C34 MODELLING ENERGY EXPENDITURE IN ALS: A PROSPECTIVE LONGITUDINAL STUDY
Kasarskis EJ1, Tandan R2, Mitsumoto H3, Simmons Z4, Bromberg M5, Matthews D2, Mendiondo M1, Kryscio R1
1University of Kentucky, Lexington KY, USA, 2University of Vermont, Burlington VT, USA, 3Columbia University, New York NY, USA, 4Penn State University, Hershey PA, USA, 5University of Utah, Salt Lake City UT, USA
E‐mail address for correspondence: [email protected]
Background: Maintaining optimal nutrition for an ALS patient requires knowledge of food intake and energy (calorie) expenditure at various stages of the illness. Weight loss and determination of BMI are indirect indices of malnutrition since loss of body mass may be due to muscle atrophy and not undernutrition. The currently used energy predictive equations, such as the Harris‐Benedict equation, are not disease‐ and state‐specific for ALS. Without a specific equation for ALS, one might inadvertently recommend under‐ or over‐feeding.
Objectives: The goal of our study was to prospectively develop empiric equations to accurately predict energy expenditure in ALS patients in various functional states as their disease progresses.
Methods: We use the doubly‐labelled water (DLW) method to determine the absolute caloric expenditure over a 10‐day period (Total Daily Energy Expenditure, TDEE), which serves as the reference against which we will model various clinical factors to predict TDEE. To begin the modelling process, we determine lean body mass (LBM) using DXA and BIS methods to estimate basal metabolic rate (BMR) using the Wang and other equations. We determine: resting metabolic rate (RMR) using indirect calorimetry in sitting and supine positions, physical activity using accelerometers and the Bouchard rating of physical activity by the subject, and clinical data using rating scales of cramps, spasticity, fasciculations, and pseudobulbar behaviours which represent potential sources of caloric expenditure that do not result in functional movement.
Results: To date, we have enrolled 39 of a projected 100 subjects at five medical centers. The mean age was 58.6 years; 57% were male, 87% were Caucasian, 27% had unimpaired ambulation, and 37% had an entry FVC >80% of predicted, indicating that we successfully recruited a broad clinical spectrum of ALS subjects. Determination of LBM by DXA and BIS was highly correlated. The difference between TDEE and measured RMR varied widely between 181 and 1461 kcal/day (774±392 kcal/day). The difference between TDEE and the predicted RMR using the Harris‐Benedict equation varied between −67 and +1817 kcal/day (665±517 kcal/day). Other predictive equations yielded similar discrepancies.
Conclusions: The differences between measured or predicted RMR and TDEE represent the calories expended on a daily basis by the ALS patient that are not accounted for using these standard methods in clinical nutritional practice. As more patients are studied and longitudinal measurements obtained over 12 months, we will be able to analyse the contribution of the various clinical factors to account for the discrepancies between the absolute TDEE and the predicted energy needs using the standard approaches. We plan to develop new predictive equations based on these data and also disease‐specific modifications of the widely‐used Harris Benedict equation. The results of our study will standardize nutritional care in clinical practice and in drug trials.
Acknowledgements: We wish to thank the ALS participants and their families, the research study staff at the participating medical centers, and the staff of their GCRCs. Supported by: NIH grant RO1 NS045087‐02, the Cynthia Shaw Crispen Endowment, the ALS Hope Foundation, and GCRC grants MO1 RR02602, RR00645, RR00064, RR00109, and RR10732
C35 VALIDATION OF BIOELECTRICAL IMPEDANCE SPECTROSCOPY FOR MEASURING LEAN BODY MASS AND TOTAL BODY WATER IN ALS
Tandan R1, Simmons Z2, Mitsumoto H3, Bromberg M4, Kasarskis E5
1University of Vermont, Burlington, VT, USA, 2Hershey Medical Center, Hershey, PA, USA, 3Columbia Neurological Institute, New York, NY, USA, 4University of Utah, Salt Lake City, UT, USA, 5University of Kentucky, Lexington, KY, USA
E‐mail address for correspondence: [email protected]
Background: Malnutrition due to energy (calorie) imbalance, and dehydration, are common in amyotrophic lateral sclerosis (ALS). Cachexia due to loss of lean body mass (LBM) and fat mass is an ultimate consequence of the disease. Bioelectrical impedance spectroscopy (BIS) is a non‐invasive and relatively inexpensive technique for measuring body composition and total body water (TBW), variables that are essential for the understanding of body composition and caloric balance.
Objective: To prospectively validate the accuracy of BIS in measuring LBM and TBW, against the reference gold standard techniques of doubly labelled water (DLW) and dual X‐ray energy absorptiometry (DXA), in ALS.
Methods: We are currently undertaking a longitudinal, prospective study of caloric balance and changes in body composition in patients at five ALS centers. For the purposes of this validation study, data on demographics, disease characteristics, body composition measured by BIS, DXA and DLW; and TBW estimated by BIS and DLW are being collected.
