Epidemiology of Clostridium difficile-associated disease (CDAD): a shift from hospital-acquired infection to long-term care facility-based infection.

BACKGROUND: Previous epidemiological studies on Clostridium-difficile-Associated Diarrhea (CDAD) have focused on hospitalized patients with nosocomial transmission. However, increasing numbers of patients with CDAD are being admitted to acute-care hospitals from long-term care facilities (LTCFs) and the local community. The purpose of our study was to study the changing epidemiological trends of CDAD patients admitted to an acute-care hospital and examine factors contributing to this shift in epidemiology.
MATERIALS AND METHODS: This IRB-approved retrospective study included 400 randomly selected patients with a diagnosis of CDAD, admitted to an acute-care hospital between January, 2005 and December, 2010. CDAD was defined as ≥3 episodes of loose stools in <24 h with a positive Clostridium difficile stool toxin assay. The patients were divided into three groups: hospital-acquired CDAD, long-term care facility (LTCF)-acquired CDAD, and community-acquired CDAD. The groups were compared in terms of demographics, ICU admissions, hospital length of stay, co-morbidities, presenting complaint, and medication use. Patients who were hospitalized in the preceding 12 weeks or who had history of antibiotic use in the prior 8 weeks were excluded.
RESULTS: Final analysis included 258 toxin-positive CDAD patients. Only 53 (20.6%) patients had hospital-acquired CDAD. Patients from LTCFs (n=119, 46.1%) and the community (86 patients, 33.3%) comprised 79.4% of patients. The mean age for LTCF population was significantly higher than the hospital-acquired and community-acquired CDAD groups (p<0.0001). The presenting complaint was categorized as diarrhea or non-diarrheal symptom. Other non-diarrheal symptoms included fever, abdominal pain and altered mental status. Only 15.2% of LTCF patients had diarrhea as their presenting complaint (n=18) as compared to 29.1% of patients from the community (n=25; p<0.05). Most LTCF patients (n=101, 84.8%) had non-diarrheal symptoms as their presenting complaint as compared to only 61 patients from the community (70.9%) (p<0.05). Use of proton pump inhibitor (PPI) was more frequent in LTCF patients (73%) and patients with hospital-acquired CDAD (69.8%) as compared to patients with community-acquired CDAD (43%) (p<0.05). No valid indication was found for PPI use in 24.13% of LTCF patients and 32.1% of patients with community-acquired CDAD as compared to only 12.9% of patients with hospital-acquired CDAD.
CONCLUSION: These observations suggest that CDAD originated predominantly in patients from LTCFs (46.1%) and community (33.3%) rather than from hospitalized patients (20.6%). Diarrhea was the presenting complaint in LTCF patients in only 15.2% of cases. Hence, CDAD should be suspected if LTCF patients present with symptoms such as abdominal pain, fever, or altered mental status along with loose stools. Majority of the LTCF patients were found to be on PPIs, a risk factor for CDAD, with as many as 24% of these patients with no valid indication for their use.