Methods: This was a retrospective chart review of all patients with MRSA BSI diagnosed at Temple University Hospital in Philadelphia, PA from 2011 to 2013 who survived the hospital stay and were readmitted to our hospital within 30 days. Demographic data and information on initial admission and readmission were analyzed.
Results: There were 71 readmissions among 63 patients with MRSA BSI. The median age was 50 and 61% were male. 27% of readmitted patients identified as active injection drug users (IVDU). 48% of patients had an infectious disease (ID) consultation during the initial admission. 8 patients had definitive endocarditis by Duke Criteria. The most common reason for readmission (58%) was suspected infection. 13% of readmissions were for lung disease or heart failure unrelated to MRSA BSI, 6% were for elective procedures, and 4% each were for metabolic disorders and cancer complications. 14 patients (accounting for 17 readmissions) had evidence of MRSA relapse on readmission (14 with MRSA BSI, 3 without). MRSA relapse readmissions occurred earlier than other readmissions; median 2 vs 9 days, p 0.001. Of MRSA relapses, 3 had completed antibiotics, 8 were still on antibiotics, and 6 left against medical advice (AMA) without completing antibiotics. MRSA relapses did not differ from other readmissions with respect to ID consultation, IVDU status, or endocarditis. Of all patients leaving AMA, 76% were active IVDU. AMA discharges were sufficiently common in other readmissions (20%) that AMA was only modestly associated with readmission for MRSA relapse (RR 1.7, p 0.33).
Conclusion: The majority of 30-day readmissions among patients with MRSA BSI were for suspected infection and nearly a quarter of readmissions had evidence of MRSA relapse. Leaving AMA and IVDU were common among MRSA relapses, but also among other readmitted patients.
W. Jean, None
H. Pritchard, None
M. Jan, None
P. Axelrod, None