Prediction is not precision: Predictive equations poorly predict resting energy expenditure in patients with...

Patients with peritoneal malignancy often present with complex and advanced disease, placing patients at high nutritional risk. Despite this, data describing energy metabolism in this population is limited. The aim of this study was to measure and compare resting energy expenditure (REE) in patients with peritoneal malignancy presenting for cytoreductive surgery and hyperthermic intraperitoneal chemotherapy with five commonly used predictive equations.

REE was measured pre-operatively and predicted REE (pREE) was calculated using Harris-Benedict, Schofield, MifflinSt-Jeor, Henry and weight based 20 kcal/kg equations. Agreement between measured and predicted REE was assessed using paired t-tests, Bland -Altman analysis and percentage of estimations within 10% of measured values.

Forty-five patients were included (mean age 51.3 ± 12.3 years, 64% female), median peritoneal cancer index was 12.5 (IQR 5.5-33.0). Mean measured resting energy expenditure (mREE) was 1567.6 ± 511.0 kcal/day, a significantly higher mREE was observed in males compared to females (1992.8 ± 552.0 vs 1333.0 ± 294.5 kcal/day, p < 0.001)). Among the predictive equations evaluated, none were able to reliably estimate REE in individual patients. Limits of agreement were wide for all equations across sexes, indicating substantial individual variability. Individual-level accuracy was poor, with fewer than 50% of patients having predicted REE within ±10% of measured REE. Agreement at the group level was acceptable with only the Mifflin-St Jeor equation significantly different from mREE, underestimating energy expenditure by 126 ± 382 kcal/day (p = 0.032). Percent bias for all equations was within ±5% at the group level; however, sex-specific analyses demonstrated a trend toward underprediction in males and overprediction in females.

Predictive equations demonstrated poor individual accuracy and wide limits of agreement when compared to mREE in patients with peritoneal malignancy. A trend towards underestimation of REE in males and overestimation in females was observed but requires confirmation in larger studies. These findings suggest that indirect calorimetry may be able to provide further information for individualised nutrition assessments and highlights the need for further research into how this may improve clinical and nutritional outcomes.