Potential Risk Factors Associated with Thrombocytop in a Surgical Intensive Care Unit

We conducted a retrospective chart review of 193 patients admitted during a 3-month period to determine the frequency of and potential risk factors associated with thrombocytopenia, and the association of acquired thrombocytopenia with length of stay in a surgical-trauma intensive care unit (SICU) and mortality. All records were reviewed beginning 24 hours after admission. Patients were followed for the duration of SICU stay or until death. Data collected and analyzed as potential risk factors for thrombocytopenia were age, gender, admitting diagnosis, classification (trauma, surgical, medical), APACHE II score, medical history, all scheduled drugs with start and stop dates, select laboratory values, arterial or central line placement, and complications. Thrombocytopenia occurred in 25 (13%) patients. These patients were more likely (p<0.05) than those without thrombocytopenia to have the following potential risk factors: presence of a central or arterial line (76% vs 46%, p<0.025), nonsurgical diagnosis (60% vs 37%, p<0.05), diagnosis of sepsis (p<0.001), and administration of phenytoin (p<0.01), piperacillin (p<0.005), imipenem-cilastatin (p<0.001), and vancomycin (p<0.005). A longer SICU stay (mean 21 vs 4.5 days, p<0.05) and increased mortality (16% vs 4%, p<0.05) were significantly associated with thrombocytopenia. Cefazolin administration was significantly associated with nonthrombocytopenia (p<0.05). Factors not associated with thrombocytopenia were age, gender, and administration of histamine2-receptor antagonists, heparin, enoxaparin, penicillins, ceftazidime, ceftriaxone, chloramphenicol, and amphotericin B. A central or arterial line was the only factor associated with the development of thrombocytopenia in a multiple linear regression analysis (p=0.0003, multiple r=0.2580). Thrombocytopenia is not a common occurrence in the SICU, but is associated with a longer SICU stay and increased mortality.