KEY POINTS
• Current knowledge: There are few data devoted to the combined assessment of the compartmental nutritional and respiratory status of subjects with neuromuscular diseases.
• Contribution of the article to current knowledge: This study puts into perspective the relevance of compartmen tal evaluation to assess respiratory muscle function and establishes that body muscle mass is the most relevant nutritional parameter in relation to respiratory muscle strength.
Respiratory muscle weakness is the hallmark of most neuromuscular diseases (NMD), and together with mal nutrition and weight loss, respiratory muscle weakness is a significant determinant of morbidity and mortality1,2. Nutritional assessment and respiratory status are essential components of evaluating children and adults with NMD3,4.
Clinical anthropometry is complicated for patients with NMD. The loss of muscle mass can be due to muscle atrophy from the primary pathology, muscle replacement by fat and fibrosis, and undernourishment. These factors can alter the compartmental distribution of muscle and fat mass and could be not reflected in the body mass index (BMI)5-7. It has been suggested that indices that incorpo rate the compartmental distribution of muscle (such as creatinine height index, CHI) and fat (zero muscle mass, ZMM) are more sensitive8,9.
There are few data devoted to the combined assess ment of the nutritional and respiratory status of subjects with NMD10-13. This study aimed to assess the relation ship between the compartmental nutritional status and respiratory muscle function measures in amyotrophic lateral sclerosis (ALS) and Duchenne muscular dystrophy (DMD) patients. Since only the CHI% considers muscle mass, we hypothesized that the respiratory muscle force could be more related to the muscle mass than with global nutritional status indicators such as BMI. We looked for correlations between compartmental nutritional variables and respiratory indices to identify respiratory muscle weak ness determinants.
Material and methods
We conducted a cross-sectional study at a tertiary care teach ing facility in Buenos Aires, Argentina. The information was extracted from an Institutional Registry of Neuromuscular Diseases with a validated quality data control process for extensive previous research. ALS and DMD patients were se lected. All respiratory function tests, as well as anthropometric measurements correspond to routine measurements in these patients. The diagnosis of ALS was established according to the revised criteria for the diagnosis of amyotrophic lateral sclerosis, El Escorial14. Regardless of gender and duration of disease, both subjects with bulbar and spinal predominance were included. Patients who used a wheelchair, had a gastros tomy, or received mechanical ventilation at home were also included. The diagnosis of DMD was established by clinical criteria and confirmed by dystrophin immunostaining3. The Vignos Functional Scale was used to assess functional motor ability15. At the time of evaluation, none of the selected patients presented unstable cardiac condition, active recent surgery, diabetes mellitus, and thyroid disorders, neither recent (6 weeks) episodes of acute respiratory exacerbation, infection, or hospitalization. Kidney function was normal. Patients were screened for eligibility during a regularly scheduled clinic visit. The candidates for the nutritional evaluation of the compart ment were selected for their ability to perform urine collec tion indications and to live in the vicinity of the hospital. The Institutional Ethics Committee approved the study (CUDAP: TRI-UBA: 0051153/2018, #216).
We employed several measures to assess nutritional status and compartmental body mass distribution8.
The body mass index (BMI) was calculated by divid ing the patient’s weight by the height squared and was expressed as kg/m2. Normal nutrition was defined by a BMI of 18.5 to 25 kg/m2. Values greater than 25 or below 18.5 were considered overweight and underweight, respectively16.
The percent of expected weight for zero muscle mass (ZMM%) was obtained after a creatine free diet for six days and total urinary creatinine excretion was averaged daily over the next three days. It provided a reliable estimation of muscle contribution to body weight according to the formula17. The patient’s weight for zero muscle mass (pZMM) = body weight 24 h creatinine excretion in mmol * 2.2625. The theoretical bodyweight for zero muscle mass is tZMM, which is obtained by subtracting the normal theoretical muscle mass from the normal theoretical bodyweight. Thus, tZMM is obtained by the following equation: tZMM = theoretical body weight (normal values of 24 hr creatinine excretion * 2.2625). The pZMM divided by tZMM is the percent of expected weight for zero muscle mass (ZMM%)9,17. Normal nutrition was defined as having 90 to 110% ZMM%. Values higher than 110% and below 90% were considered overweight (fat mass excess) and underweight.
