MO994POST-TRANSPLANT DIABETES MELLITUS PREDICTORS AFTER PANCREAS-KIDNEY TRANSPLANT

Abstract

Abstract Background and Aims Pancreas-kidney transplantation (PKT) is an established treatment for selected patients with type 1 diabetes mellitus (T1DM) with either advanced or end-stage chronic kidney disease. We aimed to evaluate predictors for post-transplant diabetes mellitus (PTDM) at 24 months of follow-up after PKT. Method We analysed a longitudinal cohort of patients with chronic kidney disease and T1DM who underwent PKT at a tertiary referral centre from January 2011 to December 31, 2017. PTDM (post-transplant diabetes mellitus) was defined as the persistent need for pharmacologic treatment of diabetes mellitus after transplant. Continuous variables were recorded as means (±SD) or as medians (25-75% interquartile range), accordingly to data distribution. Categorical variables were recorded as proportions. Comparisons were made using t-tests, Wilcoxon rank-sum tests, or X2 test as appropriate. Unadjusted and adjusted multivariate logistic regression models were fitted to identify risk factors for PTDM at 24 months of follow-up. STATA 14.2 statistical package was used and p <0.05 was considered statistically significant. Results From a total of 72 patients who underwent PKT during the study period, 47 patients completed at least 24 months of follow-up. The mean age at transplantation was 37±8 years, 30 (64%) were male, 43 (91%) were white and 4 (9%) were black. At 24-months of follow-up, the prevalence of diabetes mellitus was 28% (13/47). Number of mismatches, accumulated prednisolone dose, tacrolimus levels, previous rejection episodes, body mass index, cytomegalovirus status, serum creatinine, lipase, amilase at discharge were not associated with PTDM. Patients with persistent diabetes at 24-months of follow-up had a significantly longer pretransplant diabetes duration (29±10 versus 23±6 years; p-value = 0.026), had higher pretransplant LDL-cholesterol (137±46 versus 100±27 years; p-value = 0.003), and lower C-peptide at post-transplant hospital discharge (3.0 [2.0-3.3] versus 5.8 [4.5-6.5]; p-value = 0.004). Likewise, in univariate logistic regression models, longer diabetes duration and higher pretransplant LDL-cholesterol were found to be risk factors for PTDM at 24-months (OR 1.11, CI95% 1.00-1.22, p-value = 0.041; OR 1.03, CI95% 1.01-1.06, p-value = 0.014; respectively), while higher C-peptide at post-transplant hospital discharge was found to be protector (OR 1.03, CI95% 1.01-1.06, p-value = 0.014). In a multivariate model including age, pretransplant diabetes duration, pretransplant LDL-cholesterol, and C-peptide at post-transplant hospital discharge, only C-peptide at post-transplant hospital discharge remained significant (OR: 0.41, CI95% 0.18-0.95, p-value = 0.04). Conclusion Despite previous results in other studies, and new data researched, in our multivariate model, only lower C-peptide at post-transplant hospital discharge revealed being a predictor of PTDM after PKT.