Main

Obesity often occurs early in the course of DMD(1). In the French DMD population the prevalence of obesity exceeds 50% at the age of 10(2). More than a cosmetic problem, it increases the handicap of muscle wasting, accentuates skeletal deformities, and complicates surgical outcome after orthopedic surgery(2). It has been shown that weight control lessens the burden of weakened muscles and improves mobility in these patients(1).

Little is known about the genesis of obesity in DMD. Obesity can be attributed to an imbalance between energy expenditure and intake(3, 4). The daily energy expenditure is likely to be reduced in DMD because 1) the dramatic muscle mass loss leads to a tremendous loss of metabolically active tissue and should decrease the REE and2) muscle weakness and the loss of ambulation presumably reduce the part of the daily energy expenditure related to activity. Overfeeding could also lead to obesity, particularly if the energy expenditure is low. Additionally, a decreased fat utilization is known to be a risk factor for the development of obesity(5) in non-DMD dMD subjects, and for weight regain after a weight loss regimen(6), and could also occur in DMD.

The purpose of this study was to test the hypothesis that muscle loss is associated with a decrease in REE in DMD. A positive result may lead to the evaluation of low REE as a risk factor in developing obesity in DMD.

METHODS

We studied REE, body composition, muscle mass, and daily energy intake in 13 DMD boys aged 8 to 13 y and 9 male age-matched controls (seeTable 1). Five DMD boys had a weight-for-age over 120% according to Edward's charts and were considered obese (ODMD)(1), the other boys were nonobese (NODMD). These DMD-specific charts adjust weight for the muscle loss, and their accuracy to define obesity was recently demonstrated(2). All DMD children but two had lost ambulation. They were regularly followed in an outpatient center, and none was institutionalized. Most of the controls were children of the members of the Pediatric Staff, without any known disease. The day of the study, none of the subjects was taking any medication, or had any intercurrent illness. All of the subjects and their parents gave informed written consent. The protocol was approved by the Lille University Hospital Ethics Committee.

Table 1 Subjects' characteristics
Full size table

Protocol. The usual energy intakes of our subjects were evaluated with a 7-d diet record, 2 wk before the study. For this purpose, their family was given a booklet the day of the prestudy visit that included a page for each day and a writing space for each meal (type of food and quantity). Quantities were evaluated according to instructions written on the front page of the booklet, and explained the day of the prestudy.

A flesh-free diet without protein restriction was started 6 d before the study, and daily urine samples were collected during the three last days for muscle mass estimation.

The night before the study, each subject ate dinner at home around 2000 h and remained fasting until completion of the study at noon the next day. Only water was allowed during this period.

REE was measured for 2 h, from 0900 to 1100 h, using an open circuit indirect calorimeter with a ventilated hood (SensorMedics, Yorba Linda, CA). During the test, the subjects were lying on their back and were asked to stay as relaxed as possible and awake. All events occurring during the test were recorded on a log book to exclude the nonsteady-state periods from the calculations. Watching television was allowed during the test.

Resistance (ohms) and reactance (ohms) were measured with a bioelectrical impedance analyzer (BIA 101, RJL systems, Detroit, MI) with four surface self-adhesive spot electrodes placed on the right hand and foot, and with a standard conduction current of 800 μA and 50 kHz.

Analytical methods. REE (kcal·h-1) was obtained from the mean of oxygen consumption (Vo2 in L·min-1) and carbon dioxide production (Vco2 in L·min-1) at steady state during the 2-h collecting time using the Weir formula without protein correction(7): REE (kcal·h-1) = [3.9× Vo2 + 1.1 × Vco2] × 60. Steady state is defined as a coefficient of variation less than 10% for both Vo2 and dilution air flow, and less than 5% for the RQ.

FFM was estimated with a bioelectrical impedance analyzer using Schaeffer's formula(8): FFM (kg) = 0.79 ×(H2/R) + 2.84 with H for height in centimeters and R for resistance in ohms. Because no body composition method has been validated in DMD, the FFM estimates are only approximate values. Muscle mass was estimated from 3-d creatinine excretion in urine assuming 1 mmol of urinary creatinine = 2.2625 kg of muscle(9). The dieticians reviewed the food records with the mothers the day of the study and analyzed them for daily energy intake and diet composition using the French Federation for Nutrition tables(10).

Statistical analysis. All results are means ± SEM. Observed means were compared using analysis of variance, observed differences were tested for statistical significance using the Scheffé F test. Significance was established at p < 0.05.

RESULTS

Patients' characteristics and body composition. Age and height did not differ between groups. Weight and weight for age ratios were significantly higher in the ODMD group than in the NODMD and control groups(Table 1 and 2). The estimated FFM was lower in the NODMD and in the ODMD groups than in controls (18.8 and 19.0 versus 25.3 kg, p < 0.05, respectively) (Table 2). FFM was not different between the two DMD groups (Table 2).

Table 2 Estimated body composition in DMD
Full size table

Indirect calorimetry. REE was 13% lower in the NODMD group than in the control group (47.5 versus 54.6 kcal·h-1,p < 0.05) (Table 3). REE was not different between the ODMD and either NODMD or control groups (Table 3). The difference between the control and NODMD groups disappeared when the REE was expressed per kg of body weight or FFM (Table 3). When expressed per kg of body weight, the REE became lower in the ODMD than in either NODMD or control groups (1.22 versus 1.85 and 1.83 kcal·kg-1·h-1, p < 0.05, respectively) (Table 3). Expressed per kg of FFM, REE became higher in the ODMD than in the control group (2.71 versus 2.19 kcal·kg-1·h-1, p < 0.05)(Table 3).

Table 3 Resting energy expenditure in DMD
Full size table

Postabsorptive RQ appeared higher in both DMD groups than in the control group; however, the difference was significant only for the NODMD group (0.88versus 0.83, p < 0.05) (Table 4). ODMD and NODMD RQ were not different (Table 4).

Table 4 Energy substrate oxidation in DMD
Full size table

Vo2 was higher in the control group than in the NODMD group (189versus 162 mL·min-1, p < 0.05)(Table 4). No differences in Vco2 were observed among the three groups (Table 4).

Urine collection. Daily creatinine excretion in urine was dramatically reduced in DMD patients, whether obese (179 ± 22 mg·d-1) or not (159 ± 18 mg·d-1), when compared with controls (594 ± 78 mg·d-1) (p< 0.05). Muscle mass estimated from the creatinine excretion in urine was 71% lower in the NODMD and ODMD groups than in controls (3.2 and 3.6versus 11.9 kg, p < 0.05, respectively)(Table 2).

No differences were observed in urea excretion in the urine between controls (11.1 ± 0.8 g·d-1), the NODMD group (9.4± 1.1 g·d-1) and the ODMD group (10.5 ± 0.7 g·d-1).

Diet records (Table 5). The observed energy and carbohydrate and fat daily intakes were significantly lower in the ODMD than in the control group (1343 versus 2038 kcal·d-1, 161 versus 250 g·d-1, and 57 versus 85 g·d-1, all p < 0.05, respectively)(Table 5).

Table 5 Energy intake and food composition in DMD
Full size table

DISCUSSION

In this study the 71% muscle mass loss observed in DMD boys aged 10 y was associated with a 13% decline in REE. Fat utilization in the postabsorptive state was decreased in the NODMD group, and obesity was not associated with an increased fat oxidation in the ODMD group. Low REE and impaired fat oxidation might contribute to the development of obesity in DMD children. More research is needed to test these hypotheses.

Because skeletal muscle is metabolically active and a major determinant of energy expenditure(11), the dramatic loss of muscle observed in DMD should decrease REE. In this study, a 71% muscle loss in a group of NODMD boys aged 10 y was associated with a 13% decrease in REE (47.5versus 54.6 kcal·h-1, p < 0.05). DMD is not known to affect the other metabolically active organs; therefore the NODMD and control groups differed only in their muscle mass and that may account for the decreased REE.

The 71% muscle mass reduction in DMD boys aged 10 y is consistent with previous studies(12–14). Muscle mass represents 20-30% of REE in adults(11, 15) and probably less in children because the relative mass of the muscle is lower in proportion of body weight(15). Assuming the contribution of skeletal muscle to REE is only 15 to 20% in children, a 75% reduction of muscle mass should lead to a REE decrease between 10 and 15%, which is consistent with our data. Considering DMD as a pathologic model of muscle reduction, the drecreased REE observed in DMD boys strongly supports the hypothesis that muscle mass is a major determinant in REE(11).

