Abstract: Objective To investigate the predictive value of absolute lymphocyte counts (ALC) for the relapse of primary immune thrombocytopenia (ITP) in children. Methods The clinical data of 189 children with initial diagnosis of ITP whose therapeutic effect was complete response or response in the Affiliated Hospital of Southwest Medical University from August 1, 2013 to August 30, 2018, were retrospectively screened and analyzed. According to the optimal cut-off value of ALC by receiver operating characteristic (ROC) curve, which can predict the relapse of ITP, the children were divided into 2 groups. Kaplan-Meier method was used to draw the non-relapse survival curves. The survival differences between two groups were compared by Log-rank test. Univariate and multivariate Cox proportional hazards regression model analysis were performed respectively to find the factors influencing the relapse of ITP. Results A total of 60 children relapsed and the recurrence time was 10.19 (1.4, 11.83) months. ROC curve showed that area under the curve (AUC) of ITP recurrence predicted by ALC was 0.757, 95%CI was 0.685-0.830, P < 0.05. The optimal cut-off value of ALC was 3.08×109/L, sensitivity and specificity values were 0.600 and 0.775. Cox proportional hazards regression model analysis showed that ALC ＜3.08×109/L was a risk factor for relapse in children with ITP (HR=2.468, 95%CI: 1.311-4.644, P＜0.05). Survival analysis concluded that non-relapse survival rate of ALC＜3.08×109/L group was significantly lower than ALC≥3.08×109/L group (Log-rank c2=24.313，P＜0.01). Conclusion ALC is a predicting factors for relapse in children with ITP.

Key words: thrombocytopenia, recurrence, child, ROC curve, immune thrombocytopenia, absolute lymphocyte counts