P601Body mass index and in-hospital mortality among patients with takotsubo syndrome: a nationwide retrospective cohort study in Japan

Abstract

Background: Although previous studies have reported ?obesity paradox? of mortality among patients with acute myocardial infarction or heart failure, no study has investigated whether ?obesity paradox? exists among patients with takotsubo syndrome (TS). Purpose: To examine the association between body mass index (BMI) and inhospital mortality among patients with TS. Methods: This is a retrospective cohort study using the Diagnosis Procedure Combination nationwide inpatient database in Japan. We identified TS patients aged ≥20 years who were admitted to acute-care hospitals from July 2010 and March 2014. Eligible patients were categorised into three groups according to BMI at admission: underweight (<18.5 kg/m2), normal-weight (18.5-24.9 kg/m2), and overweight (≥25.0 kg/m2). The outcome was in-hospital mortality. We examined the association between BMI groups and in-hospital mortality in a multivariable logistic regression model with adjustment for patient demographic and 16 comorbidities. Results: We identified 4,689 eligible patients with TS (mean age, 73.2±11.3 years; female, 78.7%; mean BMI, 21.1±3.9 kg/m2) hospitalised at 723 acutecare hospitals, including 1,171 (25.0%) patients in the underweight group, 2,855 (60.9%) patients in the normal-weight group, and 663 (14.1%) patients in the overweight group. Patients in the overweight group was younger (mean age, 71.3±12.2 years) than those in the normal-weight group (73.0±11.2 years) and the underweight group (74.9±10.7 years) (p<0.001). There was no significant difference in sex distribution among three groups. The proportions of several comorbidities (chronic pulmonary disease, rheumatic disease, psychiatric disease, and pneumonia) were significantly higher in the underweight group than in other groups. The in-hospital mortality was highest in the underweight group (6.7%), followed by the normal-weight group (4.2%), and was lowest in the overweight group (3.9%) (p for trend = 0.002). In the multivariable logistic regression analysis, adjusted odds ratios (95% confidence intervals) for in-hospital mortality were 1.46 (1.06-2.00) in the underweight group, 1.00 (reference) in the normal-weight group, and 0.93 (0.58-1.49) in the overweight group. The multivariable logistic regression analysis also showed that higher in-hospital mortality was significantly associated with higher age, male sex, lower consciousness level at admission, malignancy, chronic liver disease, chronic renal failure, sepsis, pneumonia, cerebrovascular disease, and acute renal failure. Conclusion: This large-scale cohort study suggests that there may be ?obesity paradox? of in-hospital mortality among patients with TS.