Background: The medical treatment for chronic anal fissure is modeled to eliminate sphincter spasm and hypertonia so that blood flow to the anal sphincter improves and the fissure heals. Efficacious pharmacological therapy avoids complications related to surgical treatment including incontinence and the risks of surgery and anesthesia. Conservative treatment of chronic anal fissure has been reported with wide variation in the healing rates. However there are few reports of conservatively treated chronic anal fissures fromAfrica, with only one paper identified in this study. Furthermore, the influence of clinical characteristics of fissures on the healing rates and recurrences have not been widely reported. Objective: To determine the healing rates of conservatively treated chronic anal fissure and to document the influence of fissure characteristics on healing and recurrence. Setting: The Surgical Outpatient Clinic at the Aga Khan University Hospital, Nairobi, Kenya. Patients and methods:Fifty (50) consecutive patients satisfying the criteria for chronic anal fissure were treated with local application of 0.2% glyceryl trinitrate (GTN) paste twice a day for 8 weeks and followed up for six months. Features of chronicity were defined and each given a score of 1. Patients with a score of 2 or less were compared to those with a score of 3 and above in terms of healing and recurrence. Results: Thirty nine patients of the 50 analyzed (78%) healed after the 8 weeks of treatment. At six months follow up 10 patients (26%) had relapsed. This translated to an overall success rate of 58%. Three factors significantly affected healing of a chronic anal fissure- presence of indurated fissure edges (p=0.001, OR 12.19, 95 % CI 2.57-57.94), presence of symptoms for more than three months (p=0.014, OR 8.03, 95 % CI 1.52-42.52) and a score of 3 and above (p=0.004, OR 11.45, 95 % CI 2.13-61.67). There was no feature of chronicity that predicted recurrence of chronic anal fissure after initial healing. Conclusion: GTN paste led to healing in the majority of patients. Features of chronicity may be helpful in predicting response to treatment. In the present study patients with a chronicity score of 2 and below respond better than those with a score of 3 and above.