Clostridium difficile Recurrence Is a Strong Predictor of 30-Day Rehospitalization Among Patients in Intensive Care

OBJECTIVE. While incidence, mortality, morbidity, and recurrence rates of C. difficile infection (CDI) among the critically ill have been investigated, the impact of its recurrence on 30-day rehospitalization (ReAd), an important policy focus, has not been examined. DESIGN. Secondary analysis of a multicenter retrospective cohort study PATIENTS. Adult critically ill patients who survived their index hospitalization complicated by CDI METHODS. CDI was defined by diarrhea or pseudomembranous colitis and a positive assay for C. difficile toxins A and/or B. CDI recurrence (rCDI) was defined as diarrhea, positive C. difficile toxin and need for retreatment after cessation of therapy. Descriptive statistics and a logistic regression examined ReAd rates and characteristics, and factors that impact it. RESULTS. Among 287 hospital survivors, 76 (26.5%) required ReAd (ReAd+). At baseline, the ReAd+ group did not differ significantly from the ReAd-group based on demographics, comorbidities, APACHE II scores, or ICU type. ReAd+ patients were more likely to have hypotension at CDI onset (48.7% vs 34.1%, P=.025) and to require vasopressors (40.0% vs 27.1%, P=.038); they were less likely to require mechanical ventilation (56.0% vs 77.3%, P<.001). A far greater proportion of ReAd+ than ReAd-had developed a recurrence either during the index hospitalization or within 30 days after discharge (32.89% vs 2.84%, P<.001). In a logistic regression, rCDI was a strong predictor of ReAd+ (adjusted odd ratio, 15.33, 95% confidence interval, 5.68-41.40). CONCLUSIONS. Greater than 25% of all survivors of critical illness complicated by CDI require readmission within 30 days of discharge. CDI recurrence is a strong predictor of such rehospitalizations.