Management of immunosuppressive therapy in liver transplant recipients who develop bloodstream infection

Background Data about the optimal management of immunosuppressive therapy in liver transplant (LT) recipients with bloodstream infection (BSI) are missing. We aimed to describe the management of immunosuppressive therapy at diagnosis of BSI in LT recipients and to assess its impact on 28-day mortality. Methods We performed a single-center retrospective study of all LT recipients diagnosed with BSI, over 10-year period. Multivariate Cox regression analysis of risk factors for all cause 28-day mortality was adjusted for the propensity score of being managed with "any reduction" in immunosuppressive therapy at the diagnosis of BSI. Results We identified 209 episodes of BSI in 157 LT recipients: 107 (68%) male, median age 54 (IQR 48-63) years. "Any reduction" was made in 90 (43%) cases including: dosage reduction of ≥1 immunosuppressive drug in 31 (15%), discontinuation of ≥1 immunosuppressive drug in 28 (13%), both dosage reduction and discontinuation in 13 (6%), complete withdrawal of immunosuppressive therapy in 18 (9%) cases. All-cause 28-day mortality rate was 13.4%, varying from 22% to 7% (P = .002) in cases with and without "any reduction". Cox regression showed septic shock (aHR 3.15, P = .007) and "any reduction" (aHR 2.50, P = .02) as independent risk factors for all-cause 28-day mortality, while Escherichia coli (aHR 0.38, P = .03) and source control (aHR 0.43, P = .04) were protective factors. The final model did not change after the introduction of the propensity score for "any reduction". Conclusions Any reduction in the immunosuppressive therapy was common and was associated with worse outcome in LT recipients developing BSI.