The creatinine height index (CHI%) in percent of predicted values was calculated as follows: measured 24 h urinary cre atinine divided by expected 24 h urinary creatinine of normal individuals of same sex and height, expressed as a percent age. A CHI% higher than 80% of the standard was considered as normal protein status. Values of 60% to 80% represent a mild depletion in body muscle mass, values between 60% and 40% indicate moderate protein depletion, and less than 40% indicate a severe deficit of body muscle mass16.
The respiratory muscle strength was assessed by measur ing the maximum inspiratory pressure (MIP) and the maximum expiratory pressure (MEP) from residual volume and total lung capacity. The MIP reflects the strength of the diaphragm and other inspiratory muscles, whereas the MEP reflects the strength of the abdominal muscles and other expiratory muscles. At the mouth, these static maximum pressures were obtained with a snorkel-like (flanged) mouthpiece coupled with a unidirectional Hans Rudolph valve connected to a pressure transducer (Validyne MP 45, Validyne Engineering, Northridge, California). A 1mm leak was used to prevent glottic closure. Maximum effort was encouraged verbally, with simultaneous visual feedback from a monitor. Maneuvers were separated by at least 30-s rest periods and continued until no further increase in pressure could be obtained. Data were recorded in a digital format (MP100 Workstation, Biopac Systems, Goleta, California). The signals from the pressure transduc ers were filtered with low pass filters (30 Hz). They then were passed through an analog-to-digital conversion board (Biopac Systems) at a 60-Hz sampling rate for acquisition. Reference values of maximal static pressures were obtained from Evans’s equations18.
The spirometry (flow-volume curves) were taken accord ing to standard methodology. Subjects were seated and wearing nose clips. We reported the forced vital capacity (FVC) (Vitalograph Compact II Buckingham, U.K.); normal values were taken from NHANES III19. FVC, expressed ei ther as an absolute value or a percentage of the predicted value for age and sex, was recorded for all patients. In a few poor collaborative patients, spirometry and maximal static mouth pressures were repeated in another day to obtain reliable data.
The peak expiratory flow rate (PF) was obtained with por table devices (Personal Best, normal or low range). PF with the cough (PCF) maneuver was obtained by adjusting a peak flow meter to a facial mask.
Descriptive statistics were used to summarize the groups’ characteristics. Mean values of the data are followed by the standard deviation of the mean (SD). The comparison was tested for statistical significance using the Wilcoxon signed-rank or Mann- Whitney rank-sum test. Slopes were determined by linear regression analysis (SigmaPlot 12.0 Jandel Scientific Software, San Rafael, CA). The interpretation of the strength of the correlation (r-value) was made according to: > 0.7 strong, 0.5 moderate, 0.3 weak. Differences with p< 0.05 were considered significant. For alpha = 0.05, the desired power test would be 0.8.
Results
The DMD patients not receiving steroid treatment were excluded from the study. During the study period, a total of 74 patients were studied. Four were eliminated because of doubts about diet compliance or urine collec tion. All subjects in the DMD group started to be treated with steroid therapy (deflazacort 0.5-1.0 mg/kg) at 5 to 7 years old. A total of 36 ALS and 34 DMD patients were included in this study.
The physical characteristics and the pulmonary func tion test data of those enrolled in the study are shown in Table 1. There were no significant differences between our ALS and DMD groups regarding pulmonary function testing. Ten of the 36 patients with ALS were wheelchair users, four had gastrostomy tubes, and nine used long-term ventilatory assistance at home. Twenty-one of the 34 patients with DMD were wheelchair users, and the Vignos scale mean value was 6.6 ± 3.3.
The PCF in ALS subjects was 290.4 ± 157.0, while in DMD subjects was 296.0 ± 80.2 liters/min (p < 0.688). These values represent a mean increase of 21.3 for ALS (p < 0.154) and 39.2% for DMD patients (p< 0.001) of the PF, respectively. The FEV1/FVC ratio was within the predicted values.
Compared with the ALS group, the BMI was margin ally lower in DMD (p < 0.048), while the nutritional com partmental variables of both groups were significantly different (Table 2).
According to CHI%, and ZMM% both ALS and DMD groups showed a decrease in the body muscle mass and an excess in that body fat. The mean CHI% was higher in patients with ALS (p < 0.001), while the mean ZMM% was higher (p < 0.001) in patients with DMD (Table 2).
The relationship between nutritional compartmental parameters is shown in Table 3. In both groups, strong correlations uphill (positive) were found between ZMM% and BMI (r > 0.700, p < 0.001). Regarding DMD patients, no correlation was found between CHI% and BMI (r 0.091, p = 0.613), while in ALS patients, it was a sig nificant and weak positive linear relationship (r = 0.458, p < 0.007).
