Sarcopenia refers to the progressive decline in muscle mass, strength (force-generating capacity), and quality that occurs with aging (1) and is related to physical disability, impairments in function and mobility, and frailty in older adults. (2-4) In a large epidemiological study, (2) it was estimated

that approximately 50% to 53% of people aged 60 years or older had sarcopenia, defined as a skeletal muscle mass index that was 1 standard deviation or more below the sex-specific index for young adults. The economic impact of sarcopenia in the United States in the year 2000 was estimated to be $18.5 billion, accounting for 1.5% of total health care costs (5) and signifying the importance of this condition in the general population.

As well as in aging, the loss of muscle mass is a common finding in chronic conditions such as cancer, (6) HIV infection, (7) chronic heart failure, (8) and chronic obstructive pulmonary disease (COPD) (9) and is related to exercise limitations, quality of life, and mortality. 7,8,10) Because of the widespread prevalence of muscle atrophy and its impact on the health and independence of aging adults and people with chronic conditions, it is essential that physical therapists accurately assess and appropriately manage muscle atrophy.

There are several methods for quantifying muscle mass; these include dual x-ray absorptiometry, bioelectrical impedance, ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI). Dual x-ray absorptiometry and bioelectrical impedance provide estimates of fat-free mass for the whole body or body segments, (11,12) whereas ultrasound has been used to measure muscle thickness and examine muscle architecture. (13) For examining the relationships between muscle mass and muscle strength or for determining the effectiveness of interventions in improving muscle mass, more specific measurements of muscle size, such as the cross-sectional area (CSA) and volume of individual muscles, are necessary.

Muscle volume and CSA can be obtained from CT and MRI scans and are commonly used in research settings. Muscle volume is the preferred measure of muscle size because it allows the entire muscle to be captured from multiple images and, therefore, takes into account the entire size of the muscle, regardless of its shape. (14) However, to obtain measurements of muscle volume, multiple images along the length of a limb are required, resulting in increases in the time required for and the cost of scanning and analysis. In addition, with CT, longer scan times result in greater exposure to radiation, thereby increasing the risk to the patient. (15) To reduce the time and cost of imaging and analysis, a single cross-sectional slice is commonly used as an indicator of total muscle size. For the lower limbs, the midthigh CSA is commonly used as an indicator of the muscle size of the thigh. (4,10,16) However, it has not been clearly delineated whether the midthigh is the best representative level for all muscle groups of the thigh, such as the quadriceps femoris, hamstring, and adductor muscle groups, which show differences in anatomical distributions along the length of the thigh. (17)

Because of the cost of and the technical expertise required for the use of the instrumentation mentioned above, physical therapists working in typical clinical settings do not have access to these specific measurements of muscle CSA or volume. Anthropometric measurements (ie, limb circumference and skinfold thickness) often are used to estimate muscle mass with reference equations (18); alternatively, limb circumference alone is used as an estimation of the size of the underlying muscle. However, there are many assumptions underlying the use of anthropometric measurements to estimate muscle mass. For example, it is assumed that the muscle has a circular shape and that subcutaneous tissue is uniformly distributed around the circumference of the limb and represents all of the adipose tissue in the section. (19,20) These assumptions may not hold true, especially in older adults or people with chronic conditions who experience muscle atrophy, such as people with COPD.

The purpose of this study was to determine the relationships among muscle volume, muscle CSA, and thigh circumference measurements taken at different levels of the thigh. In addition, we examined whether a published reference equation for estimating midthigh muscle CSA provides a reasonable estimation of the muscle CSA obtained from MRI in older adults who are healthy and in people with COPD.

Method

Subjects

Twenty people with COPD (9 men and 11 women) and 20 adults who were healthy and who were matched for sex, age, and body mass index (BMI) to the people with COPD volunteered to participate in this study. People with COPD were included in the study if they had moderate to severe (stage II or III) disease, according to Global Initiative for Obstructive Lung Disease guidelines (21) (forced expiratory volume in 1 second [[FEV.sub.1]] <80% of that predicted and FEV1/forced vital capacity [FVC] <70%). People with COPD were excluded if they had an acute exacerbation of COPD (defined as a worsening in condition from the typical daily state that was acute in onset and necessitated a change in medications), (22) had taken oral corticosteroids (eg, prednisone) in the 6 months prior to the study, or had participated in a formal exercise rehabilitation program in the 1-year period immediately prior to the study.

None of the participants were currently smoking. Participants were asked to complete a health risk screening questionnaire prior to inclusion in the study. (23) Those who indicated the presence of comorbid cardiovascular disease (eg, heart failure, previous myocardial infarction, or cardiovascular surgery), neurological conditions (eg, stroke or Parkinson disease), or lower-extremity musculoskeletal problems (knee or hip injury or arthritis) were excluded from the study. Each participant provided written informed consent prior to participation.

