BACKGROUND There is paucity of data on long-term probability of remission in chronic idiopathic thrombocytopenic purpura (ITP). Aim was to study the course and factors influencing remission of chronic ITP. Chronic ITP was defined as thrombocytopenia persisting >6 months following initial diagnosis. METHODS Case-records of children with chronic ITP, aged <14 years, were reviewed in this retrospective study (1987-2006). RESULTS Two hundred seventy children were followed. Median age at diagnosis was 6 years. Median duration of follow up was 30 months (range 6-166). Isolated thrombocytopenia (even if <10 x 10(9)/L) in the absence of "significant" bleeds, by itself was not considered an indication for drug therapy. Sixty-seven (24.8%) children attained complete remission (CR) over a median period of 18 months (range 7-120). The probabilities of remission at 5 years for males and females were 24% and 39.6%, respectively (P = 0.01). The probability of achieving remission at 10 years in children <8 and > or =8 years was 51.2% and 34%, respectively (P = 0.02). The probability of remission at 5 years for children who received some treatment, versus no treatment was 31.4% and 27%, respectively (P = 0.8). Nine of 18 children, who underwent splenectomy, achieved CR. Intracranial hemorrhage (ICH) occurred in 11 (4%) cases. The time of occurrence of ICH from onset of symptoms varied from 6 to 55 months. CONCLUSIONS The predicted spontaneous remission rate with chronic ITP was 30% and 44% at 5 and 10 years, respectively. Platelet count at diagnosis and the treatment administered did not influence remission outcomes. Age <8 years and female gender were predictors of a favorable outcome.