OBJECT: The aim of this study was to investigate the cause and outcome of intracranial hemorrhage (ICH) in children with cancer. METHODS: The charts of 51 children who underwent treatment for both cancer and ICH between January 1985 and January 2003 were retrospectively reviewed. Assessment tools included the Karnofsky Performance Scale (KPS), Glasgow Coma Scale (GCS), and the Fisher exact and Student t-tests. Among the 51 cases, 30 involved brain tumors, 19 leukemia, and two lymphoma. The treatment group (Group 1) comprised 36 patients who suffered ICH during cancer treatment; the posttreatment group (Group 2) consisted of the 15 patients who suffered ICH after the completion of cancer treatment. The types of ICH included 22 cortical, four subcortical, 17 subdural, five brainstem, one subarachnoid, one epidural, and one ventricular. Thrombocytopenia was present in nine patients (25%) in Group 1. More patients in Group 2 (87%) than in Group 1 (44%) underwent cranial radiation treatment. Patients in Group 1 experienced a higher incidence of coagulopathy (37%) and ICH-related death (25%) than those in Group 2 (0 and 7%, respectively). Decrease in KPS and GCS scores of greater than 30 and greater than 3, respectively, at the time of ICH were indicators of increased mortality. Of the 17 children with subdural ICH, 13 suffered the hemorrhage following treatment for hydrocephalus and three patients suffered ICH associated with thrombocytopenia. In the 33 children alive at the 3-month follow-up examination after the ICH, no difference existed in the mean KPS scores pre- and post-ICH. CONCLUSIONS: Treatment for hydrocephalus, coagulopathy, thrombocytopenia, and hemorrhage into the tumor were the most probable causes of ICH among patients in Group 1. Radiation-induced vasculopathy was a possible cause of ICH in the patients in Group 2. Significant decline in the patient's neurological status at the time of ICH is a poor prognostic factor, but those patients who survive cancer and ICH are likely to regain neurological function.
OBJECT: The aim of this study was to investigate the cause and outcome of intracranial hemorrhage (ICH) in children with cancer. METHODS: The charts of 51 children who underwent treatment for both cancer and ICH between January 1985 and January 2003 were retrospectively reviewed. Assessment tools included the Karnofsky Performance Scale (KPS), Glasgow Coma Scale (GCS), and the Fisher exact and Student t-tests. Among the 51 cases, 30 involved brain tumors, 19 leukemia, and two lymphoma. The treatment group (Group 1) comprised 36 patients who suffered ICH during cancer treatment; the posttreatment group (Group 2) consisted of the 15 patients who suffered ICH after the completion of cancer treatment. The types of ICH included 22 cortical, four subcortical, 17 subdural, five brainstem, one subarachnoid, one epidural, and one ventricular. Thrombocytopenia was present in nine patients (25%) in Group 1. More patients in Group 2 (87%) than in Group 1 (44%) underwent cranial radiation treatment. Patients in Group 1 experienced a higher incidence of coagulopathy (37%) and ICH-related death (25%) than those in Group 2 (0 and 7%, respectively). Decrease in KPS and GCS scores of greater than 30 and greater than 3, respectively, at the time of ICH were indicators of increased mortality. Of the 17 children with subdural ICH, 13 suffered the hemorrhage following treatment for hydrocephalus and three patients suffered ICH associated with thrombocytopenia. In the 33 children alive at the 3-month follow-up examination after the ICH, no difference existed in the mean KPS scores pre- and post-ICH. CONCLUSIONS: Treatment for hydrocephalus, coagulopathy, thrombocytopenia, and hemorrhage into the tumor were the most probable causes of ICH among patients in Group 1. Radiation-induced vasculopathy was a possible cause of ICH in the patients in Group 2. Significant decline in the patient's neurological status at the time of ICH is a poor prognostic factor, but those patients who survive cancer and ICH are likely to regain neurological function.