Height and weight were measured with shoes off and light clothing. Participants underwent spirometry according to the standards described by the American Thoracic Society (24) to measure [FEV.sub.1] and FVC to define the presence and severity of COPD.

MRI

Magnetic resonance imaging was used to determine the muscle CSA and muscle volume of the quadriceps femoris, hamstring, and adductor muscles. A 1.5-T MRI scanner * was used to acquire 5-mm axial contiguous slices from the anterior superior iliac spine to the femoral-tibial joint line on both thighs. The images were T1 weighted (echo time = 8 milliseconds; repetition time=650 milliseconds), with a 40-[cm.sup.2] field of view and a matrix of 512X384 pixels (in-plane spatial resolution=0.78x1.78 mm). The MRI scan resulted in approximately 100 slices for each participant.

For each participant, 2 landmarks were identified on the MRI scan: 2 cm below the anterior inferior iliac spine (proximal slice) and the superior aspect of the patella (distal slice). The distance between the proximal and distal slices was divided equally to select 17 slices from the MRI scan. The muscle CSA and volume of the quadriceps femoris, hamstring, and adductor muscles were determined from measurements taken on these 17 selected slices.

[FIGURE 1 OMITTED]

Muscle CSA and Volume

NIH Image, version 1.31, ([dagger]) was used to manually outline each of the quadriceps femoris (rectus femoris), vastus (vastus lateralis, vastus intermedius, and vastus medialis), hamstring (semitendinosus, semimembranosus, and biceps femoris long head and short head), and adductor (pectineus, adductor longus, adductor magnus, and adductor brevis) muscles on the 17 selected slices. Individual muscles are shown on 3 axial images in Figure 1.

The total volume of each muscle (cubic centimeters) was calculated by summing the product of the measured muscle CSA and the slice thickness for all 17 slices, and the volume of the gaps between the slices was estimated with the truncated cone formula, as follows (25):

Volume of gap = 1/3 x [CSA1 +CSA2

+ [square root of (CSA1 x CSA2)]] x gap thickness,

where CSA1 is the CSA of the slice immediately above the gap and CSA2 is the CSA of the slice immediately below the gap. The gap thickness between measured slices was 2.0 to 2.5 cm. This method for estimating muscle volume was previously validated. (14) The interrater reliability of CSA measurements established for 65 regions of interest by 2 raters showed an average percent error between the 2 raters of 0.4%, with an r value of .99.

The CSAs measured for the quadriceps femoris, hamstring, and adductor muscles on 3 of the 17 slices, located at 30%, 50%, and 80% of the thigh length (Fig. 1), were used for further muscle volume comparisons. The 50% slice, or midthigh level, was chosen because it is commonly used in the literature to quantify thigh muscle CSA. The distal slice, closer to the knee (80% of the thigh length), was chosen for the hamstring muscles because this was the level at which the hamstring muscle CSAs were the largest in our subjects. The proximal slice (30% of the thigh length) was the slice at which the CSAs of the adductor muscle group tended to be the largest.

Thigh Circumference

Thigh circumference was measured by outlining the outside border of the thigh on the MR images at 5, 10, and 15 cm above the superior aspect of the patella by using NIH Image. To ensure the validity of circumference measurements obtained from MR images, a tape measure was used to compare these measurements with the measurements obtained in vivo from 10 of the subjects who were healthy. Measurements were obtained with the participant lying supine with the knee in slight flexion and supported by a rolled towel. Ink marks were made at 5 and 10 cm above the superior border of the patella, and a steel tape measure was placed with the lower border level with the mark. Three recordings were obtained at each distance, with the tape measure being removed each time, and an average was calculated. The tape measure was pulled taut so that no buckles occurred, but not so tight that it caused an indentation in the skin. The average percent errors between thigh circumference measurements obtained from MR images and those obtained with the tape measure were 3.2% at 5 cm and 4.2% at 10 cm.

Estimation of Muscle CSA From Anthropometric Measurements

Measurements of thigh circumference and skinfold thickness were made on all participants with the protocol established by Housh et al. (18) This protocol was chosen because it allows the prediction of quadriceps femoris and hamstring muscle CSAs individually rather than predicting the total thigh muscle CSA (26) and because it was developed from muscle measurements obtained by MRI. In short, thigh circumference (tape measure) and skinfold thickness (Harpenden skinfold calipers ([double dagger]) were measured 3 times each at the point midway between the inguinal crease and the lateral femoral condyle. The average thigh circumference and the average skinfold thickness measurements were entered into the following regression equations developed by Housh et a1 (18) to estimate the CSAs of the quadriceps femoris and hamstring muscles at 50% of the thigh length:

Quadriceps muscle CSA= (2.52 x midthigh circumference [cm]) -(1.25 x anterior thigh skinfold thickness [mm])-45.13

and

Hamstring muscle CSA = (1.08 x midthigh circumference [cm])-(0.64 x anterior thigh skinfold thickness [mm])-22.69.

