Serial ΔCt (tcdB) dynamics as a correlate of disease severity but not an independent predictor of poor outcome in hospitalized patients with clostridioides difficile infection: a retrospective cohort study

Abstract Background Clostridioides difficile infection (CDI) is the leading cause of healthcare-associated diarrhoea worldwide. The quantitative PCR cycle threshold (Ct) value generated at diagnosis is routinely discarded, and no consensus exists for serial Ct monitoring during treatment. This study evaluated the association of ΔCt (change between initial and follow-up Ct of tcdB) with disease severity, and separately examined host factors associated with poor clinical outcome, in a real-world Chinese tertiary-care inpatient cohort. Methods Retrospective cohort study of 141 consecutive adults with laboratory-confirmed CDI at Ruian People’s Hospital (January 2020–December 2023). All patients received oral metronidazole (500 mg three times daily) or vancomycin (125 mg four times daily). A follow-up PCR was performed at 7–21 days after treatment initiation. Disease severity was categorised as non-severe (WBC ≤ 15,000 cells/µL and serum creatinine < 1.5 mg/dL) or severe (WBC > 15,000 cells/µL or serum creatinine ≥ 1.5 mg/dL) per IDSA/SHEA criteria. Poor outcome was the composite of in-hospital death or discharge against medical advice. Results Of 141 patients enrolled (median age 78 years [IQR 65–84]; 59.6% aged ≥ 75 years), 101 (71.6%) had non-severe and 40 (28.4%) had severe CDI. ΔCt (tcdB) was significantly lower in severe than non-severe CDI (+ 0.3 [IQR − 1.0–1.3] vs. + 5.3 [IQR 3.0–6.6]; Spearman r = − 0.53, p < 0.001), whereas initial Ct did not differ by severity ( p = 0.486). Fifty patients (35.5%) experienced poor clinical outcomes (in-hospital death or discharge against medical advice); ΔCt was not significantly associated with poor clinical outcomes ( p = 0.217). On multivariable logistic regression, only advanced age (≥ 75 years) (aOR 8.13, 95% CI 2.45–27.01) and ICU admission (aOR 23.21, 95% CI 7.84–68.66) were independently associated with poor clinical outcomes; non-CDI antibiotic polypharmacy (≥ 3 agents) and diabetes mellitus were significant only in univariate analysis. Exploratory ROC analysis demonstrated strong discrimination of disease severity by ΔCt (AUC 0.837, 95% CI 0.759–0.907) but limited discrimination for poor clinical outcomes (AUC 0.563, 95% CI 0.458–0.663). Conclusions Serial ΔCt monitoring at 7–21 days provides clinically relevant information on microbiological response and disease severity but does not independently predict clinical outcomes. Advanced age (≥ 75 years) and ICU admission, rather than ΔCt, were independent predictors of poor clinical outcomes. ΔCt may therefore complement, rather than replace, established clinical indicators for CDI risk assessment, and the clinical utility of ΔCt-based monitoring requires prospective validation using predefined outcome-based endpoints.