Massive Bleeding in Children With Cancer or Hematopoietic Cell Transplant: International, Multicenter Retrospective Study, 2017-2021.

Objectives: To characterize the epidemiology and management of massive bleeding events in children with cancer and/or hematopoietic cell transplant (HCT).

Design: Multicenter, retrospective cohort study.

Setting: Nineteen pediatric hospitals in Europe and United States.

Subjects: Children ages 0-21 years old with malignancy and/or HCT and massive bleeding admitted from January 1, 2017, to December 31, 2021.

Interventions: None.

Measurements and main results: Demographics, oncologic history, laboratory values, interventions, and PICU outcomes were collected. One hundred fifty-two bleeding episodes from 135 patients were analyzed. The median (interquartile range [IQR]) age was 7 years (2-14 yr). Forty-three percent (58/135) were female sex. Nineteen percent of children (26/135) had death attributable to hemorrhage. Forty percent had solid tumors and one-third had undergone at least one HCT. The majority of bleeding events occurred in the PICU (81/152, 53%). The median (IQR) platelet count at time of bleeding was 52 × 109/L (24-115 × 109/L), prothrombin time 18.5 seconds (15.2-24.8 s), activated partial thromboplastin time 42.2 seconds (33.2-56.0 s), and international normalized ratio 1.51 (1.21-2.11). To treat these bleeding events, 99% (148/152) of the time children received RBC transfusions, 84% (126/152) of the time plasma transfusions, 88% (132/152) of the time platelet transfusions, and less than one-fifth hemostatic medications. Half (77/152, 52%) of the time the children received high plasma ratios and half (73/152, 49%) received high platelet ratios. Pulmonary bleeding, oral/nasal bleeding, and receipt of prothrombin complex concentrate were each associated with greater odds of death attributed to hemorrhage: odds ratio (95% CI), respectively: 5.44 (2.250-13.171; p < 0.001); 3.30 (1.20-9.09; p = 0.021); and 3.24 (1.18-8.93; p = 0.023).

Conclusions: Children with malignancy and/or HCT have a high mortality rate from hemorrhage despite being hospitalized at the time of their bleeding event. The majority of children received balanced resuscitation. Definitive trials are needed to determine optimal hemostatic resuscitation practice in this population.