Data Analysis

Data analysis was done with the Statistical Package for the Social Sciences, version 13.0. ([section]) Descriptive statistics (mean and standard deviation) were used to describe the sample characteristics and muscle volumes. Correlations were determined for BMI, muscle CSA, thigh circumference, and muscle volume as Pearson product moment correlation coefficients. The following descriptors were used to assess the strength of the correlations: low=.26-.49, moderate=.50-.69, high=.70-.89, and very high=.90-1.00. (27) The Bland-Altman procedure was used to compare the anthropometric estimations of quadriceps femoris and hamstring muscle CSAs (ie, the alternative measurements) with the MRI measurements of the CSAs (ie, the criterion measurements). The Bland-Altman plot displayed the difference between the alternative and the criterion measurements on the y-axis against the average of the 2 measurements on the x-axis. (28) The limits of agreement (ie, mean [+ or -] 2 standard deviations) were constructed around the mean difference between the 2 measurements. If the mean difference between the 2 measurements was different from 0, then there was either underestimation or overestimation of the criterion measurement. (28)

[FIGURE 2 OMITTED]

Results

The characteristics of older adults who were healthy and people with COPD are provided in Table 1. The groups were well matched for sex, age, and BMI. People with COPD were classified as having moderate to severe disease on the basis of their lung function according to international guidelines for the diagnosis of COPD. (21) The volumes of the quadriceps femoris, hamstring, and adductor muscles are shown in Table 1. People with COPD had significantly smaller muscle volumes compared with the older adults who were healthy for all 3 muscle groups (P<.01).

Significant differences in muscle CSAs were observed at specific levels, depending on the muscle group (Tab. 1). The adductor and hamstring muscle CSAs were significantly different between groups only at the proximal (30%) and distal (80%) landmarks, at which the CSAs were the largest, respectively. The quadriceps femoris muscle CSA was significantly smaller in people with COPD at both the midthigh and the distal landmarks.

The midthigh CSA was estimated from anthropometrie measurements (thigh circumference and skinfold thickness) with a published equation. (18) The mean quadriceps femoris and hamstring muscle CSAs estimated with the published equation were 64.9 [+ or -] 17.8 and 23.3 [+ or -] 5.6 [cm.sup.2], respectively, for the people with COPD and 56.7 [+ or -] 36.8 and 19.5 [+ or -] 16.9 [cm.sup.2], respectively, for the older adults who were healthy. These values were not significantly different (Tab. 1), indicating that the equation was unable to detect a smaller muscle size in people with COPD than in the older adults who were healthy. Similarly, measurements of the thigh circumference at 5, 10, and 15 cm were not different between the groups (Tab. 1).

Relationships Among BMI, Muscle Volume, and Muscle CSA

Low to moderate correlations were found between BMI and specific measurements of muscle size. In people with COPD, significant correlations were found between BMI and the volume of the hamstring (r=.48, P= .03) and adductor (r= .49, P=.03) muscles (Tab. 2). Moderate correlations (r [greater than or equal to] .50) were found between BMI and the muscle CSA at various levels: quadriceps femoris muscle at 80% (r=.52, P= .02), hamstring muscle at 50% (r=.56, P=.01), hamstring muscle at 80% (r= .51, P= .02), and adductor muscle at 50% (r=.58, P= .008).

In older adults who were healthy, the volumes of the quadriceps femoris and hamstring muscles were significantly correlated with BMI (r=.52, P=.02, and r=.50, P=.03, respectively). Moderate correlations were also seen between BMI and the quadriceps femoris and hamstring muscle CSAs at 30'% and 80% as well as the adductor muscle CSA at 30% only (Tab. 2).

Relationships Between Muscle CSA and Muscle Volume

The majority of the correlations between muscle CSA and muscle volume were moderate to very high, with the exception of the adductor muscles at 80% for both groups and the hamstring muscles at 30% only for the people with COPD (Tab. 3). When the entire sample of subjects was considered, the quadriceps femoris and hamstring muscles showed significant correlations between muscle volume and muscle CSA at all 3 levels (Tab. 3). The adductor muscle volume was significantly correlated with the adductor muscle CSA at 30% (close to the hip) and 50% (midthigh) but not at 80% (close to the knee). Representative scatter-plots of the correlations between muscle volume and muscle CSA at 50% are shown in Figure 2.

