The incidence of thrombocytopenia in neurosurgical patients prescribed famotidine is unknown. Using hospital records of neurosurgery patients treated between July 2001 and July 2002, a retrospective cohort study was performed comparing platelet counts in patients treated with famotidine with a similar group of patients who were not prescribed an H2 antagonist. Patients were excluded if: 1) platelets were less than 150,000 prior to famotidine administration; 2) pre-drug and post-drug platelets were not drawn; 3) they were concurrently taking a potential thrombocytopenic inducing drug; or 4) disseminated intravascular coagulation, thrombocytopenic purpura, or any other confounding hematologic disorder developed. Seventeen of 50 (34%) patients on famotidine developed thrombocytopenia compared with 11 of 98 (11.2%) of those untreated (P = 0.002). In this retrospective study, neurosurgical patients on famotidine developed thrombocytopenia statistically significantly more often than those untreated. Although no clinically significant sequelae developed as a result of the thrombocytopenia, if these findings are confirmed by a prospective study, proton pump inhibitors and sucralfate, with their similar efficacy, may be a better choice for gastrointestinal prophylaxis in neurosurgical patients.