Modification of Parks Classification of Cryptoglandular Anal Fistula

BACKGROUND: The Parks classification has been used for the classification of anal fistula for several years, but it does not allow for risk factors for failure after surgery. OBJECTIVE: This study aimed to describe a modification of the Parks classification of anal fistula and examine its predictive validity in the assessment of the outcome of anal fistula in terms of failure of healing and fecal incontinence. DESIGN: This is a retrospective review of a prospective database. SETTING: This study was conducted in the Colorectal Surgery Unit, Mansoura University Hospitals. PATIENTS: Adult patients with anal fistula who underwent surgery were included. INTERVENTIONS: Five risk factors for failure after fistula surgery were identified from the literature and were examined by multivariate analysis of our patients. Four risk factors proved to be significant independent predictors of failure: secondary extensions, horseshoe fistula, previous fistula surgery, and anterior anal fistula in women. We modified the Parks classification by dividing the transsphincteric type into high and low and by grouping supra- and extrasphincteric anal fistulas into 1 group. The first 3 stages were subdivided according to the absence or presence of predictors of failure. MAIN OUTCOME MEASURES: The primary outcome measured was the validity of the modified Parks classification with regard to the rates of failure and fecal incontinence after surgical treatment of each stage of anal fistula. RESULTS: A total of 665 patients with cryptoglandular anal fistula were included. Failure rates increased from 2.3% (95% CI, 0.9%-4.7%), to 17.4% (95% CI, 10.8%-25.9%), 19.5% (95% CI, 15%-24.6%), and 30.7% (95% CI, 9.1%-61.4%) across the 4 stages. The area under the receiver operating characteristic curve was 0.90 (95% CI, 0.85-0.94) indicating the strong discriminative ability of the final multivariable predictive model. The increase in failure and incontinence rates across the fistula stages was significant. LIMITATIONS: This is a retrospective, single-center study. CONCLUSION: Inclusion of predictors of poor outcome into the modified classification helped differentiate simple and complex fistulas within each stage and between the different stages, which can help in assessment and decision making for anal fistula. See Video Abstract at http://links.lww.com/DCR/B441.