Relationships Between Thigh Circumference and Muscle Volume

Low correlations (r=.14-.47) were observed for thigh circumference and total muscle volume both in the older adults who were healthy and in the people with COPD (Tab. 4). Representative scatterplots showing the relationship between muscle volume and thigh circumference at 5, 10, and 15 cm are shown in Figure 3.

Estimation of Quadriceps Femoris and Hamstring Muscle Midthigh CSAs

With the Bland-Altman procedure, the level of agreement between the estimation of quadriceps femoris and hamstring muscle CSAs with anthropometric measurements and measurements obtained by MRI was assessed (Fig. 4). The mean difference between the measurements, as seen on the y-axis, was negative for both the quadriceps femoris and hamstring muscles, indicating that the equation tended to overestimate the actual muscle CSA obtained by MRI. One woman with COPD had quadriceps femoris and hamstring muscle CSAs that were larger than those of others in the COPD group; the difference between the measurements was greater than the mean difference for the group minus 2 standard deviations.

[FIGURE 3 OMITTED]

Discussion

Measurement of muscle size is an essential aspect of the clinical assessment of older adults, who may be experiencing sarcopenia, and people with chronic conditions that result in muscle atrophy, such as people with COPD. The results of our study show that anthropometric measurements, such as thigh circumference and skinfold thickness, do not provide an adequate representation of muscle size in older adults or people with COPD. With MRI, muscle CSA measurements obtained from a single slice are representative of muscle volume; however, the most representative slice varies depending on the muscle group of interest. Therefore, the level at which the CSA is measured should be chosen on the basis of which muscle group is being quantified.

The midthigh CSA often is used to represent the size of the thigh muscles (29) or to calculate the specific torque (ie, torque/CSA) of a muscle group, such as the quadriceps femoris muscle group. (30) However, the relationship between muscle CSA and muscle volume had not previously been examined in older adults or people with COPD. Muscle CSA and muscle volume were closely correlated; this result was expected because multiple CSAs were used to calculate the total muscle volume. Therefore, the correlation between volume and CSA is circular in nature because volume measurements are dependent on CSA measurements. For the quadriceps femoris and hamstring muscles, there were strong correlations between the CSAs at all 3 levels and muscle volume. However, the adductor muscle volume was most strongly correlated with the CSAs at the proximal slices (30% and 50%) and showed a low correlation at the distal slice, at which some people did not show the presence of any adductor muscles. Therefore, if one is interested in the size of a specific muscle group on a single image, it is necessary to choose the level at which the CSA is maximized for that particular muscle group.

[FIGURE 4 OMITTED]

Body mass index is considered to be a general measure of body composition, incorporating both fat mass and lean mass in relation to a person's stature. (31) In the present study, we found moderate correlations between BMI and some measurements of muscle volume and CSA; however, no difference in BMI was found between people with COPD and controls, despite the presence of muscle atrophy in people with COPD. It was previously shown that up to 11% of men and 24% of women who have COPD and who have a BMI within the normal range actually have a low lean body mass. (9) In addition, BMI does not appear to be an adequate marker of changes in body composition in older adults because the decrease in lean muscle mass with age can be countered by an increase in fat mass, resulting in a stable BMI over time. (31) Therefore, more specific measurements of muscle size are needed to determine the presence of muscle atrophy or sarcopenia in these populations.

Magnetic resonance imaging and CT are costly techniques that require specialized equipment, technical support, and software for analysis. (15) In the clinical setting, it is important to have low-cost, efficient, and accurate measurements that can easily be applied during an assessment. Anthropometric measurements are ideal for this situation; however, they may not provide adequate estimates of muscle size in older adults or people with COPD. We chose to use a published equation that was developed from young men to attempt to estimate midthigh CSA (18) because no reference equations have been developed for older adults or people with COPD. The equation developed by Housh et al (18) was appealing because it allowed for individual estimates of quadriceps femoris and hamstring muscle midthigh CSAs rather than only a general estimate of thigh muscle CSA.

