Resting energy expenditure in Duchenne patients using home mechanical ventilation

Study subjects

In total, 20 adult patients were studied (table 1⇓), on a consecutive basis, on the occasion of a routine follow-up. The diagnosis of DMD was based on clinical, familial, electromyographic, and genetic data. Of these, 13 patients received nocturnal-only MV (noninvasive in nine cases and via a tracheostomy in four cases). The remaining seven patients were continuously dependent on their ventilator (tracheostomy in all of these cases). All patients used volume-controlled MV as provided by an Eole3^TM ventilator (Saime, Savigny-le-Temple, France), either noninvasively or through a tracheotomy. All patients received nutritional intake orally and none had a gastrostomy. No build-up drinks or food supplements were used. They all reported eating slowly and with difficulty, but stated that these inconveniencies were largely compensated by the pleasure to eat normal meals with cooked dishes.

Exclusion criteria included a recent episode of acute infection (<6 weeks), an unstable cardiac condition, active gastrointestinal abnormalities, recent surgery, diabetes mellitus and thyroid disorders. No patients were taking corticosteroids.

The protocol was approved by the ethics committee of the Société de Reanimation de Langue Française, Paris, France.

Controls

Data obtained from 12 healthy subjects, recruited from the authors' own laboratory, served as controls for the validation of the measurement techniques.

Methods

The serum concentrations of albumin, pre-albumin and C-reactive protein were measured from venous blood samples.

The fat free mass (FFM) and the fat mass were determined using dual frequency (5 kHz and 100 kHz) bio-electrical-impedance analysis, (Analycor®; Eugédia-Spengler, Chambly, France) performed with the subjects in the supine position 16.

Resting energy expenditure (REE) was calculated from the results of indirect calorimetry and urinary nitrogen production, according to Ben-Porat et al. 17. Oxygen consumption (V′_O2) and carbon dioxide production were measured using an open-circuit metabolic cart (Sensormedics Horizon System; Sensormedics, Anaheim, CA, USA) and a ventilated canopy 18. The subjects were studied in the supine position, after becoming accustomed to the equipment, in a quiet room at 24°C. The measurements were performed between 08:00–09:00 h, after an overnight fast of ≥12 h. In the nocturnal-only ventilation patients, MV was interrupted ≥12 h before the measurements were made.

To estimate calorific intake, the patients were instructed to record precisely their food intake over 3 days during the week immediately preceding the study. The records were reviewed in detail with the assistance of food models 19. The daily energy and protein intakes were calculated, expressed in kJ·24 h⁻¹, and averaged over the 3 days.

Predicted values

The measured resting energy expenditure (mREE) was compared to the theoretical energy expenditure (tREE) as described by the Harris and Benedict 20 equation. The use of body weight in the Harris and Benedict equation is not appropriate for DMD patients due to their pathological changes in body composition. Therefore, FFM adjusted resting energy expenditure (fREE) was computed using the relationship observed between REE and FFM in the control subjects 18, 21, 22.

Analysis

All the variables are described as mean±sd. The relationship between REE and FM was assessed by Pearson correlation coefficients, after verifying that these variables followed a Gaussian distribution. Differences between correlation coefficients were tested for significance using the Fischer r-to-z transformation. Within group comparisons of observed versus predicted values were made using a paired t-test. Comparisons between the groups were made using an unpaired t-test. Statistical significance of differences was assumed when the risk of type I error was <5%.

Body weight, body composition and serum proteins

Compared with the 12 age-matched controls, DMD patients were smaller in height and had a lower body weight. The reduction in weight was disproportionately greater than the reduction in size, hence, a significantly smaller BMI. Most of the reduction in BMI was accounted for by a reduction in FFM (table 1⇑). All serum albumin values except one were >35 g·L⁻¹. All serum pre-albumin values were >0.2 g·L⁻¹, except in three cases, including one with low albumin. All serum C-reactive protein values were <5 mg·L⁻¹.

Resting energy expenditure

In the control subjects, mREE slightly but significantly exceeded tREE, as previously described 23. mREE was strongly correlated with FFM (R = 0.97; 95% confidence interval (CI) 0.89–0.99; p<0.0001; fig. 1⇓).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1—

Relationship between measured resting energy expenditure (mREE) and fat free mass (FFM) in the control subjects (□) and the Duchenne patients studied during spontaneous breathing (▪) or using mechanical ventilation (▓). The curvilinear lines indicate the 95% confidence interval of the correlation between mREE and FFM in the two groups.

