A prospective cohort study on BMI levels and risk of acute pancreatitis

B Zhang ,Shou Ling Wu ,Jaiying Fan ,Y M Zhang ,Qi Sun ,Bo Tong ,Xiaoxi Fan ,Xiuli Men ,Xiaofeng Jiang ,J Zhang ,Rui Ji ,S Wang ,S H Chen ,Guoling Zhu

doi:10.3760/cma.j.cn112338-20201027-01286

Abstract

Objective: To investigate the effects of body mass index (BMI) levels at different baseline on the risk of new-onset acute pancreatitis (AP). Methods: The subjects were from the Kailuan Study Cohort and divided into 3 groups according to baseline BMI levels: BMI<24 kg/m2, normal weight; BMI 24-28 kg/m2, overweight; BMI≥28 kg/m2, obesity. The incidence of new-onset AP in these three groups was analyzed. The survival curve was plotted by Kaplan-Meier method, the cumulative incidence was calculated and tested by log-rank method. Multivariate Cox proportional hazards regression model was used to calculate HR of baseline BMI levels for AP. Results: A total of 123 841 subjects were included and followed up for (11.94±2.13) years, during which, 395 cases were found with AP. The incidence of AP was 2.67 per 10 000 person years in total population, and the incidences of AP were 2.20, 2.72 and 3.58 per 10 000 person-years in the normal, overweight and obesity groups, respectively. The cumulative incidences of AP was 0.32%, 0.40% and 0.49% in normal, overweight and obesity groups, respectively, which showed a significant inter-group difference by log-rank test (χ 2=13.17,P<0.01). The results of multivariable adjusted Cox proportional hazards regression model analysis indicated that obesity group (HR=1.45, 95%CI: 1.10-1.92) had a higher risk for AP compared with the normal BMI group. The subgroup analyses by age and sex showed that compared with the normal weight group,the HRs for AP in the obesity group was 1.58(95%CI:1.14-2.19) and 1.40(95%CI:1.03-1.90) among subjects younger than 60 years old and male subjects, respectively. After excluded onset AP within two years from baseline,with a control group from normal weight,the results of multivariate Cox proportional hazards regression model analysis indicated that the AP in the obesity group was 1.60 (95%CI: 1.18-2.15). Conclusion: Obesity may increase the risk of developing AP, particularly among young and middle-aged men.

Full Text