Feed

Background Anorectal abscess and fistula-in-ano represent acute and chronic manifestations of cryptoglandular anorectal sepsis. Fistula formation after incision and drainage remains an important cause of morbidity and recurrent sepsis. Evidence regarding demographic, socioeconomic, dietary, hydration, bowel habit, and clinical factors associated with fistula formation is limited, particularly in the Indian population. This study assessed the six-month incidence of fistula-in-ano following abscess drainage and factors associated with its development. Methods A prospective observational study was conducted among 65 adults undergoing incision and drainage of anorectal abscess at JSS Medical College and Hospital, Karnataka, India. Patients were followed for six months. Demographic, socioeconomic, comorbidity, dietary, lifestyle, bowel habit, abscess, and microbiological variables were recorded. Associations with fistula formation were assessed using Pearson chi-square, Fisher's exact, or Fisher-Freeman-Halton exact tests, as appropriate. Significant variables were further assessed using univariate binary logistic regression. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. A two-sided p <0.05 was considered significant. Results Fistula-in-ano developed in 22/65 patients (33.8%); of the remaining, 39 (60.0%) achieved complete healing without fistula and four (6.2%) developed recurrent abscess without fistula. Male sex (OR 7.20, 95% CI 1.36-38.20; p=0.014), age 61-70 years (OR 6.60, 95% CI 1.23-35.44; p=0.028), abscess size >3 cm (OR 6.88, 95% CI 1.98-23.88; p=0.002), mixed diet (OR 2.99, 95% CI 1.03-8.66; p=0.043), constipation (OR 6.64, 95% CI 1.71-25.76; p=0.006), diabetes mellitus (OR 4.28, 95% CI 1.33-13.75; p=0.014), and water intake <1.5 L/day (OR 3.40, 95% CI 1.01-11.45; p=0.048) were associated with higher odds of fistula formation. Abscess type was significantly associated with fistula formation (p=0.007). After recoding, patients with ischiorectal, submucosal, or supralevator abscesses had higher odds than perianal abscesses (OR 19.20, 95% CI 2.06-179.08; p=0.009). Lower socioeconomic status was significantly associated with fistula formation on Fisher-Freeman-Halton exact testing (p=0.003) but was not included in logistic regression because no fistula events occurred in the upper socioeconomic group, resulting in complete separation and unstable parameter estimates. Abscess position was significant on categorical analysis (p=0.013), although posterior location was not significant on regression (p=0.113). Among the 22 patients who developed fistula-in-ano, 13 (59.1%) had posterior internal openings and 16 (72.7%) had intersphincteric fistulas. Conclusion One-third of patients developed fistula-in-ano within six months following drainage of anorectal abscess. Male sex, older age, larger abscesses, diabetes, constipation, mixed dietary pattern, low water intake, and non-perianal abscess type were associated with increased odds of fistula formation. These findings identify potentially high-risk clinical and modifiable characteristics that may help guide closer follow-up and patient counselling after abscess drainage. However, the associations should not be interpreted as independent predictors because multivariable analysis was not performed. Larger multicentre prospective studies with longer follow-up and multivariable modelling are required to validate these findings and establish independent risk factors.