Predictors of Hospital-Acquired C difficile Infection in Children With Diarrhea Identified

Abdominal pain, prolonged hospitalization and antibiotic use prior to diarrhea onset, and receipt of meropenem and clindamycin were found to be independent predictors for hospital-acquired Clostridioides difficile infection (CDI) among children with antibiotic-associated diarrhea. These study findings were published in the American Journal of Infection Control.

Researchers conducted a retrospective case-control study to determine predictors of hospital-acquired CDI in pediatric patients with antibiotic-associated diarrhea. Included patients (age, ≥2-18 years) were evaluated for CDI via polymerase chain reaction testing more than 3 days after hospitalization. Patients who tested positive for hospital-acquired CDI were matched in a 2:1 fashion on the basis of age and sex against those without the infection (controls). Univariate and multivariate logistic regression were used to evaluate significant predictors.

The final analysis included 65 patients in the hospital-acquired CDI group and 130 in the control group. Overall, the median patient age was 72 (IQR, 46-123) months, and 64.1% were boys or men. Laboratory findings, including leukocyte and fecal leukocyte count, as well as C-reactive protein, serum creatinine, and albumin levels, did not significantly differ between patients in the CDI and control groups. Underlying diseases and comorbidities also did not significantly differ between these patients.

Patients in the CDI vs control groups were significantly more likely to have fever (30.8% vs 10.0%; P <.001), abdominal pain (64.6% vs 29.2%; P <.001), and grossly bloody stool (12.3% vs 1.5%; P =.003). The length of hospitalization prior to diarrhea onset also was significantly increased among patients with CDI (14 [IQR, 9-18] vs 7 [IQR, 5-10] days; P <.001), and a greater percentage of these patients were hospitalized for at least 14 days prior to diarrhea onset (56.9% vs 10.8%; P <.001).

Further comparisons showed a significantly higher rate of antibiotic use for at least 10 days prior to diarrhea onset in patients with vs without CDI (68.7% vs 23.1%; P <.001). Patients with CDI also were significantly more likely to have received clindamycin (29.2% vs 11.5%; P =.004) and meropenem (38.4% vs 3.8%; P <.001).

Factors found to be independent predictors of hospital-acquired CDI were as follows:

• Abdominal pain (odds ratio [OR], 7.940; 95% CI, 3.254-19.374);
• Hospitalization for at least 14 days prior to diarrhea onset (OR, 3.441; 95% CI, 1.034-11.454);
• Antibiotic use for at least 10 days prior to diarrhea onset (OR, 6.775; 95% CI, 1.882-24.388); and
• Receipt of meropenem (OR, 4.001; 95% CI, 1.098-14.577) and clindamycin (OR, 14.842; 95% CI, 4.496-49.000)

A scoring system to predict hospital-acquired CDI risk was developed using these factors, which had a sensitivity of 89.2%, a specificity of 74.6%, a positive predictive value of 63.7%, and a negative predictive value of 93.2%.

Limitations include the retrospective design, the single-center setting, and the relatively small sample number of patients with hospital-acquired CDI.

“The results of our study provide further evidence that limiting high-risk antibiotics such as carbapenems and clindamycin may reduce the incidence of hospital-onset CDI in pediatric patients,” the researchers concluded.