Measured versus predicted resting metabolic rate in patients hospitalised due to a diabetic foot ulcer: a prospective observational cohort study

Abstract

Background & Aims

This study aimed to assess the eventual discrepancy between predicted resting metabolic rate (p-RMR) using equations (Harris-Benedict (HB), Mifflin-St. Jeor (MSJ), World Health Organization (WHO), Cunningham (CU), Lührmann (LU) and Owen) and the gold standard of the measured resting metabolic rate (m-RMR) using indirect calorimetry (IC), in patients admitted to the Antwerp University Hospital (UZA) with a diabetic foot ulcer (DFU). Secondly, the impact of nutritional status on this discrepancy was assessed.

Methods

This prospective observational cohort study recruited patients admitted to the UZA due to a DFU from November 1st, 2021, to June 30th, 2023. A thorough nutritional assessment (nutritional questionnaires, anthropometry and bio-electrical impedance analysis) was performed upon admission. The Global Leadership Initiative on Malnutrition (GLIM) criteria were used to diagnose malnutrition. On admission, daily caloric requirements were measured by IC and predicted using six equations (MSJ, HB, WHO, CU, LU and Owen). Data were analysed using paired t-tests and Bland-Altman plots to assess statistically significant and clinically significant (>150 kcal/day) differences.

Results

A total of 59 subjects were enrolled in this study, including 46 men (78%). Patients included had a median age of 74 (38–90) years, had diabetes for 21±15 years and had a median Body Mass Index (BMI) of 27,1 kg/m². Fifty-four percent were malnourished on admission. The mean m-RMR using IC was 1570±290 kcal/day while the mean p-RMR were 1584±241 kcal/day (MSJ), 1625±263 kcal/day (HB), 1685±210 kcal/day (WHO), 1596±203 kcal/day (CU), 1666±249 kcal/day (LU) and 1525±222 kcal/day (Owen). The WHO equation overestimated the m-RMR to the greatest extent (+115±213 kcal/day; P<0,001), followed by the Lührmann equation (+96±229 kcal/day; P=0,002). The equation most precisely predicting the m-RMR values was the MSJ equation (+14±213 kcal/day; P=0,610). The MSJ, WHO and Owen equations differed on average more from the m-RMR in patients who were malnourished (MSJ: +24±212 kcal/day; P=0,523) (WHO: +137±217 kcal/day; P=0,001) (Owen: -63±217 kcal/day; P=0,109) than in patients who were normally fed (MSJ: +2±218 kcal/day; P=0,955) (WHO: +90±211 kcal/day; P=0,036) (Owen: -23±188 kcal/day; P=0,524). For the HB, CU and LU equations, the p-RMR differed on average more from the m-RMR in patients who were normally fed (HB: +57±216 kcal/day; P=0,185) (CU: +42±189 kcal/day; P=0,261) (LU: +96±229 kcal/day; P=0,038) than in malnourished patients (HB: +54±218 kcal/day; P=0,174) (CU: +12±216 kcal/day; P=0,760) (LU: +96±232 kcal/day; P=0,027). For all equations, a subset of patients (WHO: 27/59; HB: 28/59; MSJ: 25/59; Owen: 22/59; CU: 20/59; LU: 28/59) exceeded the predefined clinically significant threshold of 150 kcal/day.

Conclusion

In patients admitted with a DFU, the MSJ, HB, WHO, CU and LU equations overestimated the m-RMR, while the Owen equation underestimated the m-RMR using IC. Of these equations, the MSJ appeared to be the most appropriate. More than 50% of the patients enrolled in this study were malnourished upon admission, and the predictive equations were less reliable in malnourished patients. Further research with strong quality design is needed to evaluate the discrepancy between these predictive equations in malnourished patients, as those individuals require precise nutritional therapy the most.