Intracranial Hemorrhage (ICH) in Patients (pts) with Newly Diagnosed Leukemia: Incidence and Effect on Outcome.

Hemorrhage and infection carry significant morbidity and mortality in pts with leukemia. ICH can be a major clinical problem despite the current advances in supportive care. To analyze the incidence, outcome of ICH, and to identify potential risk indicators, we have conducted a retrospective analysis of pts with leukemia at M.D. Anderson Cancer Center (MDACC). We first reviewed all brain CTs of pts newly referred to MDACC from 1/2003 to 11/2003 with diagnosis of leukemia, myelodysplastic syndrome, or myeloproliferative disorder. Among 1185 pts (275 AML, 78 ALL, 163 CML, 296 CLL, 206 MDS, 47 MPD, 111 Other), 13 (1%) had ICH (8 AML, 3 APL, and 2 CML-myeloid blastic crisis). Based on this preliminary analysis, we limited this analysis to pts with newly diagnosed AML who received induction chemotherapy between 1/1998 and 11/2003. Among 993 such pts, 28 (3%) had ICH (parenchymal bleed in 12, subarachnoid hemorrhage in 4, subdural hematoma in 6, hemorrhagic infarct in 4, subarachnoid hemorrhage + subdural hematoma in 1, subarachnoid hemorrhage + subdural hemorrhage + parenchymal hemorrhage in 1) during the induction period (within 6 months after induction chemotherapy). Pt characteristics are shown in the table. Median time from induction therapy to ICH was 27 (0–122) days. In 16 (57%) pts death occurred before response to therapy could be determined (0–37 days). Thirteen out of 16 (81%) pts died of ICH as an immediate cause of death, and three of 16 (19%) died in the setting of multi-organ failure. Of 12 remaining pts whose disease status was assessed, 8 pts (66%) achieved complete remission (CR). Univariate analysis including age, gender, APL vs non-APL, cytogenetics, performance status, protective environment treatment, presence of antecedent hematologic disorder (AHD) and/or MDS, WBC< hemoglobin, platelet, PT, PTT, b2MG, bilirubin, creatinine, LDH, % of bone marrow blasts, % of peripheral blood blasts, and CR vs non-CR, indicated that APL, cytogenetics, thrombocytopenia, PT, bilirubin, and non-CR were associated with ICH. Differences in PT and bilirubin between the two groups, however, were clinically not significant. In a multivariate logistic regression analysis, APL remained as the only major predictive factor (odds ratio 3.35). In conclusion, ICH is limited to pts with myeloid malignancies, with an incidence of 3% in AML pts receiving induction therapy, with APL pts (13%) at greatest risk. ICH is associated with poor prognosis related to immediate mortality.