Approximately 34% of adult Clostridioides difficile infections (CDIs) are community associated, and possibly many more are underdiagnosed or underreported.Reference Lessa, Mu and Bamberg1,Reference Balsells, Shi and Leese2 Although many health systems have developed inpatient antimicrobial stewardship programs (ASPs) to help optimize CDI care, little information is available regarding community-associated CDIs managed in ambulatory care. Outpatient ASPs in the United States have largely focused on improving antibiotic prescribing by targeting specific conditions (eg, upper respiratory tract, otitis media, and pharyngitis) for improvement.Reference Sanchez, Fleming-Dutra, Roberts and Hicks3,Reference Barlam, Cosgrove and Abbo4 An opportunity exists for outpatient ASPs to optimize CDI prescribing strategies within this setting. We evaluated the management of CDI in the outpatient setting.
Methods
This study was a retrospective cross-sectional study to evaluate the management of patients diagnosed with a first episode of CDI in an ambulatory care setting between January 1, 2018, and June 30, 2019. The study was conducted at the Henry Ford Health System, which comprises >40 ambulatory medical centers located in southeastern and south-central Michigan. Clinics were categorized as primary care or specialty clinics. Specialty clinics included visits to a nonprimary care, medicine subspecialty clinic (eg, gastroenterology). Patients included were 18 years or older; they had a clinical diagnosis of C. difficile infection; and treatment had been initiated by the ambulatory clinic. Exclusion criteria were patients with severe CDI,Reference McDonald, Gerding and Johnson5 fulminant CDI,Reference McDonald, Gerding and Johnson5 immunocompromised patients, or patients who had undergone a fecal microbiota transplant. Immunocompromised was defined as acquired immunodeficiency virus, solid organ transplant, hematopoietic stem cell transplant, leukopenia/neutropenia, immunosuppressant drugs, inflammatory bowel disease, or prednisone ≥20 mg/day for >2 weeks. Clinical diagnosis was defined as patients with ≥3 stools reported in 24 hours and a positive CDI test using a 2-step algorithm performed by the health system’s core clinical microbiology laboratory.Reference McDonald, Gerding and Johnson5 Before patients were included, 3 pharmacy residents were trained and validated on data extraction, and they performed manual chart review to ascertain prescriber documentation of ≥3 stools in 24 hours. Demographic and clinical data were recorded, including age, sex, prior antibiotic exposure, CDI drug regimen and duration, clinical response, recurrent CDI within 14–60 days, and unanticipated emergency department or urgent care visits within 14 days after treatment initiation. Prior antibiotic exposure was defined as any systemic antibiotics within the previous 60 days. Clinical response was defined as symptom improvement or resolution and the absence of subsequent healthcare treatment within 14 days of treatment initiation (ie, telephone consultation for physician, revisit to physician). Appropriate management was defined as vancomycin 125 mg by mouth every 6 hours for 10–14 days per national practice guidelines.Reference McDonald, Gerding and Johnson5 Metronidazole 500 mg by mouth every 8 hours for 10–14 days was considered appropriate if medical record documentation specified that it had been prescribed as an alternative for cost, allergy, or limited resource availability. Data were analyzed using descriptive statistics: incidence, proportions, measures of central tendency, and dispersion. Our institution review board approved this study.
Results
We identified 126 patients with CDI diagnosed in an ambulatory clinic. Their median age was 58 years (IQR, 46–69) and 73% were women. The clinic most frequently visited was internal medicine (n = 50 patients, 39.7%), followed by specialty (n = 46 patients, 36.5%), family medicine (n = 24 patients, 19%), and walk-in (n = 6 patients, 4.8%). Furthermore, 49 patients (38.9%) had documented prior antibiotic exposure within 60 days, and the most common were amoxicillin/clavulanate (n = 16 patients, 32.6%), clindamycin (n = 7 patients, 14.2%), and ciprofloxacin (n = 6 patients, 12.2%). Table 1 provides a summary of CDI treatment and outcomes. Metronidazole (n = 82 patients, 65%) was prescribed most often for CDI treatment, followed by vancomycin (n = 42 patients, 33.4%). Also, 2 patients (1.6%) received regimens of either ciprofloxacin or ciprofloxacin and metronidazole for treatment. Overall, 37 patients (29.3%) were prescribed the appropriate antimicrobial therapy and duration. One patient had a documented cost barrier to receiving vancomycin therapy in their medical record. Among the subgroup of patients diagnosed with CDI in 2019, 13 of 33 (39%) were prescribed vancomycin.
Table 1. Outpatient C. difficile Infection (CDI) Therapy and Outcomes (n = 126 patients)
Moreover, 26 patients (20.6%) experienced clinical treatment failure. Clinical response occurred in 34 of 37 patients (91.9%) who received appropriate treatment and in 66 of 89 patients (74.2%) who did not (P = .025). Also, 10 patients (8%) had unanticipated emergency department or urgent care visits within 14 days, and 26 patients (20.6%) experienced recurrent CDI. The average time to recurrence was 30 days.
Discussion
We report a high proportion of suboptimal CDI management in ambulatory care clinics. Metronidazole was the most commonly prescribed regimen; it was associated with a 73% clinical response. Not surprisingly, vancomycin resulted in superior outcomes with a 92.9% clinical response. Overall, these prescribing patterns resulted in preventable poor outcomes: 20.6% of patients experienced clinical treatment failure, 8% required revisit to the ED or clinic, and 21% experienced recurrent CDI.
This study had several limitations. We included patients diagnosed as early as January 2018. The system ASP maintains CDI guidelines, which were revised in early January 2018, consistent with national guidelines.Reference McDonald, Gerding and Johnson5 However, we did not observe any evidence of improved prescribing for patients treated in 2019, 1 year after the guideline change. Among patients treated in the first 3 months of 2018, 8 received vancomycin and 18 received metronidazole, consistent with the overall cross-section (40 of 126, 32% prescribed vancomycin). Notably, 61.1% of patients had no documented recent antibiotic exposure. The assessment of prior antibiotic exposure was limited to chart documentation and accessible insurance claims, and the 38.9% rate observed is likely an underestimate. Additionally, metronidazole was defined as inappropriate if chart documentation did not mention the rationale for alternative therapy, and this approach may overestimate inappropriate prescribing. This study was observational and retrospective in nature, but it provides useful insight on ambulatory CDI management. We hypothesize that ambulatory care providers are unfamiliar with the updated recommendation to prescribe oral vancomycin first. E-mail newsletter education regarding the revised guidelines was provided to inpatient and outpatient prescribers in the health system in early 2018, but it appears to have been ineffective to communicate this practice change.
Ambulatory CDI treatment may represent a missed opportunity for institutional ASPs to minimize associated morbidity. A focused effort is needed to improve the quality of CDI management in outpatient setting.
Acknowledgments
We thank Nada Rida, PharmD, Kristin Soreide, PharmD, and Chuyin Fan, PharmD for assistance with gathering of this data.
Financial support
No financial support was provided relevant to this article.
Conflict of Interests
S.L.D. has received consulting fees from Spero Therapeutics, Tetraphase Pharmaceuticals and Allergan. All other authors have no conflicts of interest relevant to this article.
