Conference Coverage

PAS: Antibiotic risk factors for recurrent C. difficile unique in children




SAN DIEGO – Unlike in adults, inpatient administration of “high-risk” antibiotics such as cephalosporins, clindamycin, and aminopenicillins are not risk factors for recurrent Clostridium difficile infection in children, results from a multicenter study demonstrated.

“There are previous studies in which recurrent Clostridium difficile infection risk prediction models have been compiled for adult hospitalized patients,” lead study author Anthony Goudie, Ph.D., said in an interview in advance of the annual meeting of the Pediatric Academic Societies. “Our work is the first known risk prediction model specifically looking at the pediatric population.”

Dr. Anthony Goudie

Dr. Anthony Goudie

Dr. Goudie of the Center for Applied Research and Evaluation in the department of pediatrics at the University of Arkansas for Medical Sciences, Little Rock, said that an adult risk prediction model (J. Hosp. Med. 2014;9:418-23) for recurrent Clostridium difficile infection (rCDI) found no association between clinical diagnoses during hospitalization for the incident CDI (iCDI). However, the model did find a strong positive association between rCDI and increasing patient age, number of recent prior hospitalizations, and the administration of gastric acid suppressors, high-risk antibiotics, or fluoroquinolones during the iCDI hospitalization.

To test the validity of adult rCDI risk factors in a pediatric population, Dr. Goudie and his associates conducted a secondary analysis of data from 43 children’s hospitals in the United States reporting to the Pediatric Health Information System between 2009 and 2013. They defined iCDI by the discharge ICD-9 code 00.845 and the administration of metronidazole or oral vancomycin during the hospital stay, and they defined rCDI as readmission with CDI within 42 days of the end of iCDI treatment. Next, the researchers statistically modeled all cases of iCDI for rCDI and tested for predictors among several factors including age, race/ethnicity, insurance status, use of antibiotics including cephalosporins, clindamycin, and aminopenicillins, and all of the predictors associated with rCDI in the adult risk prediction model.

Of 7,798 iCDI cases reported during the study period, 567 (7.3%), experienced a subsequent rCDI. No association between rCDI was seen among children who were administered histamine-2 receptor blocker (H2RB) gastric acid suppressors (P = .81), or any of all generations of cephalosporins, clindamycin, aminopenicillins, aminoglycosides, carbapenems, and fluoroquinolone antibiotics (all P greater than .10). A positive association (P less than .05) was seen among children 1-3 years, those covered by Medicaid insurance, those who had community-acquired iCDI, those who had a prior hospitalization, those administered tetracycline antibiotics, and those who were administered proton pump inhibitors.

In addition, rCDI was significantly associated with a number of iCDI comorbid diagnoses, including Hodgkin’s disease; non-Hodgkin’s lymphoma; spondylosis, intervertebral disk disorders, or other back problems; cancer of bone or connective tissue; cancer of the brain or nervous system; maintenance chemotherapy and radiotherapy; and leukemia (P less than .05 for all).

“Of all of our results, we were especially surprised to find that compared to children covered by private health insurance, those covered by Medicaid had higher odds of having rCDI,” Dr. Goudie said. “While more study is needed, we think that Medicaid in this context may be a proxy for socioeconomic status and at least a part of the solution to reducing rCDI may lie outside the control of hospital-based quality improvement interventions.”

He acknowledged certain limitations of study, including the fact the researchers “have no clinical data on children who may have been seen at other hospitals before or after the incident or rCDI [was] identified at the children’s hospital.”

The researchers reported having no relevant financial conflicts.

[email protected]

On Twitter @dougbrunk

Next Article: