PCR Cycle-Threshold-Derived Toxin Identifies Patients at Low-Risk for Complications of C. difficile Infection Who Do Not Require Treatment

Background Literature suggests that toxin detection differentiates those who require treatment for C. difficile infection (CDI) from those who do not. In-house studies have shown that free toxin can be predicted with high negative predictive value at a predefined cycle threshold (CT) using Xpert tcdB PCR (Cepheid, Sunnyvale, CA). In October 2016, CT-toxin was added to the PCR result and a comment recommends against CDI therapy if CT-toxin is negative (CTtox-). Here we evaluate the effect of this reporting on treatment rates and outcomes of CTtox- patients. Methods Patients tested from October 2016 to Apr. 2017 with a positive Xpert PCR and CTtox- result were included. Clinical data were collected by retrospective chart review and analyzed with the Chi squared and Student t-tests using SPSS. Due to multiple comparisons, α=0.01. Results Of 1516 Xpert PCR tests, 248 (16.4%) were positive and 98 (39.5%) were CTtox-. Of these, 54 (55.7%) were treated. Patient characteristics and data at testing are shown below. There were no cases of CDI-related septic shock or toxic megacolon on review. Time to diarrhea resolution was significantly shorter in untreated patients and there was no difference in crude mortality or later onset of CTtox+ CDI. Treated n = 54 (%) [SD] Untreated n = 43 (%) [SD] p Male 25 (46) 24 (56) 0.35 Age (Yrs) 57.9 [19.7] 52.0 [20.8] 0.15 Immunocompromised 36 (67) 21 (49) 0.08 BMT/SOT 18 (33) 5 (12) 0.02 Cancer 16 (30) 9 (21) 0.33 Other 6 (11) 10 (23) 0.11 WBC (K/μL) 9.9 [7.7] 7.9 [5.3] 0.20 Creatinine (mg/dL) 1.1 [0.9] 1.0 [0.7] 0.34 Initial Stools/Day 6.2 [2.8] 4.9 [2.8] 0.05 Days to <3 Stools 2.6 [1.8] 1.6 [0.8] <0.01 Hospitalized in past mo. 31 (57) 33 (77) 0.05 Antibiotics in past mo. 35 (65) 29 (67) 0.79 CDI in past 6 mos. 7 (13) 7 (16) 0.64 In-Patient 23 (42) 30 (70) 0.02 CDI Within 90 Days 3 (6) 4 (9) 0.70 Crude Mortality (30d) 6 (11) 7 (16) 0.46 Patient data and outcomes in those with PCR+/CTtox- C. difficile test. Immunocompromised (Other) includes patients with cirrhosis, IBD, and rheumatological diseases on immunosuppressants. Conclusion This study demonstrates the impact of stand-alone PCR assay with toxin prediction on reducing CDI therapy rates and provides further evidence that PCR+/toxin- patients are at low risk for CDI-related complications and do not require treatment, though more data is needed in transplant populations. Disclosures N. Banaei, Cepheid: Collaborator, Research Contractor and Scientific Advisor, honorarium for advisory role and Research support