New insights into total and resting energy expenditure using state-of-the-art methods in cancer survivors: a cross-sectional study.

Precise measurement of energy expenditure is essential for guiding nutritional care after cancer treatment. However, commonly used predictive equations may be inaccurate for individuals recovering from cancer. Leveraging state-of-the-art methods can offer valuable insights into actual energy requirements of cancer survivors upon treatment completion. To characterize total (TEE) and resting energy expenditure (REE) and to assess the accuracy of predictive equations against measured values in post-treatment colorectal cancer survivors (CRCS). In this cross-sectional study, four TEE and 22 REE equations were compared against doubly labelled water (DLW) and whole-room indirect calorimeter (WRIC), respectively. Accuracy was assessed via paired t-test, Bland-Altman analysis, and the proportion of predictions within 10% of measured values. Pearson correlations investigated relationships between percentage bias and variables of interest (e.g., age, weight, BMI, body composition). Twenty participants (equal sex distribution; mean±SD age: 61.4±14.1y; BMI: 28.8±6.4 kg/m2) were included. Most had a history of colon cancer (55%), and stage III disease (75%). Predictive equations (pTEE range: ∼2060-2500 kcal/day; pREE range: ∼1230-1730 kcal/day) commonly underestimated measured TEE (∼2460±680 kcal/day) (n=2, 50%) and REE (∼1700±330 kcal/day) (n=19, 86.4%). Dietary Reference Intake (DRI) equations with estimated physical activity level had the highest individual-level accuracy for TEE prediction but still resulted in substantial intra-individual variability (up to ∼1400 kcal error). BMI and body composition were positively related to percentage bias in TEE but not REE equations. For REE, the Johnstone and Harris-Benedict equations showed the best individual-level agreement, but exhibited high intra-individual variability, with errors up to ∼460 kcal and 530 kcal, respectively. A majority of CRCS exhibited higher energy expenditure than estimated by standard prediction equations, underscoring the need to validate these equations in populations with cancer to optimize accuracy. Improved methods for assessing energy expenditure are needed to guide long-term survivorship care.