The muscle loss observed in DMD should decrease the metabolically active mass, a major determinant of REE(11, 16). Body weight may be used as an index of metabolically active mass to adjust and compare REE in subjects with a same body composition. When expressed per kg of body weight, the difference in REE between the NODMD and the control group disappears. If the assumption that body composition differs only slightly between these two groups is true, then the lower REE in NODMD might be attributed to their lower metabolically active mass. However, expressed per kg of body weight, the REE of the ODMD becomes lower than the two other groups. The dilution of the metabolically active mass by adipose tissue, which has a low O2 consumption(15), explains the low REE values obtained in the ODMD group. To take into account body composition when estimating the metabolically active mass, it has been proposed to adjust REE for FFM(16). REE expressed per kg of FFM, is higher in DMD boys than in controls; however, the difference is significant only for the obese DMD group. This might reflect a greater proportion of organ mass or metabolically active mass to FFM or an increased metabolism in these tissues. FFM might also explain the difference between the REE of controls and nonobese DMD children. However, FFM estimates should be analyzed with extreme caution in DMD. Although FFM is likely decreased in DMD(13), to date no method for FFM determination has been validated in DMD. FFM estimates in the DMD group should therefore be considered only as approximate values. Because muscle represents 50% of FFM in our control group as in adults(15), a 71% reduction of muscle mass should lead to 35% reduction of FFM. In contrast, NODMD FFM was decreased by only 24%. This discrepancy might reflect the replacement of the muscle by fibrous tissue(13) or the failure of the method used in determining body composition in the DMD population. Although consistent with the lower REE observed in DMD boys, the predictive value of FFM on REE in DMD remains to be evaluated by further studies.

In contrast with previous studies in non-DMD obese children(17, 18) and adults(19), the REE in the ODMD group was not found higher than in the NODMD group. In non-DMD obese people, the higher REE is attributed to the simultaneous increase of FFM with fat mass(3). In the present study, neither muscle mass nor FFM estimates were found to be increased in the ODMD group when compared with the NODMD group; this might reflect the impaired protein synthesis in DMD(20).

A low daily energy expenditure is known to be risk factor for developing obesity in adults(4) as well as in children(21). In our study, a 2-y follow-up showed that four out of five NODMD had become overweight (mean weight for age ratio: 161%). The low REE associated with muscle mass loss as well as a low level of spontaneous activity related to the handicap might lead to a low daily energy expenditure and then constitute risk factors to develop obesity in DMD. The present data warrant further studies to test the role of a low REE in the genesis of obesity in DMD.

The RQ is an index of glucose to fat oxidation ratio. RQ appeared higher in both DMD groups than in the controls; however, the difference reached statistical significance only for NODMD (0.88 versus 0.83,p < 0.05, respectively).

These results are unlikely to be attributable to the pulmonary complications of the disease. Daytime blood gas tension is normal in mildly affected patients, and respiratory complications occur in patients older than those in our study(22). Moreover, hypercapnia is unusual, and Vco2 was not decreased in the DMD groups. Differences in diet composition may affect slightly postabsorptive RQ; however, dietary assessment did not allow us to detect any differences between the ODMD and the control groups. In addition, urea excretion into the urine, an index of protein oxidation, was not different between the NODMD and the control groups, suggesting that the higher RQ cannot be attributed to a tremendously increased protein oxidation. Finally, transient ventilation changes might affect RQ through Vco2 if the measurement does not encompass these periods. However, a careful analysis of raw data and the use of steady state values may minimize this bias.

Because the measurements were obtained in the postabsorptive state(requiring at least a 12-h fast), the measured lipid oxidation should be higher than in any other period of the day. These results suggest a low lipid utilization in the NODMD group. Moreover, contrary to observations in obese children and adults(18, 23), the RQ were not lower in the ODMD group than in the NODMD group (Table 4), suggesting that the increase in lipid stores does not enhance lipid oxidation in DMD. Thus, a low lipid utilization might be associated with DMD at an early stage of the disease.

A high 24-h RQ was shown to be a risk factor for the development of obesity in a longitudinal study on a population of Pima American Indians with a 3-y follow-up(5). Postabsorptive RQ was also shown to be a good predictor for weight regain after a weight loss regimen in women(6). However, more data involving RQ measurements over a longer period of time are needed to corroborate a low lipid utilization and to test its eventual role in the development of obesity in DMD.

Observed energy and carbohydrate and fat daily intakes were lower in the ODMD group than in the controls. Most dietary studies have failed to show a higher energy intake in obese children and adults(24). The comparison between daily energy expenditure measured with doubly labeled water and observed energy intakes suggests that obese people underreport their intakes(25). Although the present study is consistent with this behavior, underreporting cannot be demonstrated. Diet records rely on information provided by the subject himself (or as in this study, by his parents or persons in charge of the child during school hours), and therefore their reliability is affected, especially for quantities. Although averaged over a 7-d period, diet records concerning DMD children should be analyzed with caution in the present study.

In summary, this study shows that 1) muscle loss is associated with a low REE in NODMD boys and 2) fat utilization might be low in the postabsorptive state in DMD boys. However, further data are needed to test the role of the low REE and high postabsorptive RQ in the genesis of obesity in DMD. We observed no significant muscle mass gain from being obese for these boys; therefore, obesity should be considered only as a factor increasing the handicap. Dietary counseling providing a normal balance of each nutrient, avoiding an excessive fat intake, and an energy intake adapted to their body composition might be useful to prevent obesity in DMD.