The Association Between BMI and Mortality in Surgical Patients

Background: While obesity is commonly associated with increased morbidity and mortality, in patients with chronic diseases, it has have been associated with a better prognosis, a phenomenon known as the 'obesity paradox'. Objective: We investigated the relationship between mortality, length of hospital stay (LOHS), and body mass index (BMI) in patients hospitalized to general surgical wards. Methods: We extracted data of patients admitted to the hospital between January 2011 and December 2017. BMI was classified according to the following categories: underweight (< 18.5), normal weight (18.5–24.9), overweight (25–29.9), obesity (30–34.9) and severe obesity (≥ 35). Main outcomes were mortality at 30-day mortality and at the end-of-follow-up mortality), as well as LOHS. Results: A total of 27,639 patients (mean age 55 ± 20 years; 48% males; 19% had diabetes) were included in the study. Median LOHS was longer in patients with diabetes vs. those without diabetes (4.0 vs 3.0 days, respectively), with longest LOHS among underweight patients. A 30-day mortality was 2% of those without (371/22,297) and 3% of those with diabetes (173/5,342). In patients with diabetes, 30-day mortality risk showed a step-wise decrease with increased BMI: 10% for underweight, 6% for normal weight, 3% for overweight, 2% for obese and only 1% for severely obese patients. In patients without diabetes, 30-day mortality was found to be 6% for underweight, 3% for normal weight and 1% across the overweight and obese categories. Mortality rate at the end-of-follow-up was 9% of patients without diabetes and 18% of those with diabetes (adjusted OR = 1.3, 95% CI, 1.2–1.5). In patients with diabetes, mortality risk showed an inverse association with respect to BMI: 52% for underweight, 29% for normal weight, 17% for overweight, 14% for obesity and 7% for severely obese patients, with similar trend in patients without diabetes. Conclusions: The results support the ‘obesity paradox’ in the general surgical patients as those with and without diabetes admitted to surgical wards, BMI had an inverse association with short- and long-term mortality.