In our sample, the equation tended to overestimate the CSAs of both the quadriceps femoris and the hamstring muscles, to such an extent that it was unable to detect a difference in muscle size between adults who are healthy and people with COPD. This finding is similar to the findings of previous studies that compared estimates made from MRI or CT with anthropometric estimates in middle-age adults, (19) people with various levels of adiposity, (20) and people who had COPD and muscle atrophy. (10) Prediction equations, such as that described by Housh et al, (18) are typically developed in young men who are healthy and are based on several assumptions about the shape of the muscle and the distribution of adipose tissue that may not hold true as people age or under conditions in which muscle atrophy occurs, such as COPD. For example, in a young person, the majority of adipose tissue in the thigh lies within the subcutaneous tissue compartment, whereas in older adults and people who have atrophy, there is a greater distribution of intermuscular fat (4) that is not captured in measurements of limb circumference or skinfold thickness. Furthermore, we have found that people with COPD have greater intramuscular fat infiltration, as seen on T1-weighted MRI, compared with matched controls who are healthy. (32) This feature can lead to an overestimation of muscle CSA by the prediction equation because thigh circumference and skinfold thickness do not properly reflect the amount or location of adipose tissue and the size of the underlying muscle. The development of population-specific reference equations for older adults who are healthy and people with COPD would benefit clinicians in assessing muscle CSA with a simple, but accurate, method.

Thigh circumference is another measurement that can be easily applied in the clinical setting and that is often used by clinicians to assess muscle atrophy or changes in muscle size with training. From the present study, however, it does not appear that thigh circumference provides a sensitive measure of muscle size because it was not able to distinguish muscle size in people with COPD from that in controls who are healthy. Similarly, Arangio et a1 (33) found that in people who had undergone anterior cruciate ligament reconstruction and experienced thigh muscle atrophy, thigh circumference showed a difference of only 1.8% between the injured (atrophied) leg and the noninjured leg, whereas muscle CSA obtained by MR/ showed a difference of 8.6%. Furthermore, we observed only a single significant correlation: between thigh circumference and muscle volume (ie, thigh circumference at 10 cm in the sample of people with COPD). Therefore, it appears that thigh circumference is not a good indicator of muscle size and should not be used alone as a clinical measure of muscle atrophy.

Conclusion

Muscle CSA at the midthigh showed moderate to high correlations with the volume of the quadriceps femoris, hamstring, and adductor muscles both in older adults who were healthy and in people with COPD. However, if a single slice is going to be used to represent a specific thigh muscle group, then a location at which the muscle CSA is at its largest should be considered. Doing so may be particularly important if comparisons of muscle size are going to be made before and after an intervention. Anthropometric measurements, such as thigh circumference and skinfold thickness, are easily applied in the clinical setting but do not provide an adequate representation of muscle size in older adults or in people who have muscle atrophy (ie, people with COPD). Specific reference equations need to be developed for these populations so that muscle size can be assessed simply but accurately with anthropometric measurements in a clinical setting.

Dr Mathur and Dr Reid provided concept/ idea/research design. Dr Mathur and Ms Takai provided writing. Dr Mathur provided data collection. All authors provided data analysis. Dr Mathur, Dr Maclntyre, and Dr Reid provided project management. Dr Maclntyre and Dr Reid provided facilities/ equipment. Dr Reid provided institutional liaisons. Ms Takai, Dr Maclntyre, and Dr Reid provided consultation (including review of manuscript before submission).

The Clinical Research Ethics Board of the University of British Columbia approved the study.

This article was submitted February 12, 2007, and was accepted September 17, 2007.

DOI: 10.2522/ptj.20070052

References

(1) Marzetti E, Leeuwenburgh C. Skeletal muscle apoptosis, sarcopenia and frailty at old age. Exp Gerontol. 2006;41:1234-1238.

(2) Janssen I, Heymsfield SB, Ross R. Low relative skeletal muscle mass (sarcopenia) in older persons is associated with functional impairment and physical disability. J Am Geriatr Soc. 2002;50:889-896.

(3) Roubenoff R. Sarcopenia: a major modifiable cause of frailty in the elderly. J Nutr Health Aging. 2000;4:140-142.

(4) Visser M, Goodpaster BH, Kritchevsky SB, et al. Muscle mass, muscle strength, and muscle fat infiltration as predictors of incident mobility limitations in well-functioning older persons. J Gerontol A Biol Sci Med Sci. 2005;60:324-333.

(5) Janssen I, Shepard DS, Katzmarzyk PT, Roubenoff R. The healthcare costs of sarcopenia in the United States. J Am Geriatr Soc. Jan 2004;52(1):80-85.

(6) Argiles JM, Busquets S, Felipe A, Lopez-Soriano FJ. Muscle wasting in cancer and ageing: cachexia versus sarcopenia. Adv Gerontol. 2006;18:39-54.

(7) Roubenoff R. Acquired immunodeficiency syndrome wasting, functional performance, and quality of life. Am J Manag Care. 2000; 6:1003-1016.

(8) Piepoli MF, Kaczmarek A, Francis DP, et al. Reduced peripheral skeletal muscle mass and abnormal reflex physiology in chronic heart failure. Circulation. 2006; 114:126-134.

