Identification of children with acute lymphoblastic leukemia at low risk for tumor lysis syndrome

Background and Objective: Tumor lysis syndrome (TLS) could occur before, during or after the initiation of chemotherapy in patients with cancers especially those with hematologic malignancies. This study was designed to determine the prevalence and predictors of TLS in children with Acute Lymphoblastic Lymphoma (ALL) and to develop a sensitive prediction rule to identify patients at low risk of TLS. Methods: In this cross-sectional study 160 children diagnosed by ALL in Ali-Asghar Children Hospital, Tehran (1996-2010) were recruited. TLS was defined as having two or more of the certain criteria. Predictors of TLS were determined using univariate and multiple logistic regression analyses. Results: TLS was diagnosed in 41 cases (25.6%). The most common laboratory abnormality was hypocalcaemia (30%) in these patients. The results of univariate analysis showed that splenomegaly (OR, 2.38; p=0.005), mediastinal mass (OR, 4.45; p=0.003), T-cell phenotype (OR, 4.66; p=0.001), central nervous system involvement (OR, 10.93; p=0.001), lactate dehydrogenase ≥2000 U/L (OR, 3.88; p=0.003), and white blood count (WBC) ≥20×109/L (OR, 4.18; p<0.001) were predictors of TLS in these cases. Multiple regression analysis of variables that were available at presentation identified CNS and renal involvement, mediastinal mass, and initial WBC ≥ 20 × 109/L as independent predictors of TLS. When all 4 of those predictors were absent at presentation (n= 83 patients), the negative predictive value of developing TLS was 92.22%, with a sensitivity of 82.93%. Conclusions: It could be suggested to evaluate the risk of TLS in all patients with hematologic malignancies before starting chemotherapy. Finding a model of independent factors to define a group of ALL children at low risk of TLS could be used to prevent the monitoring and high cost prophylactic treatment modalities.