Analyse of patient characteristics and aetiological causes of enterocutaneous fistulas and their impacts on in-hospital mortality: a ten-year retrospective cohort study.

Introduction: This study aimed to compare patient characteristics according to the primary aetiology including gunshot wounds in inpatient individuals diagnosed with enterocutaneous fistula (ECF) or enteroatmospheric fistula (EAF) and to evaluate the impacts of these characteristics on all-cause in-hospital mortality.

Methods: This is a single-centre hospital-based retrospective cohort study conducted with adult patients who were hospitalised for treatment of ECF or EAF. The patients were allocated to three study groups according to their primary aetiology (surgery-related group, gunshot-related group and other-cause group). The demographics and clinical features of the patients were compared between the study groups, furthermore, the impacts of these characteristics on in-hospital mortality were evaluated using Cox Regression Analysis.

Results: Gunshot-related fistulas were more likely to originate from small intestines, whereas surgery-related fistulas originated from all anatomic sites and those related to other aetiologies often originated from large intestines (p = 0.006). Severe malnutrition was more prevalent in the other-cause group (26.1%) (p < 0.001). Sepsis occurred in 24.3%, 68.8% and 47.8% of the patients (p = 0.008); and the median length of stay was 18.0, 45.0 and 32.0 days (p = 0.025) in the surgery-related group, the gunshot-related group, and the other-cause group, respectively. While the surgery-related and gunshot-related groups had similar and low mortality rates (2.7% and 6.7%, respectively), patients with other reasons had the highest mortality (30.4%). The increase in the duration of output (DOO) was associated with decreased mortality [HR (95%CI): 0.55 (0.39-0.79) p = 0.001], whereas being severely malnourished and having an aetiology of other causes were associated with increased mortality [HR (95%CI): 25.29 (5.20-123.09) and p < 0.001, and HR (95%CI): 9.06 (1.11-73.86) and p = 0.040, respectively].

Conclusions: Patient characteristics, clinical manifestations and treatment approach may differ according to primary aetiology in patients with ECF or EAF. Primary aetiology, the decrease in DOO and severe malnourishment have negative impacts on in-hospital mortality.