Energy Expenditure in Kidney Failure: Implications for Management
Renal replacement therapy, in the form of dialysis or transplantation, is the cornerstone of management for end-stage renal disease. UK renal registry shows nearly half of those needing renal replacement therapy are treated by dialysis – predominantly by haemodialysis. Patients on renal replacement therapy have increased mortality risk compared to age matched general population. Moreover, some specific subgroups of patients on haemodialysis have increased risk of mortality than expected. The survival benefit seen in women in the general population is attenuated resulting in similar survival for men and women on haemodialysis therapy. In addition, obese individuals and those of non-Caucasian origin have better survival outcome. Though the underlying reason for these findings is not clear and is likely to be multi-factorial, it has been hypothesised that this paradox could be due to the current practice of normalising dialysis dose to total body water. A number of metabolic factors – body surface area, resting energy expenditure and total energy expenditure – have been proposed as alternative to total body water for scaling dialysis dose. There were two overarching aims of this work – one was to study the effect of declining renal function on resting and total energy expenditure and to study the influence of various energy expenditure measures on uraemic toxin generation. The second was to study the impact on survival outcome of using these alternate parameters for normalising dialysis dose and to derive dialysis dose adjustments based on these metabolic parameters. In order to study these aims, studies were designed to explore different aspects of energy expenditure measures along with a longitudinal study to examine the impact of these parameters on survival outcome. The relationship between energy metabolism, body composition and uraemic toxin generation was studied with a retrospective analysis of 166 haemodialysis patients in whom urea generation rate was used as surrogate marker of uraemic toxin generation. It was found that total energy expenditure and fat-free mass predicted uraemic toxin generation after adjustment for other relevant variables. This study provided the preliminary data which was useful in designing further studies for this work. The effect of renal function on resting and total energy expenditure was studied in 80 patients with varying stages of chronic kidney disease who were not on renal replacement therapy. Resting and total energy expenditures were measured directly using gold-standard methods. It was found that declining renal function did not have a significant influence on either of these measures. This supports the hypothesis that metabolic rate is the driving force for glomerular filtration rate and not vice-versa. The directly measured energy expenditure measures were also found to have a moderately strong relationship with urea generation rate in these patients not on renal replacement therapy. The impact of physical activity on uraemic toxin generation, and thereby dialysis requirement, was studied in a prospective cross-sectional study of 120 haemodialysis patients in whom the physical activity was measured by an accelerometer device. Results from the study showed physical activity level to be a significant predictor of uraemic toxin generation after adjustment for gender and body size differences. This study results stressed the importance of adjusting dialysis dose based on individual’s physical activity level. To study the impact of using metabolic factors as normalising parameter for scaling dialysis dose on survival outcome, a large-scale longitudinal study was conducted with 1500 maintenance haemodialysis patients recruited for the study. Dialysis dose-related parameters and survival outcomes were collected at baseline and at various time points during the follow-up period of 18 months. Study results were analysed in two parts - the theoretical basis for using these metabolic factors as scaling parameters was explored which showed that current minimum target dialysis dose risks under-dialysis in certain subgroups of patients and using these alternative parameters may provide a more equivalent dialysis dose across individuals of different body sizes and gender. With these results arguing for potential use of the alternative parameters, the impact on survival of using them were examined. It was found that all three parameters performed better than the current parameter (total body water) with regards to predicting mortality. Total energy expenditure was found to be the best parameter with the lowest hazard ratio for risk of death. The study data was also analysed to derive an algorithm for adjustment of minimum target dialysis dose based on body size and physical activity level. This newly derived minimum dose target was also shown to impact on survival with those underdialysed based on this criteria having poorer survival outcomes. To understand the impact of whole body protein turnover on resting energy expenditure and uraemic toxin generation, a cross-sectional study was conducted on 12 patients with advanced CKD – 6 each in pre-dialysis CKD and haemodialysis group. It was found that haemodialysis patients had higher rate of protein turnover compared to pre-dialysis patients. Whole body protein turnover was found to contribute significantly to resting energy expenditure and had a moderately strong relationship with urea generation rate. In the course of these studies, two questionnaire tools have been validated for use for clinical and research purposes – one is a self-report comorbidity questionnaire and the other, the Recent Physical Activity Questionnaire. The comorbidity questionnaire was developed as part of this work and was validated against Charlson Comorbidity Index. The Recent Physical Activity Questionnaire was validated for physical activity data collection and energy expenditure calculation against the gold-standard doubly labelled water method. In conclusion, it has been demonstrated that metabolic factors such as body surface area, resting energy expenditure and total energy expenditure are more closely related to uraemic toxin generation compared to total body water. It has also been demonstrated that physical activity contributes to metabolic waste production and may necessitate changes in dialysis requirement. It has been shown that these metabolic factors, when used as scaling parameter for dialysis dosing, may predict survival better than the current parameter in use. The algorithm for dialysis dose adjustment and the questionnaires validated in this work have provided novel tools for further research studies and clinical practice. The central hypothesis of this work is that some metabolic factors may be better markers of uraemic toxin generation compared to total body water. It is hypothesised that modifications in dialysis practice based on these factors may improve the quality of haemodialysis and favourably impact on survival outcome for patients with end-stage renal disease. The work presented here largely supports this hypothesis.