Acute Clostridioides difficile Infection in Hospitalized Persons Aged 75 and Older: 30-Day Prognosis and Risk Factors for Mortality

Objectives: To assess the 30-day mortality predictive markers in the oldest patients with Clostridioides difficile infection (CDI) and to analyze the accuracy of the European severity risk markers in this population.

Design: Observational prospective multicenter cohort study conducted by the French Infectious Diseases Society and Geriatrics Society networks. An electronic questionnaire was sent to members of both societies regarding their participation. Each investigator used an online survey to gather the data.

Setting and participants: Patients aged ≥75 years hospitalized in French geriatric or infectious wards with confirmed diagnosis of CDI between March 1, 2016 and May 1, 2017.

Methods: Clinical and laboratory parameters included medical history and comorbidities with the Cumulative Illness Rating Scale (CIRS). Criteria increasing the risk of severe disease were recorded as listed in the European guidelines. Therapeutic management, recurrence, and mortality rates were assessed at day 30 after diagnosis.

Results: Included patients numbered 247; mean age was 87.2 years (SD 5.4). Most of the CDI incidences (66.4%) were health care-associated infections, with 81% diagnosed within 30 days of hospitalization; CIRS mean score was 16.6 (SD 6.6). Markers of severity ≥3 included 97 patients (39.3%). Metronidazole was the main initial treatment (51.0%). C difficile infection in the older adult was associated with a 30-day mortality of 12.6%. Multivariate analysis showed that baseline CIRS score [hazard ratio (HR) 1.06 per 1-point increase, 95% confidence interval (CI) 1.00-1.12] and evidence of cardiac, respiratory, or renal decompensation (HR 3.04, 95% CI 1.40-6.59) were significantly associated with mortality.

Conclusions and implications: European severity markers are adequate in the oldest old. Organ failure and comorbidities appeared to be the main markers of prognosis, and these should raise the awareness of practitioners. Although antibiotic treatment was not predictive of mortality, our results point out the lack of adherence to current guidelines in this population.