In DMD patients as a whole, mREE was significantly lower than in controls (4559±853 kJ·24 h⁻¹ versus 7407±1312 kJ·24 h⁻¹, respectively; p<0.0001). It was also markedly lower than tREE, by 22% on average (4559±853 kJ·24 h⁻¹ versus 5766±1187 kJ·24 h⁻¹, respectively; p<0.0001). Conversely, mREE and fREE were not significantly different (4559±853 kJ·24 h⁻¹ versus 4826±1274 kJ·24 h⁻¹, respectively; p = 0.4; table 2⇓). As in the controls, mREE and FFM were correlated (R = 0.79; 95% CI 0.50–0.93; p<0.0001), but the strength of this association was significantly lower than in the controls (R difference 0.17; 95% CI 0.10–0.94; p = 0.0264). When the continuous ventilatory support patients were removed from the analysis, the correlation between mREE and FFM (R = 0.830; 95% CI 0.420–0.959) was not significantly different from the controls.

In the seven patients continuously dependent on MV, mREE was significantly lower than in the 13 patients using nocturnal-only ventilation (4000±960 kJ versus 4900±650, respectively; p = 0.0200), with the difference averaging 19%. Body weight, BMI and FFM did not differ significantly between the two patient subgroups. In the continuously ventilated patients, the correlation between mREE and FFM did not reach significance (R = 0.705; 95%CI -0.249–0.965; p = 0.1286), but the number of observations was small.

Energy intake

On average, the energy intake of DMD patients was 1.2±0.4 greater than their mREE (5536±1432 kJ·24 h⁻¹ versus 4559±853 kJ·24 h⁻¹, respectively; p = 0.008; table 2⇑). This was true in nocturnal-only ventilated patients, as well as in continuously ventilated patients.

Only four patients had an energy intake lower than their mREE. In two cases, the difference did not exceed 1%. One patient was on a diet due to obesity (BMI = 29) and one patient had an energy intake 9% lower than his mREE. This was the only patient with a low albumin (30.8 g·L⁻¹; lower limit of the normal range in the laboratory: 41 g·L⁻¹) and a low pre-albumin (0,181 g·L⁻¹; lower limit of the normal range in the laboratory: 0.203 g·L⁻¹).

Methodological considerations

Height can underestimate arm span in patients suffering neuromuscular diseases, due to scoliosis. All patients in the current study had undergone surgical correction of scoliosis (on average when aged 11–13 yrs, and before spinal curvature had reached 25°). Patients height was determined with a tape measure during stretching, in order to adjust for scoliosis. Therefore, the authors are confident that the BMIs provided were not an over overestimation. The most common way to predict REE in clinical practice is to apply the Harris and Benedict equations 20, which take into account sex, age, height and body mass. The present results (averaging a 22% difference between measured values and Harris and Benedict predicted values) corroborate observations made in other settings 21, and confirm that this approach is erroneous in the presence of a reduced muscle mass. Conversely, predicting REE from its relationship to FFM established in controls appears to be as valid in patients with DMD as in other types of patients 18, 21, 22. This study assessed FFM from bio-electrical impedancemetry. This has not been formally validated in patients with neuromuscular diseases and could have specific drawbacks in this setting. However, compared with other techniques, such as dual-energy X-ray absorpiometry 2, total-body potassium determination 24, whole-body densitometry 2, magnetic resonance imaging or deuterium dilution 21, 25, bio-electrical impedance analysis is the only practical method in ventilator-dependent patients who have reached a stage where full assistance is required to move from the bed or wheelchair. Furthermore, some of the other techniques cannot be performed for technical reasons in these patients (for example, all study patients had undergone surgical correction of scoliosis with Harrington or Luque rods making magnetic resonance imaging impossible). Although the patients who were not continuously ventilator-dependent could have been assessed by some of the above techniques, the choice of bio-electrical impedance allowed the authors to assess FFM homogeneously in the two patient populations. There is a clear need for a validation of bio-electrical impendance analysis of FFM for neuromuscular patients.

