Journal of Intensive Medicine (Oct 2024)

Outcomes of intracranial hemorrhage in critically ill patients with acute leukemia: Results of a retrospective cohort study

  • Antoine Herault,
  • Yannick Hourmant,
  • Etienne Lengliné,
  • Antoine Lafarge,
  • Eric Mariotte,
  • Michael Darmon,
  • Sandrine Valade

Journal volume & issue
Vol. 4, no. 4
pp. 508 – 514

Abstract

Read online

Background: Admission to the intensive care unit (ICU) is frequently required for patients with acute leukemia (AL) because of life-threatening complications such as intracranial hemorrhage (IH). In this study, we evaluated the impact of IH on survival and neurological outcomes in this population. Methods: This was a single-center retrospective cohort study including adult patients with AL requiring ICU admission and experiencing IH between 2002 and 2019 at Saint Louis Hospital. Leukemia type was determined according to the French–American–British classification. Brain imaging (either computed tomography or magnetic resonance imaging) was available for all the patients. The primary endpoint of the study was to describe the clinical and biological characteristics and evaluate the mortality and neurological outcome of patients hospitalized in the ICU with newly diagnosed AL and IH. The secondary endpoint was to identify predictive factors of IH in these patients. Results: Thirty-five patients with AL were included, median age of the patients was 59.00 (interquartile range [IQR]: 36.00–66.00) years. Twenty-nine patients (82.9%) had acute myeloid leukemia, including 12 patients with acute promyelocytic leukemia. Thrombocytopenia was constant, and 48.5% of patients had disseminated intravascular coagulation (DIC). At ICU admission, the median Sequential Organ Failure Assessment score was 5 (IQR: 3–9). The median time between AL onset and IH was 2.0 (IQR: 0.0–9.5) days. The ICU and hospital mortality rates were 60.0% (n =21) and 65.7% (n=23), respectively. In univariate analysis, mechanical ventilation and stupor were associated with mortality, but DIC and acute promyelocytic leukemia were not. Upon multivariate analysis, stupor or coma was the only factor significantly associated with a poor outcome (odds ratio = 8.56, 95 % confidence interval: 2.40 to 30.46). Conclusion: IH is associated with a high mortality rate in AL patients, with stupor or coma at the onset of intracranial bleeding being independently associated with poor outcomes.

Keywords