Predictive factors for gastrostomy at time of diagnosis and impact on survival in patients with amyotrophic lateral sclerosis.

BACKGROUND: Gastrostomy is recommended in patients with Amyotrophic Lateral Sclerosis (ALS) in the presence of weight loss over 10% as compared to usual weight, repeated aspirations or meal time duration longer than 45 min. Currently, the impact of gastrostomy on survival of ALS patients is not clear. AIMS: i) to describe diagnosis factors associated with the indication for gastrostomy ii) to evaluate survival of ALS patients with gastrostomy indication according to their acceptance of feeding tube placement. METHODS: Patients with ALS were included and followed in the ALS referral centre of Limoges's teaching hospital between 2006 and 2017. Neurological, nutritional and respiratory status was assessed prospectively from diagnosis to death. Statistical analysis was performed using Mann-Whitney test, Chi2 tests, Cox model and multivariate logistic regression. RESULTS: Two hundred and eighty-five patients were included. Among the 182 for whom gastrostomy was indicated, 63.7% accepted the placement. The median time was 7.3 months [IQR: 3.2-15.0] and 2.7 months [IQR: 0.9-5.8] respectively from diagnosis to indication and from indication to placement. Weight loss >5% significantly increased the risk of death by 17% (p < 0.0001). At time of diagnosis, bulbar onset, a loss of one point in the body mass index or on the bulbar functional scale were all positively associated with indication for gastrostomy (aOR = 10.0 [95%CI: 1.96-25.0]; p = 0.002, aOR = 1.17 [95%CI: 1.02-1.36]; p = 0.025 and aOR = 1.19 [95%CI: 1.06-1.32]; p = 0.002, respectively). However, gastrostomy placement did not have any impact on survival (aHR = 1.25 [95%CI: 0.88-1.79]; p = 0.22). CONCLUSION: Both neurological and nutritional criteria were associated with an indication for gastrostomy at diagnosis. Gastrostomy placement had no impact on survival. The study of earlier gastrostomy placement might be of interest in further prospective studies.

Increased resting energy expenditure compared with predictive theoretical equations in amyotrophic lateral sclerosis.

OBJECTIVES: Approximately 50% to 60% of amyotrophic lateral sclerosis (ALS) is characterized by an increase in metabolic rate. The Harris and Benedict (HB) 1919 formula is the equation mainly used to calculate resting energy expenditure (cREE) compared with measured REE (mREE) by indirect calorimetry (IC), but other formulas are also applied in current practice. The present study aimed to assess mREE in patients with ALS compared with 12 cREE formulas and study the relevant threshold of REE variation to screen patients with a higher evolving risk. METHODS: Nutritional assessments and body composition (by bioimpedance analysis) were performed in patients with ALS. mREE was measured by IC, and cREE was calculated using the HB 1919, HB 1984, World Schofield, De Lorenzo, Johnstone, Mifflin, World Health Organization/Food and Agriculture Organization, Owen, Fleisch, Wang, Rosenbaum, and Nelson formulas. Functional and respiratory evolution and survival by log-rank test according to two thresholds of REE variation (10% and 20%) were studied. RESULTS: A total of 315 patients with ALS were included in the study. The median mREE was 1503 kcal/24 h (range, 1290-1698 kcal/24 h), which was higher than all predictive equations (P < 0.0001). Depending on the predictive equation, REE variation >10% and 20% was found in 35.2% to 76.3% and 14.6% to 53.3% of patients with ALS, respectively. Patients with an REE variation >20% with HB 1919 and HB 1984 had a lower survival. Moreover, with this same threshold and the Mifflin formula, patients had higher functional and respiratory evolutions and lower survival. CONCLUSIONS: The increase in metabolic rate is present according to the different cREE formulas used compared with IC. In clinical practice, REE formulas (e.g., HB 1919, HB 1984, or Mifflin) can be used as a reference value compared with IC to screen patients with ALS with an REE variation >20% and a higher evolving risk.

Resting energy expenditure equations in amyotrophic lateral sclerosis, creation of an ALS-specific equation.

INTRODUCTION: Resting energy expenditure (REE) formulas for healthy people (HP) are used to calculate REE (cREE) in amyotrophic lateral sclerosis (ALS) patients. In 50-60% of ALS cases an increase of measured REE (mREE) in indirect calometry (IC) compared to cREE is found. The aims here were (i) to assess the accuracy of cREE assessed using 11 formulas as compared to mREE and (ii) to create (if necessary) a specific cREE formula for ALS patients. METHOD: 315 Patients followed in the ALS expert center of Limoges between 1996 and 2014 were included. mREE assessed with IC and cREE calculated with 11 predictive formulas (Harris Benedict (HB) 1919, HB 1984, WSchofield, De Lorenzo, Johnstone, Mifflin, WHO/FAO, Owen, Fleisch, Wang and Rosenbaum) were determined at the time of diagnosis. Fat free mass (FFM) and fat mass (FM) were measured with impedancemetry. A Bland and Altman analysis was carried out. The percentage of accurate prediction ±10% of mREE, and intraclass correlation coefficients (ICC) were calculated. Using a derivation sample, a new REE formula was created using multiple linear regression according to sex, age, FFM and FM. Accuracy of this formula was assessed in a validation sample. RESULTS: ICC ranged between 0.60 and 0.71 (moderate agreement), and percentage of accurate prediction between 27.3% and 57.5%. Underestimation was found from 31.7% to 71.4% of cases. According to these unsatisfactory results we created an ALS-specific formula in a derivation sample (130 patients). ICC and percentage of accurate prediction increased in a validation sample (143 patients) to 0.85 (very good agreement) and 65.0% respectively, with 17.5% underestimation. CONCLUSION: REE formulas for HP underestimate REE in ALS patients compared to mREE. Our new ALS-specific formula produced better results than formulas for HP. This formula can be used to estimate REE in ALS patients if IC is not accessible.