inability of
the biceps skinfold to determine intrathoracic fat
deposition. Chest CTs (both conventional and highresolution) reveal a high prevalence of subpleural fat

1300
Downloaded From: http://journal.publications.chestnet.org/ by a Grantham & District Hosptial User on 12/04/2014

Clinical Investigations

Table 4-Regression Correlation Coefficients -All Subjects*

Pulmonary
Function

Age

FVC

Height

% BF

BMI

0.90

-0.11
-0.05

0.004

FEVy
RV
TLC

0.36
0.83

-0.03
-0.11

Constant

Multiple
R

Multiple

Biceps

-0.03
-0.11

0.16
5.42
1.09
2.42

0.69
0.42
0.69
0.70

0.48
0.18
0.48
0.49

R2

*n=42; multiple regression correlation coefficients determined for each of the pulmonary function tests using all anthropometric variables and
age. No equation was generated to predict FEV1/FVC.

deposition in patients undergoing evaluation for asbestos pleural disease.20'2' In these investigations, no
correlation between body size and subpleural fat
deposition was noted, but a formal correlation analysis was not performed. The large amount of intersubject variability seen in FVC in normal subjects22
and the differences in FVC, FEV1, and TLC detected
in this study, may reflect alterations in elastic recoil'8
due to small amounts of subpleural fat. Determination of intrathoracic fat by CT or MRI may be useful in further quantifying this relationship.
Smokers could have biased the results obtained,
decreasing spirometry and increasing lung volumes
from cigarette-induced emphysema and chronic
bronchitis even in young, asymptomatic subjects.23 A
partial correlation analysis controlling for cigarette
smoking suggests that the contribution of this factor
is minimal in this study population. However, additional nonsmokers must be studied to clarify this issue. Cigarette smokers are also known to have higher
WHRs than nonsmokers, which declines after cessation of smoking.24'25 Furthermore, upper body fat
deposition is known to be more prevalent in men than
in women;26 therefore, the observations in this study
need to be confirmed for women.
In this study, the distribution of body fat was found
to effect spirometry and lung volumes in normal,
nonobese adults men, and to a lesser extent in mildly
obese men. The best measure of fat distribution influencing pulmonary function tests in this study
population was the biceps skinfold, which is a measure of upper body fat distribution.'7 The findings of
this study need to be applied to a larger sample of
nonsmoking men, to a more ethnically and anthropometrically diverse population, and to women. In
any event, we believe caution is warranted when
standard equations are used to predict pulmonary
function tests in an anthropometrically diverse population pending further study.

ACKNOWLEDGMENTS: This work was supported in part by

the Allen & Hanburys division of Glaxo, Inc. The authors wish to
thank Bryant Stamford, PhD, for his advice and assistance in the
completion of this study.

REFERENCES
1 Kissebah AH, Freedman DS and Peiris AN. Health risks of
obesity. Med Clin N America 1989; 73:111-38
2 Guilleminault C, Simmons FB, Motta H. et al. Obstructive sleep
apnea syndrome and tracheostomy: long term follow-up experience. Arch Intern Med 1981; 141:1228-231
3 Lopata M, Onal E. Mass loading, sleep apnea, and the pathogenesis of the obesity hypoventilation syndrome. Am Rev
Respir Dis 1982; 126:640-45
4 Ray C, Bray G, Hansen J, et al. Effects of obesity on respiratory
function. Am Rev Respir Dis 1983; 128:501-6
5 Leech JA, Ghezzo H, Stevens D et al. Respiratory pressures and
functions in young adults. Am Rev Respir Dis 1983; 128:17-23
6 Vague J. The degree of masculine differentiation of obesities:
A factor determining predisposition to diabetes, atherosclerosis,
gout, and uric calculous disease. Am J Clin Nutr 1956; 4:20-34
7 Kissebah A, Vydelingum N, Murray R, et al. Relation of bodyfat distribution to metabolic complications of obesity. J Clin
Endocrinol Metab 1982; 54:254-60
8 Zerah F, Harf A, Perlemuter L, et al. Effects of obesity on respiratory resistance. Chest 1993; 103:1470-76
9 Evans DJ, Hoffman RG, Kalkhoff RK, et al. Relationship of
body fat topography to insulin sensitivity and metabolic profiles
in premenopausal women. Metab 1984; 33:68-75
10 Enzi G, Baggio B, Vianello A, et al. Respiratory disturbances in
visceral obesity. Int J Obes 1990; 14(suppl 2):26
11 Muls E, Vryens C, Michels A, et al. The effects of abdominal
fat distribution measured by computed tomography on the
respiratory system in non-smoking obese women. Int J Obes
1990; 14(suppl 2):136
12 Crapo RO, Morris AH, Gardner RM. Reference spirometric
values using techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 1980; 123:659-64
13 Crapo RO, Morris AH, Clayton PD, et al. Lung volumes in
health nonsmoking adults. Bull Europ Physiopath Resp 1982;
18:419-25
14 Durnin JVGA, Womersley J. Body fat assessed from total body
density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J
Nutr 1974; 32:77-97
15 Norusis MJ. SPSS Introductory Statistics Student Guide. Chicago: SPSS 1990; 283-306
16 Van Itallie TB. Health implications of overweight and obesity
in the United States. Ann Intern Med 1985; 103:983-88
17 Mueller WH, Wear ML, Hanis CL, et al. Which measure of
body fat distribution is best for epidemiologic research? Am J
Epidemiol 1991; 133:858-69
18 Anthonisen NR. Tests of mechanical function. In: Fishman AP,
ed. Handbook of Physiology: The Respiratory System, Vol. III,
Part 2. 1986; Bethesda, Md: American Physiological Society
753-84
19 Despres J-P, Prudhomme D, Pouliot M-C, et al. Estimation of
deep abdominal adipose-tissue accumulation from simple anthropometric measurements in men. Am J Clin Nutr 1991;
CHEST / 107 / 5 / MAY, 1995

Downloaded From: http://journal.publications.chestnet.org/ by a Grantham & District Hosptial User on 12/04/2014

1301

54:471-77
20 Sargent EN, Boswell WD, Ralls PW, et al. Subpleural fat pads
in patients exposed to asbestos: distinction from non-calcified
pleural plaques. Radiol 1984; 152:273-77
21 Friedman AC, Fiel SB, Fisher MS, et al. Asbestos-related pleural disease and asbestosis: a comparison of CT and chest radiography. AJR 1988; 150:269-75
22 Hyatt RE. Forced expiration. In: Fishman AP, ed. The Handbook of Physiology: The Respiratory System Vol II, Part 1. Bethesda, Md: American Physiological Society 1986; 295-314
23 Niewoehner DE, Kleinerman J, Rice DB. Pathologic changes in

the peripheral airways of young cigarette smokers. N Engli

Med 1974; 291:7-55-58
24 Shimokata H, Mueller DC, Andres R. Studies in the distribution
of body fat: III. Effects of cigarette smoking. JAMA 1989;
261:1169-173
25 Barrett-Connor E, K-T Khaw. Cigarette smoking and increased
central adiposity. Ann Intern Med 1989; 111:783-87
26 Kissebah AH, Evans DJ, Peiris AN, et al. Endocrine characteristics in regional obesities; Role of sex steroids. In: Vague J, ed.
Metabolic Complications of Human Obesities. Amsterdam:
Elsevier Science Publishers 1985; 115

1 302
Downloaded From: http://journal.publications.chestnet.org/ by a Grantham & District Hosptial User on 12/04/2014

Clinical Investigations