RESUMO
BACKGROUND: Clostridioides difficile infection results in life-threatening short-term outcomes and the potential for subsequent recurrent infection. Predicting these outcomes at diagnosis, when important clinical decisions need to be made, has proven to be a difficult task. METHODS: 52 clinical features from existing models or the literature were collected retrospectively within ±48 h of diagnosis among 1660 inpatient infections. A modified desirability of outcome ranking (DOOR) was designed to encompass clinically-important severe events attributable to the acute infection (intensive care transfer due to sepsis, shock, colectomy/ileostomy, mortality) and/or 60-day recurrence. A deep neural network was constructed and interpreted using SHapley Additive exPlanations (SHAP). High-importance features were used to train a reduced, shallow network and performance was compared to existing conventional models (7 severity, 7 recurrence; after summing DOOR probabilities to align with conventional binary outputs) using area under the ROC curve (AUROC) and DeLong tests. FINDINGS: The full (52-feature) model achieved an out-of-sample AUROC 0.823 for severity and 0.678 for recurrence. SHAP identified 13 unique, highly-important features (age, hypotension, initial treatment, onset, PCR cycle threshold, number of prior episodes, antibiotic exposure, fever, hypotension, pressors, leukocytosis, creatinine, lactate) that were used to train a reduced model, which performed similarly to the full model (severity AUROC difference P = 0.130; recurrence P = 0.426) and significantly better than the top severity model (reduced model predicting severity 0.837, ATLAS 0.749; P = 0.001). The reduced model also outperformed the top recurrence model, but this was not statistically-significant (reduced model recurrence AUROC 0.653, IDSA Recurrence Risk Criteria 0.595; P = 0.196). The final, reduced model was deployed as a web application with real-time SHAP explanations. INTERPRETATION: Our final model outperformed existing severity and recurrence models; however, it requires external validation. A DOOR output allows specific clinical questions to be asked with explainable predictions that can be feasibly implemented with limited computing resources. FUNDING: National Institutes of Health-Institute of Allergy and Infectious Diseases.
RESUMO
OBJECTIVE: We sought to validate available tools for predicting recurrent C. difficile infection (CDI) including recurrence risk scores (by Larrainzar-Coghen, Reveles, D'Agostino, Cobo, and Eyre et al) alongside consensus guidelines risk criteria, the leading severity score (ATLAS), and PCR cycle threshold (as marker of fecal organism burden) using electronic medical records. DESIGN: Retrospective cohort study validating previously described tools. SETTING: Tertiary care academic hospital. PATIENTS: Hospitalized adult patients with CDI at University of Virginia Medical Center. METHODS: Risk scores were calculated within ±48 hours of index CDI diagnosis using a large retrospective cohort of 1,519 inpatient infections spanning 7 years and compared using area under the receiver operating characteristic curve (AUROC) and the DeLong test. Recurrent CDI events (defined as a repeat positive test or symptom relapse within 60 days requiring retreatment) were confirmed by clinician chart review. RESULTS: Reveles et al tool achieved the highest AUROC of 0.523 (and 0.537 among a subcohort of 1,230 patients with their first occurrence of CDI), which was not substantially better than other tools including the current IDSA/SHEA C. difficile guidelines or PCR cycle threshold (AUROC: 0.564), regardless of prior infection history. CONCLUSIONS: All tools performed poorly for predicting recurrent C. difficile infection (AUROC range: 0.488-0.564), especially among patients with a prior history of infection (AUROC range: 0.436-0.591). Future studies may benefit from considering novel biomarkers and/or higher-dimensional models that could augment or replace existing tools that underperform.
