The aim of the study is to evaluate demographics, epidemiology, clinical characteristics, treatment and outcomes of Clostridium difficile infection (CDI) in patients with and without concurrent cancer. This is a prospective cohort study of consecutive primary CDI episodes in adults (January 2006-December 2016). CDI was diagnosed on the presence of diarrhoea and positive stool testing for toxigenic C. difficile. Univariate analysis assessed differences between cancer and non-cancer patients. Risk factors of all-cause 30-day mortality were determinate using the logistic multivariable procedure. In total, 787 CDI episodes were recorded, 191 in cancer patients (median age 64, IQR 50-73). Of these, 120 (63%) had solid and 71 (37%) haematological malignancies (24 received a stem cell transplant). At the CDI diagnosis, 158 (82.7%) cancer patients had prior antibiotics and 150 (78.5%) were receiving proton pump inhibitors. Fifty-seven (80.3%) patients with haematological and 52 (43.3%) with solid malignancies were under chemotherapy at diagnosis; 25 (35.2%) with haematological and 11 (9.2%) with solid malignancies had an absolute neutrophil count < 1000/mm3. Overall, 30-day mortality was higher in cancer patients than in those without (19.2 vs. 8.6% respectively, p < 0.001); recurrence rates did not vary significantly (11.1 vs. 11%, p = 0.936). By type of neoplasm, 30-day mortality was higher in patients with haematological malignancies and solid tumours than in patients without cancer (respectively, 25.4 vs. 8.6%; p < 0.001 and 15 vs. 8.6%; p < 0.001). Our results suggest that the prognosis of CDI (30-day mortality) is poorer in patients with cancer than in those without although percentages of recurrent infection are similar in these two patient populations.
Keywords: Clostridium difficile infection; Epidemiology; Haematology patients; Oncology patients; Outcome.