Development of a Predictive Energy Equation for Maintenance Hemodialysis Patients: A Pilot Study

Modeling a Predictive Energy Equation Specific for Maintenance Hemodialysis

“…In 2 separate studies, inflammation was associated with an acceleration in energy expenditure among CKD patients . Similar to our preliminary findings, this sample of patients receiving MHD were largely inflamed, and the CRP levels remained a strong predictor of mREE. As such, our results suggest that the REE among these patients is elevated and that any of the existing predictive energy equations that do not account for these factors could grossly underestimate energy determinations with potentially harmful side effects.…”

“…On the basis of our findings, we continue to support the hypothesis that the addition of laboratory data produces better‐fitting models of mREE among patients receiving MHD than with demographic and anthropometric characteristics alone, as is commonly done in predictive formulas for healthy populations . Hence, we were able to explain a greater proportion of the variance in mREE among patients diagnosed with CKD, with a higher degree of precision than any of the other earlier investigations . In so doing, we confirmed similar predictors of mREE from our pilot study and were able to create and internally validate a model within a larger sample of patients receiving MHD.…”

“…In terms of its overall precision, our equation explains more of the variance of mREE than our earlier model in a similar MHD population . We also gained a higher proportion of participants for whom the equation “accurately” estimated REE, which is higher than other commonly reported equations, even in general populations . Nonetheless, there are individuals who fell outside the band of acceptability, and additional exploration into improving the bias is warranted and limits the broad application of such findings without further inquiry.…”

“…Because SCr values are routinely collected for the purpose of determining dialytic adequacy and protein nutriture, we chose to evaluate its ability to predict mREE in this sample. In our predictive modeling, SCr continues to be an alternate parameter that may be used when either CRP or A1c is unavailable or missing . SCr is convenient and easily accessible, and its inclusion improved the practicality of the predictive model.…”

“…To further our understanding of energy expenditure in CKD, we hypothesized that there were several clinical factors that influence energy requirements . We recently published a preliminary predictive model that is disease specific with better precision than some of the common formulas utilized, but we were not able to fully validate it for clinical use . Subsequently, in a larger sample, we were able to measure REE and identify key clinical factors similar to our pilot work.…”