Abstract and Introduction
Abstract
Objectives: The 2010 Infectious Diseases Society of America/Society for Healthcare Epidemiology of America treatment guidelines for Clostridium difficile infections (CDI) recommend oral metronidazole for mild-to-moderate disease and oral vancomycin for severe disease. Given that disease severity is easily determined by the peripheral white blood cell count and serum creatinine level, a computerized decision support (CDS) pathway to guide treatment is inherently appealing. Because providers often override or ignore the computer-based alerts, the proposed CDS pathway should be justified before implementation.
Methods: We undertook this study to ascertain the frequency of nonadherence to CDI guidelines. Between October 1, 2007 and September 30, 2008, a total of 229 cases were screened and 78 cases were included in the study, which took place at a 661-bed acute tertiary care teaching hospital.
Results: During the year-long study of CDI cases at our tertiary care hospital, 61.5% (48/78) of the patients received an antibiotic regimen that was not recommended by the 2010 guidelines. Among the 35 patients with mild-to-moderate CDI, 85.7% (30/35) received the recommended treatment of oral metronidazole monotherapy; in contrast, among the 43 patients with severe disease, none (0/43) received the recommended treatment of oral vancomycin monotherapy (P < 0.01). Moreover, 17.9% (14/78) of patients received concurrent oral metronidazole and vancomycin, a regimen that is not recommended anywhere in the Infectious Diseases Society of America/Society for Healthcare Epidemiology of America guidelines and which may be associated with a poor outcome. Patients who received combination oral metronidazole and vancomycin were not more likely to have comorbidities or severe CDI compared with those who received a single antibiotic agent.
Conclusions: As a result of this study, we plan to educate our providers on the treatment of CDI through a CDS pathway in an effort to increase guideline adherence, decrease inappropriate antibiotic use, and potentially improve patient outcomes.
Introduction
Clostridium difficile infections (CDI) are a significant cause of healthcare-associated morbidity and mortality. Fluoroquinolone use has been associated with the emergence of a virulent strain, denoted NAP1/027. This strain has been associated with a higher rate of colectomies and deaths compared with previous outbreak strains. The Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA) issued clinical practice guidelines for CDI in 1995 and 2010. The 1995 guidelines recommend metronidazole for the majority of cases, whereas vancomycin is reserved for metronidazole failures. The 2010 guidelines recommend oral metronidazole for mild-to-moderate infection and oral vancomycin for severe infections. Vancomycin administered per os and per rectum, if ileus is present, with or without intravenously administered metronidazole, is recommended treatment for severe, complicated CDI.
Among those cases of CDI with mild-to-moderate and severe disease, the classification is based on the peripheral white blood cell count and serum creatinine level. These guidelines were affected by data showing that oral vancomycin was superior to oral metronidazole in cases with severe CDI; in contrast, among cases with mild CDI, metronidazole was equally efficacious as oral vancomycin. The combination of oral metronidazole and oral vancomycin was associated with a worse outcome than treatment with a single oral agent. Neither the 1995 nor the 2010 IDSA/SHEA guidelines recommend concurrent oral vancomycin and metronidazole for mild-to-moderate or severe CDI. Despite these data, because of the onslaught of severe nosocomial epidemics of CDI, we anecdotally noted the use of concurrent oral metronidazole and vancomycin at our institution. Given that the addition of computerized physician order entry/clinical decision support (CPOE/CDS) pathways often leads to providers either overriding or ignoring the alerts, we first sought to determine the frequency of nonadherence to the CDI treatment guidelines. We chose a time period that occurred after the introduction of CPOE but before the implementation of significant antibiotic-related CDS pathways.