Background:

C. difficile is a predominant cause of nosocomial diarrhea. It is most commonly associated with prior antibiotics use, and seen in conjunction with other risk factors such as increasing age, severe underlying illness, impaired gastrointestinal motility, tube feedings, and proton pump inhibitors. Recently, the emergence of a hyper–virulent strain, NAP–1, and its increased severity and mortality has heightened concern about this entity. New guidelines have been proposed emphasizing a more precise approach to diagnose, risk stratifying patients and use of oral vancomycin over metronidazole in severe cases. We were interested in determining if these recommendations were being followed in our institution.

Methods:

Retrospective chart review of 100 adult patients admitted through the ER clinically suspected by the admitting physician to have C. difficile infection. We reviewed patients’ clinical presentations (fever, diarrhea, abdominal pain, and ileus), lab findings (WBC, albumin, creatinine, and stool toxin assay), any endoscopic findings present. We reviewed the admitting physician’s approach to diagnosis and therapy. Patients were stratified into mild, moderate or severe infection.

Results:

Average age was 63 years. 54% were male, 77% were admitted to general medicine wards, and 23% were admitted to the ICU. 66% had been exposed to antibiotics within the past 6 weeks. The major presentations were diarrhea 97% and abdominal pain 38%. The stool toxin assay was performed on 95 patients at least once and was positive in 26%. Classified by CDI severity score system 51% were mild to moderate, and 49% were severe. The potentially offending antibiotic was stopped 6/66 times. Metronidazole was used in 88% of patients with mild disease, and vancomycin in 11%. In patients with severe disease, metronidazole was still predominant, used in 77%. 94% patients defined as having mild disease responded well to the antibiotics, but in patients with severe disease, only 60% improved on metronidazole and 72% on vancomycin. There were 10 mortalities in total, 8 of whom were among severe cases. Of those 8 patients, 7 were treated with metronidazole.

Conclusions:

This study shows the diagnosis of C. difficile was not robust, and the treatment inconsistent with the current guidelines. Our study has significant limitations; it is unclear in this retrospective study how many of our patients truly had C. difficile or whether the outcomes were related to the infection. However, we suspect that our approach and findings are consistent with what occurs in other community hospitals. We are concerned that many physicians are unaware of the change in risk associated with C. difficile and believe that, in the interest of patient safety that it is imperative to adopt a structure approach to diagnosis and to treatment of C. difficile.