To the Editor We read with interest the publication by Polage and colleagues1 on overdiagnosis of Clostridium difficile infection (CDI) in patients with a positive nucleic acid amplification test (NAAT) result and a negative toxin immunoassay result. Although they showed more CDI-related complications and deaths in patients with positive toxin immunoassay results compared with patients with NAAT-positive and toxin immunoassay–negative results, we question whether their findings could have been owing to a confounding variable inherent to their study design. Specifically, the authors made the assumption that the matrix of stool specimens has no bearing on the ability of the immunoassay to detect toxin. However, stool is known to contain interfering substances that can inhibit diagnostic assays.2,3 Without controlling for the stool matrix (eg, Bristol scale, consistency, and other characteristics), the authors might have unintentionally compared patients with noninhibitory stool matrix types (patients with toxin-positive results) to patients with inhibitory stool matrix types (patients with toxin-negative results). Contending that noninhibitory stool matrix types (eg, watery diarrhea) correlate with more severe diarrheal disease compared with inhibitory stool matrix types, the authors might have compared outcomes in patients with more severe diarrheal disease to patients with less severe disease, independent of actual toxin concentrations. Therefore, further studies are needed to address the potential effect of stool matrix on toxin immunoassay results. Meanwhile, given that the best laboratory assay for diagnosing CDI is unknown, improving the selection of patients for CDI testing (excluding those with <3 episodes of loose stools in 24 hours and those who are taking laxatives) would limit testing to those who are most likely to benefit from anti-C difficile treatment.
Banaei N, Schroeder LF. Toxin Immunoassays and Clostridium difficile Infection. JAMA Intern Med. 2016;176(3):413. doi:10.1001/jamainternmed.2015.8525