Clostridium difficile infection (CDI) is commonly diagnosed with the polymerase chain reaction (PCR), but this test finds a high percentage of false positives, so their use and interpretation in CDI is a challenge in the clinical practice. That is why it is necessary to define an algorithm to optimize the use of PCR that considers clinical characteristics to classify patients with diarrhea as CDI or without CDI.
To identify a predictive algorithm with the clinical features that best classify patients with CDI vs. without CDI, to help physicians in making decisions to request PCR.
Materials and methods
A case-control study was conducted at Fundación Valle del Lili. The population was inpatients between 2012 and 2016, with 18 or more years, and diarrhea, abdominal pain, or other nonspecific gastrointestinal symptoms who underwent PCR. Cases were defined as patients with positive PCR for C. difficile and as controls patients with negative PCR for C. difficile. Predictive algorithms to classify patients was constructed using a classification tree, classification and regression tree (CART).
A total of 149 patients were included (48 cases with positive PCR and 99 controls with negative PCR). The CART has a high capacity to classify patients with a negative PCR correctly. It includes variables about the history of antibiotics use, the use of proton-pump inhibitor, the use of ranitidine, and the use of antifungal drugs. The CART showed sensitivity 64.6%, specificity 85.8%, positive predictive value 68.8%, negative predictive value 83.3%. and AUC 79.7%.
CART had good specificity and a high negative predictive value; it could be considered as an algorithm to identify conditions that indicate when it is not necessary to perform a PCR test in a patient symptom of CDI.