Impact of neutropenia on clinical manifestations and outcome of Staphylococcus aureus bloodstream infection: a propensity score-based overlap weight analysis in two large, prospectively evaluated cohorts

Objectives: This study aimed to investigate whether neutropenia influenced mortality and long-term outcomes of Staphylococcus aureus bloodstream (SAB) infection. Methods: Data from two prospective, multicentre cohort studies (INSTINCT and ISAC) conducted at 20 tertiary care hospitals in six countries between 2006 and 2015 were analyzed. Neutropenic and severely neutropenic patients (defined by proxy of total white blood cell count <1000/µl and <500/µl, respectively, at onset of SAB infection) were compared with a control group using a propensity score model and overlapping weights to adjust for baseline characteristics. Overall survival and time to SAB infection-related late complications (SAB infection recurrence, infective endocarditis, osteomyelitis, or other deep-seated manifestations) were analyzed with Cox regression and competing risk analyses, respectively. Results: Of the 3187 included patients, 102 were neutropenic and 70 severely neutropenic at the time of SAB infection onset. Applying overlap weights yielded two groups of 83 neutropenic and 220 nonneutropenic patients, respectively. The baseline characteristics of these groups were exactly balanced. In the Cox regression analysis, we observed no significant difference in survival between the two groups (death during follow up: 36.1% in neutropenic vs. 30.6% in nonneutropenic patients; hazard ratio (HR): 1.21; 95% CI, 0.79–1.83). This finding remained unchanged when we considered severely neutropenic patients (HR: 1.08; 95% CI, 0.60–1.94). A competing risk analysis showed a cause-specific HR of 0.39 (95% CI, 0.11–1.39) for SAB infection-related late complications in neutropenic patients. Discussion: Neutropenia was not associated with a higher survival rate during follow up. The lower rate of SAB infection-related late complications in neutropenic patients should be validated in other cohorts.