THE LANCET Summary Background Anal fissure is most commonly treated surgically by internal anal sphincterotomy. However, there is some concern over the effects of this procedure on continence. Nitric oxide donors such as glyceryl trinitrate (GTN) have been shown to cause a reversible chemical sphincterotomy capable of healing fissures in a small series of cases. This study reports a prospective, randomised, double-blind, placebo-controlled trial to test the hypothesis that topical GTN is the best first-line treatment for chronic anal fissure. Methods 80 consecutive patients were randomised to receive treatments with topical 0·2% GTN ointment or placebo. Maximum anal resting pressure (MARP) was measured with a constantly perfused side-hole catheter before and after the first application of trial ointment. Anodermal blood flow was measured during manometry by laser Doppler flowmetry. After initial treatments, patients were given a supply of ointment (either GTN or placebo) to be applied to the lower anal canal twice daily. Patients were reviewed 2-weekly. At the initial and follow up visits patients were asked to record pain experienced on defaecation on a linear analogue pain score. Endpoints were healing of the fissure or condition after 8 weeks of treatment. Findings After 8 weeks, healing was observed in 26/ 38 (68%) patients treated with GTN and in 3/ 39 (8%) patients treated with placebo (p<0·0001,  2 test). Linear analogue pain score fell significantly in both groups after 2 weeks of treatment. This fall was maintained in those treated with GTN but pain scores returned to pre-treatment values by 4 weeks on treatment with placebo. MARP fell significantly from a mean of 115·9 (SD 31·6) to 75·9 (30·1) cm H 2 O (p<0·001, Student’s paired t- test) in patients treated with GTN but no change was seen in MARP after placebo. Anodermal blood flow measured by laser Doppler flowmetry significantly increased after application of GTN ointment but was unaffected by placebo. Interpretation Topical GTN provides rapid, sustained relief of pain in patients with anal fissure. Over two-thirds of patients treated in this way avoided surgery which would otherwise have been required for healing. Long-term follow up is needed to assess the risk of recurrent fissure in patients with GTN. Lancet 1997; 349: 11–14 Introduction Anal fissure is characterised by pain on defaecation, rectal bleeding, and spasm of the internal anal sphincter (IAS). The aetiology of anal fissure is contentious; it may be due to ischaemia of the posterior commissure of the anal canal, exacerbated by hypertonicity of the internal anal sphincter. 1,2 Although conservative treatment with topical ointments is successful for most acute anal fissures, they do not usually help in the treatment of chronic anal fissures. 3–5 Surgical treatments for fissure overcome spasm of the IAS by forcible anal dilatation or internal sphincterotomy. Both anal dilatation and sphincterotomy are associated with short-term and long-term impairment of continence in up to 30% of patients. 6–8 A non-surgical method of reducing anal pressure to treat anal fissure would be useful. 9 Nitric oxide (NO) has recently been shown to be an inhibitory neurotransmitter in the IAS. 10 Organic nitrates are degraded by cellular metabolism releasing NO. 11 Glyceryl trinitrate (GTN) ointment applied to the anus causes a fall in maximum anal resting pressure (MARP) amounting to a reversible “chemical sphincterotomy”. 12 Anodermal blood flow may be inversely related to MARP because the blood supply to the mucosa comes predominantly from vessels which cross the sphincter. Increase in anodermal blood flow has been reported after lateral internal sphincterotomy and topical applications of nitrates. 13,14 Small, uncontrolled studies have suggested that GTN ointment may be an effective treatment for chronic anal fissure but no prospective randomised trial has been done. 14–17 We evaluated topical GTN ointment in the treatment of chronic fissure in a randomised, double blind, placebo-controlled trial. Methods Consecutive patients who attended sugical outpatients clinics at two hospitals with chronic anal fissure were recruited. All patients had had symptoms of anal fissure of more than 6 weeks with fibrosis at the base of the fissure and were therefore deemed to be chronic. Patients were randomly allocated to receive either 0·2% GTN ointment (Percutol, Dominion Pharma, Haslemere, UK) diluted 1 in 5 with white soft paraffin or placebo (white soft paraffin) in a double-blind study. Randomisation was done in the hospital pharmacy with computer-generated randomisation. Trial ointments were prepared in identical numbered pots by staff not involved further in the trial. Both trial ointments looked the same. The randomisation code remained in the hospital pharmacy during the trial and was not available to the investigators. 80 patients were recruited to give a 90% chance of finding a 40% difference in healing between placebo and GTN at the 5% level. The code was broken when 80 patients had completed the trial. Study protocol was approved by the Ethics Committee of the University Hospital, Nottingham and informed consent to enter the trial was obtained from each patient. On the first visit, anal manometry was done with a constantly perfused side-hole catheter. When the catheter had been in position for 20 min and a steady state had been reached, 0·5 g of trial ointment or placebo was applied to the lower anal canal. Manometry was continued for a futher 40 min. Maximum anal resting pressure (MARP) was recorded before and after application of the ointment. Anodermal blood flow was measured Vol 349 • January 4, 1997 11 Department of Surgery, University Hospital, Nottingham, NG7 2UH, UK (J N Lund FRCS, J H Scholefield ChM) Correspondence to: Mr J H Scholefield, Department of Surgery, E Floor, West Block, University Hospital, Nottingham NG7 2UH, UK A randomised, prospective, double-blind, placebo-controlled trial of glyceryl trinitrate ointment in treatment of anal fissure Jonathan N Lund, John H Sc holefield