In a previous study we tried to assess the clinical usefulness of platelet count (PIC) to confirm whether postoperative pulmonary embolism could be suspected early. Unexpectedly, the 19 patients who subsequently developed pulmonary embolism had significantly lower mean PIC levels even before surgery. In an attempt to discover whether the preoperative PIC levels were associated with a different incidence of postoperative blood loss, we decided to retrospectively study the relationship between preoperative PIC levels and the consequences of blood loss. There were 459 consecutive patients undergoing hip surgery. After excluding 5 patients who died during the first 3 postoperative days, and 16 patients who bled from a definitive anatomic site, there were 438 patients. Blood loss was considered to be excessive when two or more of the following conditions were present: (1) total transfusion requirements exceeding 1,000 ml whole blood or 2 units of packed red cells; (2) a drop in hemoglobin level of 5 g/dl or more, and (3) a hemoglobin level below 8 g/dl at any moment during the first 8 postoperative days. Blood loss was considered to be excessive in 91 patients. Preoperative PIC levels were significantly lower in these patients as compared to patients without the condition (204 ± 52 vs. 236 ± 79 × 109 liter–1; p = 0.0002). When patients were classified according to the quartiles of preoperative PIC, the odds ratio of developing excessive blood loss was 0.69 (95% CI: 0.38-1.26) in patients in the second quartile; 0.57 (95% CI: 0.30-1.06) in the third quartile, and 0.27 (95% CI: 0.13-0.57) in patients in the highest quartile. After adjusting for age, sex, type of surgery and type of prophylaxis, the preoperative PIC levels maintained a statistically significant inverse correlation with postoperative blood