(9) Vermeeren MA, Creutzberg EC, Schols AM, et al. Prevalence of nutritional depletion in a large out-patient population of patients with COPD. Respir Med. 2006; 100:1349-1355.

(10) Marquis K, Debigare R, Lacasse Y, et al. Midthigh muscle cross-sectional area is a better predictor of mortality than body mass index in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2002; 166:809-813.

(11) Wang W, Wang Z, Faith MS, et al. Regional skeletal muscle measurement: evaluation of new dual-energy X-ray absorptiometry model. J Appl Physiol. 1999;87:1163-1171.

(12) Barbosa-Silva MC, Barros AJ. Bioelectrical impedance analysis in clinical practice: a new perspective on its use beyond body composition equations. Curr Opin Clin Nutr Metab Care. 2005;8:311-317.

(13) Sipila S, Suominen H. Ultrasound imaging of the quadriceps muscle in elderly athletes and untrained men. Muscle Nerve. 1991;14:527-533.

(14) Tracy BL, Ivey FM, Jeffrey Metter E, et al. A more efficient magnetic resonance imaging-based strategy for measuring quadriceps muscle volume. Med Sci Sports Exert: 2003;35:425-433.

(15) Ross R. Advances in the application of imaging methods in applied and clinical physiology. Acta Diabetol. 2003;40(suppl 1):S45-S50.

(16) Akima H, Kano Y, Enomoto Y, et al. Muscle function in 164 men and women aged 20-84 yr. Med Sci Sports Exerc. 2001;33: 220-226.

(17) Lajeunesse D, Edwards C, Grosenick B. Realism: A Study in Human Structural Anatomy. Red Deer, Alberta, Canada: Kasterstener Publications Inc; 2003.

(18) Housh DJ, Housh TJ, Weir JP, et al. Anthropometric estimation of thigh muscle cross-sectional area. Med Sci Sports Exerc. 1995; 27:784-791.

(19) Fuller NJ, Hardingham CR, Graves M, et al. Predicting composition of leg sections with anthropometry and bioelectrical impedance analysis, using magnetic resonance imaging as reference. Clin Sci (Lond). 1999;96:647-657.

(20) Tothill P, Stewart AD. Estimation of thigh muscle and adipose tissue volume using magnetic resonance imaging and anthropometry. J Sports Sci. 2002;20:563-576.

(21) Global Initiative for Obstructive Lung Disease. Global Strategy for the Diagnosis, Management and Prevention of Chronic Obstructive Pulmonary Disease. Bethesda, Md, and Geneva, Switzerland: National Heart, Lung, and Blood Institute/ World Health Organization; 2004:6-8.

(22) Rodriguez-Roisin R. Toward a consensus definition for COPD exacerbations. Chest. 2000;117(5 suppl 2):398S-401S.

(23) American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription. Philadelphia, Pa: Lippineott, Williams & Wilkins; 2000.

(24) American Thoracic Society/European Respiratory Society. Skeletal muscle dysfunction in chronic obstructive pulmonary disease; a statement of the American Thoracic Society and the European Respiratory Society. Am J Respir Crit Care Med. 1999; 159:S1-S40.

(25) Ross R, Rissanen J, Pedwell H, Clifford J, Shragge P. Influence of diet and exercise on skeletal muscle and visceral adipose tissue in men. J Appl Physiol. 1996;81: 2445-2455.

(26) Knapik JJ, Staab JS, Harman EA. Validity of an anthropometric estimate of thigh muscle cross-sectional area. Med Sci Sports Exerc. 1996;28:1523-1530.

(27) Munro BH. Correlation. 4th ed. Philadelphia, Pa: Lippincott; 2001.

(28) Bland JM, Airman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307-310.

(29) Visser M, Kritchevsky SB, Goodpaster BH, et al. Leg muscle mass and composition in relation to lower extremity performance in men and women aged 70 to 79: the Health, Aging and Body Composition Study. J Am Geriatr Soc. 2002;50:897-904.

(30) Bernard S, LeBlanc P, Whittom F, et al. Peripheral muscle weakness in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 1998; 158:629-634.

(31) Zamboni M, Zoico E, Scartezzini T, et al. Body composition changes in stable-weight elderly subjects: the effect of sex. Aging Clin Exp Res. 2003; 15:321-327.

(32) Mathur S, MacIntyre DL, Forster BB, et al. Preservation of eccentric torque of the knee extensors and flexors in people with COPD. J Cardiopulm Rehabil. In press.

(33) Arangio GA, Chen C, Kalady M, Reed JF III. Thigh muscle size and strength after anterior cruciate ligament reconstruction and rehabilitation. J Orthop Sports Phys Ther. 1997;26:238-243.

* General Electric, 2421 N Mayfair Rd, Milwaukee, WI 53266.

