Comparative evaluation of serotonin toxicity among Veterans Affairs patients receiving linezolid and vancomycin

Despite the theoretical risk of serotonin toxicity (ST) with linezolid, "real-world" clinical evaluations of the risk of ST in patients receiving linezolid have been limited to case reports and noncomparator studies. An observational, matched-cohort study was conducted to evaluate the risk of ST among hospitalized patients who received linezolid or vancomycin at the Upstate New York Veterans Affairs Healthcare Network (Veterans Integrated Service Network 2 [VISN-2]). Matching criteria included VISN-2 hospital, hospital ward, prior hospital length of stay, age, and baseline platelet counts. The patients' electronic medical records were evaluated for symptoms consistent with ST and the Hunter serotonin toxicity criteria (HSTC) using an intensive, natural word search algorithm. The study included 251 matched pairs. Demographics and comorbidities were similar between groups. Over half of the study population received at least one concurrent medication with serotonergic activity. Receipt of agents with serotonergic activity was more pronounced in the vancomycin group, and the higher frequency was due to concomitant antihistamine and antiemetic use. Antidepressant use, including selective serotonin reuptake inhibitors (SSRIs), was similar between groups. No patients in either group were found to meet the criteria using the word search algorithm for ST. Fewer linezolid patients than vancomycin patients met the HSTC overall (3.2% versus 8.8%) and when stratified by receipt of a concurrent serotonergic agent (4.3% versus 12.4%). Of the patients meeting the HSTC, most had past or present comorbidities that may have contributed to or overlapped the HSTC. This study of hospitalized patients revealed comparably low frequencies of adverse events potentially related to ST among patients who received linezolid or vancomycin.