Energy expenditure and neuromuscular diseases

The medical literature is scarce on this topic. In adult patients with various types of neuromuscular disease, including some DMD patients, McCrory et al. 26 found a preserved REE but a reduced total daily energy expenditure. In patients with myotonic dystrophy, Josefowicz et al. 24 reported a reduction in REE expressed as a function of body surface area, but an augmentation when expressed as a function of FFM. In children with DMD, compared to age-matched controls, Hankard et al. 27 found a decreased REE. This decrease became nonsignificant after correction for FFM, suggesting that muscle attrition was a major factor in the REE. Similarly, Zanardi et al. 28 failed to demonstrate a low REE in a small group of children with DMD. This does not seem to hold true in the patients presented. Indeed, a markedly reduced energy expenditure was found, which was directly proportional to the loss in FFM, as indicated by the lack of significant difference between the measured values of REE and the values predicted by FFM adjustment procedure. The difference between the studies of Hankard et al. 27, Zanardi et al. 28 and the present study could be explained by the radically different nature of the populations studied. Hankard et al. 27 studied children with little reduction in BMI and who did not require ventilatory assistance. This was also the case for Zanardi et al. 28. Patients in the current study, however, were adults, had a dramatically reduced muscle mass and had patent chronic respiratory insufficiency. In the studies of McCrory et al. 26, Hankard et al. 27, and Zanardi et al. 28, the patients were not mechanically ventilated and had a relatively preserved FFM. In this regard, it is important to keep in mind the data provided by Zurlo et al. 29, showing that skeletal muscle can account for up to 20% of REE, and the data of Kayser 30 showing that skeletal muscle is responsible for 25% of V′_O2 of the body.

Insights into the impact of ventilatory muscle activity on resting energy expenditure

In the 13 patients studied while breathing spontaneously, the strength of the association between REE and FFM was not significantly lower than in the controls. Conversely, in the seven patients who were ventilator dependent and, thus, were studied in the absence of ventilatory muscle activity, there was no correlation between REE and FFM. Caution is called for by the small number of patients in this category, suggesting that it is possible to observe a quasi-suppression of the contribution of skeletal muscle to REE at an extreme stage of the disease when the viscera alone constitute the FFM and, thus, become the main factors influencing REE. Although the absence of ventilatory muscle activity during the metabolic measurements is one of the main differences between the two groups of patients, it is unlikely that this suffices to fully explain the observed gap in REE, even in the presence of, for example, a micro-atelectasis-related increase in the work of breathing. Indeed, the average difference in REE between the two categories of patients with DMD amounted to 19%, a figure which is much greater than the energy expenditure of the human respiratory muscles at rest, namely 1–2% of the total REE 31. Nevertheless, it seems reasonable to postulate that in patients who switch from assisted to spontaneous breathing, important fluctuations in REE can occur. Of note, the present results suggest that no specific nutritional measures are needed when patients with DMD go from nocturnal-only to continuous ventilatory support, if their condition is otherwise stable (see below).

Practical implications

This study provides a picture, although not comprehensive, of the nutritional status of adult patients with DMD having reached the stage of the disease when ventilatory support is needed. It appears that, in a stable clinical condition, nutritional intake and energy expenditure are reduced proportionately and, hence, grossly balanced. The mechanisms of this reduction are unclear, but may reflect a homeostatic adaptation to the reduction in muscle mass (of note, patients had been given general nutritional recommendations, e.g. how to balance the daily intake of the main nutrients, but were not following any kind of specific nutritional programme). Therefore, from a practical point of view the authors believe that patients with DMD should receive standard nutritional education and be told to eat ad libitum in order to satisfy perceived hunger and their weight should be precisely monitored. Any significant increase or decrease in weight, in the absence of obvious factors, such as an infectious state or sodium retention, should prompt a measure of the FFM and the energy expenditure in order to guide nutritional adaptations. In future studies, the value of fat mass as a nutritional marker in DMD should be specifically evaluated, due to its independence from the disease-related muscle loss.

Of note, it is difficult, if at all possible, to put the results from the present study and the above proposition into the perspective of the nutritional recommandations derived from the weight percentile chart proposed by Griffiths and Edwards 1, 32–35. Indeed, these data pertain to children only, whereas the youngest patient in this study was aged 21 yrs.

Alternatively, the hypothesis of a primary reduction in food intake (whatever its mechanism) as the source of an accelerated loss in muscle mass, and in turn of the REE, must be considered.

Perspectives

To better understand the nutritional processes at play in patients with Duchenne muscular dystrophy longitudinal data on energy expenditure and calorie intake are needed. Gaining a better knowledge of the changes in energy expenditure and food intake that are associated with infections, and with the course of the Duchenne muscular dystrophy-related cardiomyopathy, would be particularly important. This knowledge would help clinicians to better monitor patients with Duchenne muscular dystrophy to precisely adapt energy intake to energy expenditure in situations such as surgery, intercurrent infections, cardiac insufficiency or respiratory failure.