([dagger]) National Institutes of Health, 9000 Rockville Pike, Bethesda, MD 20892. Available at: http://rsb.info.nih.gov/ij/Java 1.3.1_03.

([double dagger]) Baty International, Victoria Rd, Burgess Hill, West Sussex RHI5 9LB, United Kingdom.

([section]) SPSS Inc, 233 S Wacker Drive, Chicago, IL 60606

S Mathur, BSc(PT), PhD, is a postdoctoral fellow at the University of Florida, Gainesville, Fla. She was a doctoral student in Human Kinetics at the University of British Columbia, Vancouver, British Columbia, Canada, at the time of the study. Dr Mathur's institutional mailing address is: Muscle Biophysics Lab, Vancouver Coastal Health Research Institute, Room 500-828 W 10th Ave, Vancouver, British Columbia, Canada V5Z 1L8. Address all correspondence to Dr Mathur at: sunitamathur@ phhp.ufl.edu.

KPR Takai, BSc(PT), MSc, is a clinical physical therapist. She was a graduate student in the School of Rehabilitation Sciences, University of British Columbia, at the time of the study.

DL MacIntyre, BSc(PT), PhD, is Associate Professor, Department of Physical Therapy, University of British Columbia, and Investigator, Rehabilitation Research Lab, GF Strong Rehabilitation Centre.

D Reid, BMR(PT), PhD, is Professor, Department of Physical Therapy, and Chair, Research Graduate Programs in Rehabilitation Sciences, University of British Columbia.

[Mathur S, Takai KPR, MacIntyre DL, Reid D. Estimation of thigh muscle mass with magnetic resonance imaging in older adults and people with chronic obstructive pulmonary disease. Phys Ther. 2008;88:219-230]

Table 1.
Characteristics of Older Adults Who Were Healthy and People
With Chronic Obstructive Pulmonary Disease (COPD) in the
Present Study (a)

Characteristic                       Value (b) for:

                                    Older Adults Who
                                  Were Healthy (n=20)

Age (y)                            64.4 [+ or -] 8.1
Sex (no. of women/men)             11/9
Height (m)                         1.67 [+ or -] 0.13
Weight (kg)                        69.0 [+ or -] 14.4
BMI (kg/[m.sup.2])                 24.3 [+ or -] 2.2
Lung function
  [FEV.sup.1], L (% predicted)     2.28 [+ or -] 0.72
                                    (81 [+ or -] 20)
  FVC, L (% predicted)             3.05 [+ or -] 1.11
                                    (83 [+ or -] 20)
  [FEV.sup.1]/FVC (%)                77 [+ or -] 9
Muscle volume ([cm.sup.3])
determined from MR]
  Quadriceps femoris            1,610.9 [+ or -] 511.7
  Hamstring                       613.8 [+ or -] 165.1
  Adductor                        775.7 [+ or -] 232.0
Muscle CSA ([cm.sup.2])
determined from MRI
at 30% (proximal)
  Quadriceps femoris               42.6 [+ or -] 11.4
  Hamstring                         4.7 [+ or -] l.8
  Adductor                         47.0 [+ or -] 10.4
Muscle CSA ([cm.sup.2])
determined from MRI at
50% (midthigh)
  Quadriceps femoris               56.8 [+ or -] 13.5
  Hamstring                        17.4 [+ or -] 3.9
  Adductor                         42.7 [+ or -] 11.5
Muscle CSA ([cm.sup.2])
determined from MRI
at 80% (distal)
  Quadriceps femoris               36.7 [+ or -] 9.1
  Hamstring                        29.8 [+ or -] 5.4
  Adductor                          4.3 [+ or -] 3.4
Muscle CSA ([cm.sup.2])
estimated from anthropometric
measurements
  Quadriceps femoris midthigh      65.0 [+ or -] 14.0
  Hamstring midthigh               23.3 [+ or -] 6.6
Thigh circumference at
indicated level (cm)
above the knee
  5                                40.4 [+ or -] 3.1
  10                               44.3 [+ or -] 2.7
  15                               47.9 [+ or -] 2.8

Characteristic                       Value (b) for:           P

                                      People With
                                      COPD (n=20)

