Predictors of Complications and Mortality in Hospital and Community Acquired Clostridium difficile Colitis - Do Their Outcomes Differ?

Purpose: To determine the factors for complications and mortality of Clostridium difficile colitis (CDC) and if the hospital acquired CDC have greater risk than community acquired. Methods: A retrospective study was conducted in an inner city urban university affiliated community teaching hospital located in an ethnically diverse neighborhood, during a 12 month period. Study included 63 patients diagnosed with Clostridium difficle colitis (CDC) by fecal toxin assays. Patients with no prior history of hospital admission or antibiotic exposure in the past 3 months were considered community acquired. Patients with recent hospital admission, acquiring infection while in the hospital stay or admission from a long term facility were considered hospital acquired. Out of 63 patients, 50 had hospital acquired CDC. Patients were compared by complication and mortality status to determine significant prognostic factors for these two outcomes using the chi-square and Wilcoxon rank sum tests. Multiple logistic regression analysis was used to determine independent risk factors for these outcomes. Results: Number of co morbidities, low albumin, African American race, sepsis, intravenous (I.V) vancomycin use and treatment at ICU/CCU were significant predictors for both complication and death. Female sex, CKD, PPI use, I.V primaxin use, creatinine, number of antibiotics used and need for colonoscopy were also significantly associated with complication. Since patients with community acquired CDC have no history of antibiotic use, effect of number of antibiotics and use of specific antibiotics on complication and mortality were determined only for patients with hospital acquired CDC. Independent significant prognostic factors for complication were low albumin, female sex, CKD, PPI use and I.V primaxin use (P<0.05). Taking these factors into account, adjusted odds ratios (OR) for complication comparing females and males, CKD and non-CKD patients, PPI users and nonusers, primaxin users and nonusers were 40.3, 13.9, 29.4 and 41.7, respectively; a significant 18% decrease in risk for complication was observed for each unit increase in albumin. Complications itself (p=0.002) together with treatment at ICU/CCU (p=0.04) were the independent significant prognostic factors for mortality. Adjusted for the effect of each other OR for death comparing patients with and without complications and by treatment at ICU/CCU were 58.8 and 13.5, respectively. Hospital acquired and community acquired CDC patients did not differ significantly on complication and mortality and on risk factors for these outcomes. Conclusion: Patients with hospital acquired CDC did not have an increased mortality compared to community acquired CDC. (Data from Wyckoff Hts. Med. Ctr.).Table: Table. Comparisons between hospital and community acquired CDC patients on outcomes and risk factorsTable: Table. Independent significant prognostic factors for complications