Our 10-year surveillance of CO-CDI at a tertiary hospital in China indicated that 89 patients (7.3%) had CDI and 4.5% of these patients experienced at least one recurrence. This observation is line with our earlier study, which identified 8.9% of toxigenic C. difficile isolates from inpatients (10). There were two reasons for the small number of CO-CDI patients in this study. Unlike other studies of CO-CDI, which recruited patients admitted within 48 h, we only analyzed individuals who visited outpatient clinics, so it is likely that most patients with severe or difficult-to-treat CO-CDI were admitted to a hospital. Another reason for our smaller number of CO-CDI cases may be the poor awareness of CDI among general healthcare workers, which could have led to missed cases. In the study by Bauer et al., there was no specific request to test for C. difficile, which could have caused six of ten cases to be missed (11).
Our outpatients with CDI were older than those without CDI (56 vs. 47 years). However, the median age of our CDI patients was similar to that of inpatients with CDI in our previous study (56 years)(12). A previous study reported that patients with CO-CDI were younger than those with HO-CDI(5). However, the present study clearly showed that CDI patients tended to be older, in that nearly half of the cases (43/89) were 60 or more years-old. Moreover, the median age of the 10 patients from the clinical hematology department was 38 years, and this biased the total median age. Epidemiological studies in Western countries showed that CDI was more common in females than males(13), and a large study of 113 laboratories across England also reported that 67% of CO-CDI cases were in females(13, 14). However, males accounted for more than half of the outpatients in the present study, similar to the proportion of inpatients in our previous study(12). This difference may be due to gender differences in care-seeking behaviors among different countries.
More than half of our cases were from the clinical gastroenterology department, indicating that diarrhea or gastrointestinal discomfort were the most common reasons for visiting the clinic. Patients visiting this department were older than those who visited other outpatient departments. This is consistent with the interpretation that older individuals are more susceptible to CDI(15). Another interesting finding of the present study is that the positive rate for CDI was highest in our clinical hematology department, and the median patient age was youngest in this department. Our review of the records of these patients indicated that most of them were follow-up patients who received chemotherapy or haematopoietic stem cell transplants (HSCTs), and these cases of C. difficile were healthcare-facility acquired CO-CDI. There is evidence that patients receiving chemotherapy for haematological malignancies or HSCTs have an increased risk for CDI(16). Thus, it is important for clinicians to consider testing for C. difficile in outpatients who develop diarrhea and are older than 60 years or immunocompromised. Recurrences of CDI are serious, and the management of these patients is challenging. A meta-analysis found that the recurrence rate of CDI was 13 to 50% among all patients after an initial episode(17). The recurrence in the present study was 4.5%, similar to that reported by Tsai et al. (4.7%)(18), but lower than reported in another study (15.9%)(19). We speculate this may be because our CO-CDI patients were not as severely ill as hospitalized patients and because some of our patients were lost to follow-up due to visiting other hospitals.
The A + B + strains were the most common types in the present study, similar to our previous study(10). The epidemiology of C. difficile is region-specific. Thus, CDI cases in Europe are mostly from ribotypes RT002 and RT056(13). In the present study, ST-54, ST-35, and ST-2 were the most common STs, similar to our previous study which identified ST-54, ST-35, and ST-37 as the three most common STs among patients hospitalized with CDI(20). Although ST-2 was not among the major STs isolated from inpatients in our previous studies(20), ST-2 is a major ST of C. difficile isolated from patients with community-associated CDI in many European countries(21). Future studies are needed to compare the differences of ST diversity of C. difficile isolates in patients with community-onset and hospital-onset CDI. Interestingly, the distributions of STs differed among our different hospital departments. For example, ST-54 was the most common in the clinical gastrointestinal department and clinical hematology department, but ST-37 was the most common in the clinical infectious disease department. In general, the STs of inpatients and outpatients are similar, and a previous study showed that healthcare facilities and community settings had 79% of the same RTs(22). This may be because outpatients who were hospitalized were reviewed after discharge, especially in the hematology clinic and the infection clinic. Fortunately, none of the isolates was identified as the hypervirulent ST-1 (BI/NAP1/027), but there was one ST-11 (RT078) isolate from the clinical hematology department.
Vancomycin and metronidazole are the only two antimicrobial agents used to treat CDI in China. Fortunately, all toxigenic C. difficile isolates from our outpatients were sensitive to these two antibiotics and the MICs were low. Compared with our previous study, the isolates from outpatients and hospitalized patients had similar resistance profiles(10). However, we observed a low resistance to tetracycline, slightly lower than reported for hospital-acquired CDI but not significantly different from reports in the Asia-Pacific region(23). These differences in antimicrobial resistance may be attributed to differences in exposure to antibiotics in outpatients and individuals in the community, and to the use of different antibiotics in different regions. Another important finding of the present study is that nearly one-quarter of the isolates were multidrug-resistant, much lower than previously reported for isolates from inpatients(10). Based on this, we speculate that most of the strains we identified were from community, where antibiotic use is much less common.
There were some limitations in this study. Firstly, we only analyzed patients visiting clinics, and this may have reduced the number of CO-CDI for those who were admitted within 48 h. Secondly, we did not classify CDI as community-acquired, or healthcare-acquired. Thirdly, we were missing data regarding antibiotic exposure, comorbidities, and outcomes, and could therefore not identify risk factors associated with CDI. Therefore, further research is needed to address these limitations.