Drug-induced thrombocytopenia in the critically ill: a case-control study

Drugs are suspected when obvious causes of intensive care unit (ICU)-acquired thrombocytopenia have been excluded. It has been estimated that 10% to 25% of cases may be drug induced.

The objectives of this study were to evaluate the risk of thrombocytopenia associated with drug classes commonly used in the ICU.

Data concerning patients admitted for more than 48 hours between 1997 and 2011 were extracted from a research-purpose database. Patients with thrombocytopenia within the first 72 hours of admission and with diagnoses or interventions considered strongly associated with thrombocytopenia were excluded. Drug exposures were compared and adjusted for confounders using conditional logistic regression.

A total of 238 cases were identified after exclusions. Each case was matched according to sex, age, admission year, and admission unit with 1 control. In univariate analysis, quinolones (odds ratio [OR] = 1.56; 95% CI = 1.01-2.40) and extended spectrum β-lactams (OR = 1.71; 95% CI = 1.00-2.93) were significantly associated with an increased risk of thrombocytopenia. After adjusting for confounders, exposure to quinolones was the only drug class with a statistically significant increase in risk of thrombocytopenia (OR = 1.697; 95% CI = 1.002-2.873; P = 0.049).

In this study of ICU-acquired thrombocytopenia, we found no association between the exposures to several antibiotic classes, anticonvulsants, antiplatelet agents, nonsteroidal anti-inflammatory agents, and heparins and thrombocytopenia. As linezolid was not studied, no conclusions can be drawn concerning this agent. The statistically significant association between quinolones and thrombocytopenia warrants further investigation.