No correlation between CHI% and BMI was observed in either of the two groups of patients (Fig. 1A). The 53% of ALS and 22% of DMD patients were located inside the three quadrants of diagnostic agreement between ZMM % and BMI method (Fig. 1B).
Correlation coefficients and equations of nutritional (BMI, ZMM%, and CHI%) and respiratory (FVC%, MIP%, and MEP%) parameters in ALS and DMD patients are displayed in Table 4.
There was a weak positive linear relationship between BMI and the respiratory variables in ALS patients. Regarding DMD patients, we did not find a significant correlation between BMI and any respiratory variable. In both groups of patients, the ZMM% index did not correlate with any respiratory variable. The CHI% showed the strongest (r > 0.7) positive linear relationship with MIP%, MEP%, and FVC% in both ALS and DMD patients (Table 4).
The regression lines and their 95% confidence inter vals for nutritional and respiratory variables are shown in Figures 2 (ALS group) and 3 (DMD group). Regarding the significant relationships (Table 4), the best reliability of the estimation procedure (95% confidence interval of the regression line) was for CHI%.
The relationship between PCF and CHI% is shown in Figure 4. In both groups, the CHI% showed the strongest (r > 0.7) positive relationship with the PCF. In the ALS group, there was a moderate positive linear relationship between BMI and PCF (0.571, p < 0.001). In both groups of patients, the ZMM% index did not correlate with PCF.
Discussion
In this study, we correlated spirometry (mainly FVC), the maximal static pressures at the mouth (MIP and MEP), and a dynamic measure of cough effectiveness (PCF) with the global nutritional status according to BMI and with compartmental measurements such as body muscle mass according to CHI% and expected weight for zero muscle mass according to ZMM%. Our study’s most important find ing is that the CHI% showed the strongest relationship with FVC%, MIP%, MEP%, and PCF%, both in ALS and DMD patients. The BMI correlated with all respiratory variables in the ALS group but not in the DMD group, while the ZMM% did not correlate with any respiratory function parameters.
Neuromuscular diseases represent a heterogeneous group of acquired or inherited conditions. The nutri tional approach is complicated for these patients6,20. Pat terns of global nutritional parameters21, anthropometric measures11, compartmental muscle mass8, truncal fat distribution12, and total body fat depend on such NMD, years of evolution, steroid therapy22 and perhaps dietary interventions23.
Even when BMI remains a useful clinical tool if these limitations are acknowledged22, classifying boys with DMD as overweight or underweight based solely on height and weight or BMI, without accurate concurrent estimates of body composition, can be problematic5,24. Our present results confirm significant differences between nutritional status diagnoses in DMD patients8. The BMI does not accurately reflect body composition and underestimates excessive body fat. We found that in ALS patients, the diagnostic concordance was higher (Fig. 2B) than in DMD patients. It is an expected result because muscle atrophy in ALS patients does not follow muscle replacement by fat and fibrosis.
Our DMD group showed a more significant body fat (Table 2). Other authors have addressed body fat’s sig nificant influence upon BMI12,23. A child with pronounced muscle wasting can look cachectic, despite having exces sive soft-tissue body fat22. Our study suggests that DMD subjects, who look like normal BMI, from the compartmen tal approach, seem probably obese. Both in ALS and DMD patients, BMI correlated positively with ZMM%. Interesting, only a few DMD subjects lied into the concordance quad rants between ZMM % and BMI (Fig. 1B).
In both groups, the body fat was definitively associated with BMI, whereas the muscle mass only was positively correlated with the BMI only in patients with ALS (Table 3). The replacement of muscle mass by fat tissue present in DMD patients does not occur in ALS patients. It is a fact that equal BMI may be present in a very different compart ment distribution.
The restrictive pattern found in our ALS patients was explained by the inspiratory and expiratory muscles’ weakness (Table 1), while in DMD patients, the re strictive pattern results from a complex interaction between respiratory muscle weakness and progressive scoliosis25.
The combined assessment of the nutritional and re spiratory status in NMD, in theory, could depend on the nutritional approach performed. BMI correlated well with all respiratory variables only in the ALS group (Table 4). Perhaps the higher proportion of total body fat in DMD subjects precludes the BMI as an indirect indicator of muscle mass. Remarkably, the CHI% was the nutritional compartmental parameter that showed the most robust positive relationship with all respiratory variables (Table 4, Figs 2, 3 and 4).
