The association between BMI and 90-day mortality in patients with and without diabetes seeking care at the emergency department

Background: The impact of body mass index (BMI) on mortality varies with age and disease states. The aim of this research study was to analyse the associations between BMI categories and short- and long-term mortality in patients with or without diabetes seeking care at the emergency department (ED) with acute dyspnoea. Population and methods: Patients aged >= 18 years at ED during daytime on weekdays from March 2013 to July 2018 were included. Participants were triaged according to the Medical Emergency Triage and Treatment System-Adult score (METTS-A), and blood samples were collected. Totally, 1,710 patients were enrolled, with missing values in 113, leaving 1,597 patients, 291 with diabetes and 1,306 without diabetes. The association between BMI and short-term (90-day) and long-term (mean follow-up time 2.1 years) mortality was estimated by Cox regression with normal BMI (18.5-24.9) as referent category, with adjustment for age, sex, METTS-A scoring, glomerular filtration rate, smoking habits and cardiovascular comorbidity in a fully adjusted model. The Bonferroni correction was also used. Results: Regarding long-term mortality, patients with diabetes and BMI category >= 30 kg/m(2) had a fully adjusted Hazard Ratio (HR) of 0.40 (95% confidence interval [CI]: 0.23-0.69), significant after the Bonferroni correction. Amongst patients without diabetes, those with underweight had an increased risk but only of borderline significance, whilst risks in those with overweight or obesity did not differ from reference. Regarding short-term mortality, risks did not differ from reference amongst patients with or without -diabetes. Conclusions: We found divergent long-term mortality risks in patients with and without diabetes, with lower risk in obese patients (BMI >= 30 kg/m(2)) with diabetes, but no increased risk for patients without diabetes and overweight (BMI: 25-29.9 kg/m(2)) and obesity.