Background: The medical management of chronic anal fissure varies among clinicians because no single treatment has yet approached an overall healing rate similar to lateral sphincterotomy. However, chemical sphincterotomy has the advantage of not carrying the presumed high risk of faecal incontinence. Glyceryl Trinitrate is commonly used in our practice with varied response. This study set out to evaluate the effectiveness of a locally compounded preparation of 0.2% topical Nifedipine in the treatment of chronic anal fissure and to compare the investigational product to 0.2% Glyceryl Trinitrate in overall healing rate and relieving pain. Design: Single centre, parallel, double-blind randomized controlled trial. Setting: The surgical outpatient department at Aga Khan University Hospital (Nairobi). Participants: Patients with chronic anal fissure were recruited from the surgical outpatient department. Intervention: A computer generated randomization sequence was used to allocate treatment. Patients were randomly allocated to receive Glyceryl Trinitrate or Nifedipine, both applied 12 hourly topically to the perianal region for a period of eight weeks. Main Outcome measure: The primary outcome measure was overall healing rate of chronic anal fissure after eight weeks of treatment. Secondary outcome measures were cumulative healing rate, time to resolution of pain and occurrence of headache. Results: There was no significant difference recorded between the test and control drugs with regard to age, gender, duration of symptoms, fissure position or entry VAS pain score. Seventy five patients (Glyceryl Trinitrate n=38; Nifedipine n=37) were randomized. The mean age was 36.3 years (SD 11.9) with a mean duration of symptoms of 37.4 weeks (SD 36). Males comprised 48.6 percent of the patients. Seven patients were excluded from analysis because of loss to follow up or missing data, leaving 68 (n=33; n=35) patients for intention to treat analysis. Healing rate was higher (p=0.001) with Nifedipine (94.3 percent) as compared with Glyceryl Trinitrate (60.6 percent). The mean time to healing was shorter (p<0.001) in patients treated with Nifedipine (6.5 ± 1.2 weeks) compared with Glyceryl Trinitrate (9.1 ± 1.9 weeks). The mean pain scores after two and four weeks of treatment differed (p<0.001) with quicker resolution in the Nifedipine group. On completion of treatment at eight weeks, pain scores were significantly lower in both groups (0.25 in Glyceryl Trinitrate and 0 in Nifedipine versus 7.0 and 7.8, respectively). Headaches were more frequent (p<0.001) with Glyceryl Trinitrate (69.7 percent) as compared with Nifedipine. Conclusion: Topical application of 0.2% Nifedipine was more effective than