Vincken et al. studied patients with a stable chronic and clinically heterogeneous NMD not presenting with respi ratory symptoms10. BMI did not correlate with respiratory muscle strength in their patients.
In patients with ALS, Silva et al. found that BMI cor related significantly with both muscle area and fat area of the arm and respiratory parameters11. Similarly to Vincken et al, the BMI did not correlate with muscle strength. It is an expected fact (at least in the DMD group) because it is a determination that does not discriminate fat from muscle. Furthermore, although it is known that severe malnutrition affects muscle strength, the low number of patients with severe malnutrition according to BMI in our DMD group makes it unlikely that a correlation with muscle strength can be established.
Canapari et al. designed his study to determine the relationship between central adiposity, BMI, and lung function in boys with DMD. FVC (% predicted) was not significantly correlated with BMI, and PCF was not associ ated with body fat parameters, including BMI12. Truncal fat distribution was a significant predictor of lower FVC. Interestingly, they suggest that an increase in BMI at a given truncal fat percentage is protective of FVC. Since BMI includes muscle mass, it may suggest that higher BMI is consistent with greater muscle mass. The authors found a lack of association of total fat mass (directly measured) with pulmonary function.
Given the progressively degenerative nature of DMD, an increasing BMI or weight may not be a reflection of increasing muscle mass. Chew et al. found that body fat mass was adversely associated with FVC13. Higher weight profile and lower adiposity have better respiratory outcomes. Our patients found that the respiratory muscle force was more related to the muscle mass than with global nutritional status indicators (Table 4).
The CHI% was the only parameter that correlated with respiratory functional abnormalities. This was valid for both ALS (Fig. 2) and DMD patients (Fig. 3). If we consider that muscle mass is adequately evaluated with the CHI% and respiratory muscle strength depends not only on the muscle’s contractile characteristics and muscle mass, this association is expected.
PCF was significantly higher than PF in the DMD group. As a whole, patients with ALS showed a signifi cantly higher dispersion of PCF relative to its mean than the DMD group, reflecting different degrees of bulbar involvement26. PF is used to assess the effectiveness of the cough as well as bulbar involvement in NMD. There is currently a 1st-grade recommendation for the applica tion of mechanically assisted cough in patients with NMD who present less than 270 L / min of PCF27. Most of our patients were mechanically cough assisted devised us ers. Remarkably in both groups, the CHI% showed the most robust relationship with the PCF%. The moderate linear relationship between PCF and BMI found only in the ALS group means perhaps the relative better bound between BMI and muscle mass in this group (Table 3). Interestingly, a few DMD subjects showed a flattening shape regression curve at almost predicted CHI% values (Fig. 4), reflecting perhaps a myopathic component of weakness, a well-known fact in DMD subjects beyond muscle mass loss(3,5,7,8).
This study has several limitations that merit consid eration. First, the findings represent the experience of a single center. Second, day-to-day variation in protein intake and exercise cause significant variation in 24-hour creatinine excretion. Further, 24-hour urine creatinine excretion is problematic because it is often inaccurate28,29. We trust in the subjects’ high motivation to comply with the diet and collect urine. To obtain urine specimens, we used convenience sampling to create samples according to the ease of access, compliance with medical indications, and people’s availability to be part of the study. Due to the sedentary characteristics of our patients, we dismiss that physical activity could be a confounding factor. Secondly, inaccuracies in anthropometric measures, especially height, may have influenced findings. Finally, since boys with DMD can move from over- to under-nutrition within their shortened life span, results should be restricted to our patients’ age. However, the DMD group receiving long-term deflazacort therapy could serve to assess the impact of the compartmental nutritional status on respira tory muscle function.
In conclusion, although the relationship between nu tritional status and respiratory muscle function has been established, data on this relationship in patients with NMD are scarce. BMI does not accurately reflect body compo sition and underestimates excess fat, it should be used cautiously to assess patients’ nutritional status with NMD. In these patients, it seems more appropriate to use indi ces that incorporate the evaluation of the compartmental distribution of muscle and fat.
This study shows that the CHI% is most powerfully related to the three indices examined with respiratory and global muscle strength values. It also puts into perspective the relevance of compartmental evaluation to assess re spiratory muscle function. It establishes that body muscle mass is the most relevant nutritional parameter concerning respiratory muscle strength. The clinical implications of our study should be explored. It must be outlined whether or not nutritional monitoring with a lot of elaborated indices than BMI and interventions targeted in body muscle mass modifies the development of respiratory muscle weakness.