IDF21-0268 Association between chronic hepatitis B and diabetes complications: Analysis of 0.5 million Chinese with type 2 diabetes

Abstract

Background: China has a high prevalence of chronic hepatitis B (CHB) affecting 6-8% of the population. Up to 10% of people living in China have type 2 diabetes. The impact of comorbid CHB on prognosis of people with type 2 diabetes have not been well quantified. Aim: The aim of the study is to compare the incidence of cardiovascular-kidney complications and death between people with comorbid CHB and type 2 diabetes and those with type 2 diabetes alone. Method: We conducted a retrospective analysis of the territory-wide Hong Kong Diabetes Surveillance Database containing 521,623 people with physician-diagnosed type 2 diabetes who underwent baseline metabolic assessment between 1 Jan 2000 and 31 Dec 2019. CHB was defined by positive hepatitis B surface antigen on laboratory test. People without documented positive hepatitis B surface antigen were considered not to have CHB. Incident complications including cardiovascular disease (CVD) (coronary heart disease, stroke, peripheral artery disease), end-stage kidney disease (ESKD) and death were identified using ICD-9 codes from hospital discharge, procedure codes, laboratory data (estimated glomerular filtration rate [eGFR] <15 mL/min per 1.73m2 for definition of ESKD) and linkage to death registry from baseline until 31 Dec 2019. Multivariate Cox proportionate regression was used to derive hazard ratios (HR) (95% confidence interval [CI]) of CHB for each of the clinical outcome, adjusted for age, sex, diabetes duration, tobacco use, alcohol use, BMI, blood pressures, HbA1c, LDL-cholesterol, HDL-cholesterol, triglyceride, eGFR, pre-existing CVD (for ESKD and mortality outcome) and drug use including glucose-lowering drugs, statins, and renin-angiotensin system inhibitors at baseline. Results: 26,016 (5.0%) people with type 2 diabetes had CHB. People with CHB were younger (58.9 vs 62.0 years), had a male preponderance (60.9% vs 50.2%) but similar diabetes duration (2.0 vs 2.0 years) compared with those without CHB. Current or past use of tobacco (36.5% vs 30.4%) and alcohol (35.8% vs 29.6%) was more frequent in the CHB group. Metabolic profile including BMI, blood pressure, HbA1c and cholesterol levels were similar between the two groups. Over 3.8 million person-year follow-up, incident CVD was recorded in 45,345 people, incident ESKD in 20,810 people, and 85,444 people died. The respective incidence rates (per 1,000 person-year [95% CI]) of CVD were 11.9 (11.4, 12.4) vs 13.2 (13.1, 13.3), ESKD 6.3 (6.0, 6.7) vs 4.9 (4.8, 4.9), and all-cause death 27.3 (26.6, 28.1) vs 20.4 (20.3, 20.6) in people with and without CHB (p<0.05 for all comparisons). In multivariate Cox regression, CHB was associated with increased risks of ESKD (HR 1.53 [1.42, 1.65], p<0.001), all-cause death (HR 1.88 [1.81, 1.96], p<0.001) and death which excluded liver disease and hepatocellular carcinoma as primary causes (HR 1.48 [1.42, 1.55], p<0.001), but not with CVD (HR 1.01 [0.95, 1.07], p=0.716) after full adjustment. Discussion: Among Hong Kong Chinese with type 2 diabetes, CHB independently increases the risk of ESKD and death but not CVD. Reduced survival in people with comorbid CHB and type 2 diabetes is not fully explained by liver disease or hepatocellular carcinoma.