Body Composition and Intradialytic Exercise in Kidney Disease: A Combined Analysis of the PEDAL and CYCLE-HD Randomised Controlled Trials

Background
Haemodialysis patients are at high risk of myopenic obesity, necessitating effective nutritional status monitoring and intervention strategies. This combined analysis of two clinical trials (PEDAL trial and CYCLE-HD study) aimed to (i) determine the clinical utility of body mass index (BMI) in comparison to fat tissue index (FTI) and lean tissue index (LTI) and (ii) assess the effect of a 6-month intradialytic exercise intervention compared to usual care on FTI and LTI.

Methods
A priori secondary endpoints in both trials included BMI, FTI and LTI. BMI was classified by World Health Organisation definitions (underweight, healthy, overweight or obese). FTI and LTI were determined by Bioelectrical Impedance Spectroscopy and classified by previous research evidence (FTI of 4–15 kg/m2 and LTI of 15–20 kg/m2 being associated with best survival). For aim (i), BMI was compared to FTI and LTI by correlation and classification. For aim (ii), changes over 6 months in FTI and LTI were compared between exercise intervention and control groups.

Results
Across both studies, 298 and 209 participants had bioelectrical impedance spectroscopy measurement at baseline and 6 months, respectively. Mean (SD) age: 58 (15) years; BMI: 28.2(6.3) kg/m2; male: 65%. At baseline, only 47 of 298 participants (16%) had an FTI and LTI associated with best survival. BMI correlated with FTI (r = 0.79; p < 0.0001). However, 34% of participants were misclassified by BMI (e.g., 9% of patients were classified as obese by BMI yet FTI revealed their body composition was healthy). BMI did not correlate with LTI (p = 0.15), and 86% of participants were misclassified by BMI (e.g., 73% of patients were classified as healthy, overweight or obese by BMI yet LTI revealed they were myopenic). There was no difference between exercise intervention and control groups in mean change (95% CI) over 6 months for LTI (−0.3 [−1.1 to 0.4] kg/m2; p = 0.4) or FTI (0.2 [−0.7 to 1.0] kg/m2; p = 0.7).

Conclusions
Worryingly, only a minority (16%) of haemodialysis patients had both LTI and FTI within the range associated with best survival. Body composition misclassification using conventional BMI cut-offs was common: despite having healthy, overweight or even obese BMI, the majority (73%) of patients had hidden myopenia according to LTI. Six months of intradialytic aerobic exercise did not improve body composition. This study identified that common measures of nutritional status in haemodialysis patients such as BMI are misleading and that interventions other than intradialytic cycling are urgently required to target myopenic obesity.