Platelet transfusion threshold in patients with upper gastrointestinal bleeding: a systematic review.

BACKGROUND

There exists uncertainty as to the optimal platelet values when managing patients with nonvariceal upper gastrointestinal (GI) bleeding. GOALS AND STUDY: A systematic review was carried out to determine the optimal approach when managing patients with thrombocytopenia in the setting of nonvariceal upper GI bleeding.

RESULTS

Eighteen of 803 potential articles were selected and reviewed, including 4 randomized controlled trials and 6 cohort studies. The only empirical clinical data available pertained to the management of hematology or oncology patients. There was no high-level evidence that determined the proper threshold of platelet transfusion specifically in GI bleeding. We were, therefore, limited to include principally consensus opinions, recommendations, and guidelines for platelet transfusion trigger as they apply to the treatment (including prophylaxis) of bleeding in general, with a paucity of data addressing major bleeding, let alone bleeding from a gastroenterologic origin. Randomized clinical trials were individually underpowered in allowing definitive conclusions, even though resulting recommendations were supported by similarly underpowered retrospective and prospective observational studies.

CONCLUSIONS

There exist a paucity of data to recommend optimal therapeutic platelet count targets in patients with active GI bleeding. Based principally on expert opinion recommendations, we propose a count of 50×10/L. Some professional associations have suggested in very specific clinical settings (postcardiopulmonary bypass surgery or central nervous system trauma) a higher value of up to 100×10/L. Properly designed randomized trials are required to more precisely address this important clinical question.