Age (y)                            68.2 [+ or -] 10.0       .161
Sex (no. of women/men)             11/9
Height (m)                         1.66 [+ or -] 0.09       .715
Weight (kg)                        72.1 [+ or -] 14.6       .352
BMI (kg/[m.sup.2])                 26.6 [+ or -] 4.7        .110
Lung function
  [FEV.sup.1], L (% predicted)     1.34 [+ or -] 0.41     < .001
                                    (51 [+ or -] 17)
  FVC, L (% predicted)             2.58 [+ or -] 0.47     < .001
                                    (78 [+ or -] 14)
  [FEV.sup.1]/FVC (%)                52 [+ or -] 14       < .001
Muscle volume ([cm.sup.3])
determined from MR]
  Quadriceps femoris            1,214.7 [+ or -] 250.7      .004
  Hamstring                       483.1 [+ or -] 109.9      .005
  Adductor                        545.5 [+ or -] 167.6      .001
Muscle CSA ([cm.sup.2])
determined from MRI
at 30% (proximal)
  Quadriceps femoris               38.2 [+ or -] 7.3        .150
  Hamstring                        4.31 [+ or -] 1.2        .457
  Adductor                         37.4 [+ or -] 10.0       .005
Muscle CSA ([cm.sup.2])
determined from MRI at
50% (midthigh)
  Quadriceps femoris               48.3 [+ or -] 10.2       .011
  Hamstring                        16.3 [+ or -] 4.0        .396
  Adductor                         37.5 [+ or -] 11.3       .163
Muscle CSA ([cm.sup.2])
determined from MRI
at 80% (distal)
  Quadriceps femoris               30.4 [+ or -] 7.4        .022
  Hamstring                        22.9 [+ or -] 5.6        .027
  Adductor                          3.8 [+ or -] 3.7        .651
Muscle CSA ([cm.sup.2])
estimated from anthropometric
measurements
  Quadriceps femoris midthigh      64.9 [+ or -] 17.8       .980
  Hamstring midthigh               23.3 [+ or -] 7.8        .994
Thigh circumference at
indicated level (cm)
above the knee
  5                                40.7 [+ or -] 4.4        .843
  10                               44.1 [+ or -] 5.1        .866
  15                               49.1 [+ or -] 5.7        .451

(a) BMI=body mass index, [FEV.sub.1]=forced expiratory volume
in 1 second, FVC=forced vital capacity, MRI=magnetic resonance
imaging, CSA=cross-sectional area.

(b) Data are reported as mean [+ or -] SD unless otherwise
indicated.

Table 2.
Correlations (r) Among Body Mass Index (BMI),
Muscle Cross-Sectional Area (CSA), and Muscle Volume

                          Correlation for:

                                     People With
                           Older       Chronic
                          Adults     Obstructive
                          Who Were    Pulmonary     Entire
                          Healthy      Disease      Sample
Parameter                  (n=20)       (n=20)      (N=40)

Muscle volume
  Quadriceps femoris      .52 (a)    .40            .16
  Hamstring               .50 (a)    .48 (a)        .23
  Adductor                .28        .49 (a)        .19
Quadriceps femoris
muscle CSA at:
 30%                      .54 (a)    .40            .28
 50%                      .44        .50 (a)        .33
 80%                      .52 (a)    .52            .31
Hamstring muscle CSA at:
 30%                      .51 (a)    .44            .33
 50%                      .38        .56 (a)        .42 (a)
 80%                      .52        .51 (a)        .40 (a)
Adductor muscle CSA at:
 30%                      .55 (a)    .26            .15
 50%                      .26        .58 (a)        .36 (a)
 80%                      .17        .35            .17

(a) P [less than or equal] .05.

Table 3.
Correlations (r) Between Muscle Cross-Sectional
Area (CSA) and Muscle Volume

                     Correlation for:

                                   People With
Muscle                               Chronic
                     Older Adults  Obstructive
                      Who Were      Pulmonary    Entire
                      Healthy       Disease      Sample
                      (n = 20)      (n = 20)     (N=40)

Quadriceps femoris
muscle CSA at:
  30%                  .95 (a)       .92 (a)     .92 (a)
  50%                  .94 (a)       .91 (a)     .89 (a)
  80%                  .84 (a)       .82 (a)     .89 (a)

Hamstring
muscle CSA at:
  30%                  .67 (a)         .41       .58 (a)
  50%                    .61         .85 (a)     .68 (a)
  80%                  .75 (a)       .81 (a)     .75 (a)

Adductor
muscle CSA at:
  30%                    .52         .75 (a)     .69 (a)
  50%                  .71 (a)       .89 (a)     .77 (a)
  80%                    .26           .37         .30

(a) P [less than or equal to] .001.

Table 4.
Correlations (r) Between Thigh Circumference and Muscle Volume

         Correlation for:

                          People
                          With Chronic
Total    Older Adults     Obstructive      Entire
Volume   Who Were         Pulmonary        Sample
(cm)     Healthy (n=20)   Disease (n=20)   (N=40)

5        .43              .38              .31
10       .34              .47 (a)          .32
15       .19              .35              .14 (a)

(a) P < .05.