Objective: Determine whether the NAP1 strain identified by polymerase chain reaction (PCR)-based stool assay is correlated with CDI severity and clinical outcomes.
Methods: Medical records of adult patients with positive stool Xpert® Clostridium difficile PCR assay for an initial episode of CDI between January 2012 and January 2013 at a tertiary care hospital in Chicago were reviewed. Two patients diagnosed with CDI caused by a non-NAP1 strain (positive Xpert® C. difficile assay but negative Xpert® C. difficile Epi assay) were included for each patient diagnosed with CDI caused by a NAP1 strain (positive Epi assay). Patient charts were reviewed for markers of severity, risk factors, treatment regimens, and outcomes.
Results: Of 494 stool specimens, 90 (18%) that were positive for C. difficile by PCR were positive for NAP1 strain. In total, 37 patients with CDI due to NAP1 were matched with 74 patients with CDI due to non-NAP1 strains. Multivariable model revealed individuals ≥65 years old were 3 times more likely to have NAP1 strain than individuals <65 (P=.02). Residents of a nursing home prior to hospitalization were 10 times more likely to have NAP1 strain than patients residing in their homes (P=.001). More NAP1 cases had a change in treatment from metronidazole to oral vancomycin plus intravenous metronidazole (P=.01). The severity of CDI, incidence of mortality and recurrent CDI were similar between groups.
Conclusions: In a nonepidemic setting, NAP1 strains were more common in older patients and individuals admitted from nursing homes. Identification of NAP1 by PCR of stool specimens was associated in a change of therapy but did not predict worse outcomes. Reporting strain results may not be clinically useful in routine settings.