| Background: Acute rejection (AR) and graft loss (GL) that occur as a complication following kidney transplantation (KT) are a major cause of concern in patients with KT. Corticosteroids used as potent immunosuppressants in preventing AR and GL are associated with potentially serious side effects such as development or progression of cardiovascular diseases (CVD), new onset diabetes (NODM), infections and malignancies. Deaths with a functioning graft account for 40% of deaths following KT and CVD are a major cause of these deaths. Trials on tacrolimus-based regimen have found no significant difference in the AR or GL rates and a significant reduction in total cholesterol with steroid withdrawal regimens compared to steroid maintenance regimens. However, majority of these trials were short-term, of a duration of <=5 years and included low immunologic risk patients. Long-term effects of these regimens on GL and cardiovascular events such as stroke, myocardial infarction (MI) and deaths due to CVD (D-Cardio) are not known.;Objectives: The objectives of the study were to determine the optimal steroid withdrawal strategy that minimizes the incidence of both graft loss as well as cardiovascular events, amongst the five regimens: 1) steroid avoidance 2) 7-days steroid-withdrawal 3) 7-days to 6-months (6-month) steroid-withdrawal 4) 6-months to 12-months (12-month) steroid-withdrawal and 5) steroid maintenance, using a discrete event simulation model.;Methods: A discrete event simulation model was developed that included the following events: AR, GL, MI, stroke, other CVD, NODM, cancer, bacterial infection (BI), cytomegalovirus infection, fracture, D-Cardio, death due to GL and death due to other reasons. The United States Renal Data System registry that follows patients with transplantation was used to derive risk estimates of patients for the above events using parametric regressions adjusting for patients' demographic characteristics, immunologic risks and comorbidities. The estimates were then used to obtain Weibull distributions to transition a cohort of 10,000 patients, individually, in the model, using minimum of sampled time approach. The model was run for 20 years for base patient with mediocre-risk frequently seen in practice and for African-American patients and patients with a history of CVD.;Results: At the end of 20 years, base patients in the 6-month and the 12-month steroid-withdrawal groups were significantly less likely to experience MI (9.6-9.8% vs 12.2%), NODM (37.2%-42.4% vs 46.4%), BI (51.7%-57.6% vs 67.4%), fractures (51.1%-54.8%% vs 59.1%) and D-Cardio (24.5%-25.7% vs 28.8%), compared to steroid maintenance. The incidence of AR and GL were significantly higher in the steroid avoidance and 7-days steroid-withdrawal group compared to the steroid maintenance group (42.6%-51.4% and 57.9%-76.4% vs 30.5% and 40.9%). Compared to base patient, patients with a history of CVD and African-American patients were more likely to have a GL (46.6%-58.0% vs 40.7%-42.2%) and NODM (44.4%-44.9% vs 37.2%-42.4%).;Conclusion: At 20 years, the 6-month and the 12-month steroid-withdrawal regimens have benefits of significantly reduced rates of cardiovascular events with no significantly worse effects on AR and GL rates compared to steroid maintenance in mediocre-risk patients. Future simulation studies on a heterogeneous patient population are needed. |