Paired t‐tests and Pearson's correlation coefficients were used to compare LBM measured by DXA and BIS, and TBW measured by DLW and BIS. The absolute and percentage mean difference from data obtained by the reference techniques was recorded.
Results: To date, we have data available on 26 of an expected 100 patients. There were 54% males and onset was bulbar in 35% of patients. Other mean±SD values were: age 57±10 years, disease duration 26±19 months, BMI 25±5kg/m2, ALSFRS 35±6 and FVC 74±15% of predicted.
The mean±SD LBM‐BIS was 2.4±3.4 kg lower than LBM‐DLW (46.0±14.0 kg vs. 48.4±12.9 kg, mean difference −5.6%, p = 0.002) (n = 25); and 1.9±4.0 kg lower than LBM‐DXA (46.8±14 kg vs. 48.7±13.7 kg, mean difference −4.1%, p = 0.04) (n = 22). Pearson's correlation between LBM‐BIS and LBM‐DLW was r = 0.97 (p<0.001), and between LBM‐BIS and LBM‐DXA was r = 0.96 (p<0.001).
The mean±SD TBW‐BIS was 0.2±4.1 l higher than TBW‐DLW (35.7±11.0 l vs. 35.5±9.3 l, mean difference 0.3%, p = 0.84) (n = 26). Pearson's correlation between TBW‐BIS and TBW‐DLW was r = 0.93 (p<0.001).
Discussion and conclusion: Compared to DLW and DXA, BIS underestimates LBM by about 4–5% in ALS; nevertheless, LBM estimated by BIS is highly correlated with that measured by the gold standard techniques of DLW and DXA. BIS is as accurate as DLW in estimating TBW. Thus, BIS can be reliably used in studies of caloric balance in ALS.
Acknowledgements: Supported by NIH Grant RO1 NS045087‐2, the Cynthia Shaw Crispen Endowment, ALS Hope Foundation, and GCRC Grants MO1‐RR00109, RR02602, RR00645, RR00064, RR10732 and CO6‐RR016499.
C36 COMPLIANCE AS IT RELATES TO SPECIFIC ELEMENTS IN THE PRACTICE PARAMETERS FOR PEG PLACEMENT
O'Connor P, Miller R, Katz J
Forbes Norris ALS Research Center, San Francisco, USA
E‐mail address for correspondence: [email protected]
Background: According to the practice parameters of the American Academy of Neurology, percutaneous endoscopy tubes (PEG) should be recommended for accelerated weight loss and symptomatic dysphagia soon after the onset of these difficulties. However, terms such as ‘accelerated’, ‘symptomatic’, and ‘soon’ leave room for interpretation. It is not known how varying degrees and types of clinical findings affect compliance.
Objective: To determine how specific clinical parameters affect compliance for PEG intervention.
Methods: We reviewed records from 63 patients who died after being cared for at a multidisciplinary ALS center. PEG had been recommended for 52 patients (81%) before death. These patients were placed in three groups: 1) compliers (C), n = 24, PEG within two months of recommendation; 2) eventual compliers (EC), n = 18, initially refused PEG, placed more than two months later; and 3) non‐compliers (NC), n = 9, recommended but not received before death. Data from all visits, from the initial recommendation to death, were reviewed (dates of recommendation, placement, and death; gender; weight loss; per cent ideal body weight and usual body weight; FVC; ALSFRS scores for swallowing and saliva). Features were compared at the time PEG was offered; C and EC were also compared at the time PEG was eventually placed. A logistic regression model was used to determine which combinations of independent variables effectively predict PEG compliance. Odds ratios were calculated for each independent variable to determine how well they predict the occurrence of PEG placement.
Results: Significant differences at the time PEG was offered were found in ALSFRS bulbar subscores between EC and C (mean scores for swallowing EC = 2.8, C = 2.2, p = 0.02 and for salivation EC = 2.9, C = 2.2, p = 0.02). There were no differences in other variables. At the time the PEGs were placed, there were no statistical differences in any parameter between C and EC. The initial EC group took a mean of 252 (60–852) days to comply with the initial recommendation and lost an additional 4% (mean) of body weight (range 18% to gained 11%) during the interval. There were no significant differences between EC and NC at the time of recommendation for PEG, although NC had higher mean ALSFRS scores for swallowing (3.3) and saliva (3.4). At the time of death, the mean swallowing score in NC was 2.8.
Discussions and conclusions: Compliance with recommendations for PEG appears to depend on the degree of dysphagia. Patients who fulfil practice parameters for PEG may be reluctant to comply if they perceive swallowing is relatively intact, regardless of weight loss or respiratory difficulty. It is not yet clear if non‐compliance is related to actual dysphagia or under‐reporting of severity.