Visceral obesity measured using computed tomography scans: No significant association with mortality in critically ill patients

Michelle R. Baggerman, Ingeborg M. Dekker, Bjorn Winkens, Steven W.M. Olde Damink, Sandra N. Stapel, Peter J.M. Weijs, Marcel C.G. van de Poll

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Introduction: The association between obesity and outcome in critical illness is unclear. Since the amount of visceral adipose tissue(VAT) rather than BMI mediates the health effects of obesity we aimed to investigate the association between visceral obesity, BMI and 90-day mortality in critically ill patients. Method: In 555 critically ill patients (68% male), the VAT Index(VATI) was measured using Computed Tomography scans on the level of vertebra L3. The association between visceral obesity, BMI and 90-day mortality was investigated using univariable and multivariable analyses, correcting for age, sex, APACHE II score, sarcopenia and muscle quality. Results: Visceral obesity was present in 48.1% of the patients and its prevalence was similar in males and females. Mortality was similar amongst patients with and without visceral obesity (27.7% vs 24.0%, p = 0.31). The corrected odds ratio of 90-day mortality for visceral obesity was 0.667 (95%CI 0.424–1.049, p = 0.080). Using normal BMI as reference, the corrected odds ratio for overweight was 0.721 (95%CI 0.447–1.164 p = 0.181) and for obesity 0.462 (95%CI 0.208–1.027, p = 0.058). Conclusion: No significant association of visceral obesity and BMI with 90-day mortality was observed in critically ill patients, although obesity and visceral obesity tended to be associated with improved 90-day mortality.

Original languageEnglish
Article number154316
JournalJournal of Critical Care
Volume77
DOIs
Publication statusPublished - Oct 2023

Fingerprint

Dive into the research topics of 'Visceral obesity measured using computed tomography scans: No significant association with mortality in critically ill patients'. Together they form a unique fingerprint.

Cite this