Platelet Transfusion Practices for Critically Ill Children: A 28-Day Point-Prevalence Study in Three European Countries in 2023.

OBJECTIVES

To study the prevalence, indications, thresholds, and outcomes of platelet transfusions in critically ill children in three European countries.

DESIGN

International 4-week point-prevalence study in 2023.

SETTING

Forty-four PICUs across Spain, the United Kingdom, and Italy.

PATIENTS: PICU patients of 1 month to 17 years old receiving platelet transfusions.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN

RESULTS

During four prespecified 7-day blocks, 110 of 2713 (4.1% [95% CI, 3.4-4.9%]) patients received at least one platelet transfusion, accounting for 253 transfusions. The primary indication was a low platelet count (PC) (54% [95% CI, 48-60%]), and 83% (95% CI, 77.7-87.4%) were prophylactic indications. Active bleeding was the main reason for transfusion in only 17% of cases (12.6-22.2). Median pretransfusion PC was 36 × 10 9 cells/L. PC differed significantly across diagnostic subgroups and indications. It was higher in postsurgery cardiac (83 × 10 9 cells/L) and nonsurgery cardiac patients (61 × 10 9 cells/L) and lower in septic shock (24 × 10 9 cells/L) and hemato-oncological conditions (23 × 10 9 cells/L). Posttransfusion (2 hr) PC increment was 42 × 10 9 /L, but by 12-24 hours it decreased to 13 × 10 9 /L, regardless of patient diagnosis or reason for transfusion. Viscoelastic testing was performed in only seven events (2.8%). The 28-day PICU mortality was 21.8% (95% CI, 14.6-30.4%). The number and total dose of transfusions were associated with mortality, even after adjusting for the Pediatric Index of Mortality III score.

CONCLUSIONS

In 44 European PICUs in 28 days during 2023, 4% of admitted patients received platelet transfusions. PC was the primary reason for transfusion, with substantial variability between subgroups and indications. There were a high number of prophylactic transfusions, and exceeding the recommended restrictive threshold was common. The observed change in the patients' PC is not related to the reason for the platelet transfusion or the patient's diagnosis. Viscoelastic testing is still rarely performed. These results support the need for pediatric trials to define safe and effective platelet thresholds and to evaluate the role of other transfusion triggers in critically ill children.