Survival predictions of different arm circumference classifications in patients under intensive care.

Abstract

Background & aims: Mid-upper arm circumference (MUAC) is a simple anthropometric measure widely used as a marker of nutritional reserves, in the context of critical illness and other clinical conditions. Although MUAC can be classified using different approaches, few studies have investigated the prognostic value of these classifications in critically ill patients. In this study, we aimed to evaluate the ability of different MUAC classifications approaches to predict survival among patients admitted to the intensive care unit (ICU).

Methods: This was a secondary analysis of a cohort study with prospective data collection. Adults of all ages admitted to a general ICU were included. MUAC was measured upon ICU admission and classified using four different approaches: two continuous and two categorical. Continuous measures included (1) absolute MUAC in centimeters (MUAC_cm) and (2) MUAC adequacy percentage (MUAC%), calculated using the 50th percentiles from Frisancho. Categorical classifications were (3) MUAC% based on Blackburn & Thornton criteria, categorized as low (<90 %), adequate (90-110 %), and excess (>110 %), and (4) using the Malnutrition Universal Screening Tool (MUST) cutoff, with low MUAC defined as <23.5 cm. Outcomes of interest were 28-day ICU mortality and in-hospital mortality.

Results: A total of 426 patients were included (53.1 % male, median age: 64 years, interquartile range: 53 to 71). Metabolic disorders were the most common cause of ICU admission (49.5 %). Cox regression analysis using continuous variables showed that for each 1 cm decrease in MUAC, the risk of ICU mortality increased by 11 % (HR _adjusted 1.11, 95 % CI 1.01 to 1.20). Furthermore, for each 1 % decrease in MUAC, the risk of ICU mortality increased by 4 % (HR _adjusted 1.04, 95 % CI 1.01 to 1.06). Low MUAC according to the MUST classification was associated with a 3.73-fold increase in the hazard of 28-day ICU mortality and a 2.70-fold increase in the hazard of in-hospital mortality. No significant associations were observed using the Blackburn & Thornton classification.

Conclusion: As a continuous variable, MUAC was an independent predictor of 28-day ICU mortality, but not of in-hospital mortality. In contrast, low MUAC classified according to the MUST criterion independently predicted both 28-day ICU and in-hospital mortality. No significant associations were observed using the Blackburn & Thornton classification. These findings highlight the prognostic utility of a simple anthropometric measure; however, they also suggest that different classification methods may yield distinct prognostic implications.