Delayed early lactate clearance after liver transplantation: A propensity score matching study
Nagai S, Takahashi K, Safwan M, Abouljoud MS, and Jafri SM. Delayed early lactate clearance after liver transplantation: A propensity score matching study. Hepatology 2018; 68(S1):105A.
Background: Lactate, a waste product of cellular metabolism, is mainly metabolized in the liver. Abnormally elevated lactate level reflects early liver graft dysfunction following liver transplantation (LT), and postoperative lactate levels considered as one of the real-time indicators for graft function. We hypothesized that trend of peri-operative lactate levels in LT recipients might be a reliable biomarker to predict post-transplant outcomes. This study aimed to investigate risk factors for early allograft dysfunction (EAD) and graft outcome after LT, focusing on perioperative lactate levels and early clearance. Methods: We retrospectively reviewed 265 decease donor LT between January 2011 and December 2014. EAD was defined as a peak values of aminotransferase >2000 IU/mL during the first week or an international normalized ratio ≥1.6 and/or bilirubin ≥10 mg/dL at day 7. Early lactate clearance was defined as reduction rate of lactate between the time of reperfusion and immediately after LT. Patients were categorized into the normal and delayed clearance groups. Cut-off value of lactate clearance was decided by Youden Index to predict occurrence of EAD. We used propensity score matching (PSM) between these two groups to estimate an impact of lactate clearance on incidence of EAD and one-year graft survival. Results: EAD occurred in 56 patients (21.8%). Cut-off values for immediate post-op lactate level and early lactate clearance predicting EAD were 3.45 mmol/L and 0.2 mmol/L/hour. After PSM, 120 patients in the normal clearance and 36 patients in the delayed clearance group were matched. Patients in the normal and delayed clearance groups were similar in regards to all covariates and the PSM model achieved absolute standard differences less than 0.10 (10%) for all covariates. The incidence rates of EAD were 23.3% (28 of 120) and 50.0% (18 of 36) in the normal and delayed clearance groups, respectively (P =0.002) (Figure 1a). One-year graft survival rates were 95.8% and 77.8%, respectively (P <0.001, log rank test) (Figure 1b). The risks of delayed lactate clearance for EAD and one-year graft loss were adjusted by intra-operative lactate level at the time of reperfusion of the liver graft. Base on these risk adjustment models, delayed lactate clearance was considered as an independent risk factor for EAD (odds ratio [OR] =3.49, 95% CI =1.55-7.85, P =0.002). The adjusted hazard of one-year graft loss significantly increased in the delayed clearance group (hazard ratio =6.69, 95% CI=2.13-20.99, P =0.001). Warm ischemia time (WIT) ≥45 minutes was significantly associated with initial high lactate level and delayed early lactate clearance (OR =3.26, P =0.002 and OR= 2.17, P =0.04 respectively). Conclusion: Delayed early lactate clearance was found to be a strong predictor for occurrence of EAD and poor graft outcomes. Shortening WIT may decrease a risk for delayed lactate clearance and improve graft outcomes.