Is topical nifedipine effective for chronic anal fissures?

Journal of Family Practice, March, 2003 by Dan Merenstein, Dan Rosenbaum

Perrotti P, Bove A, Antropoli C, et al. Topical nifedipine with lidocaine ointment vs. active control for treatment of chronic anal fissure: results of a prospective, randomized, double-blind study. Dis Colon Rectum 2002; 45:1468-1475.

* PRACTICE RECOMMENDATIONS

Patients in this study showed remarkable improvement when 1.5% lidocaine and 0.3% nifedipine were applied twice daily for 6 weeks. This extremely safe, well tolerated, and effective treatment should provide family physicians with a reliable nonsurgical method for treating chronic anal fissures.

* BACKGROUND

Acute anal fissures generally heal spontaneously with minimal or no intervention. Conversely, chronic anal fissures are traditionally treated with surgery. Therapies such as botulinum toxin, isosorbide dinitrate, and glyceryl tinitrate have shown some benefit, but their side-effect profiles are substantial. With the knowledge that topical nifedipine has been shown to relax smooth muscle, lower anal resting pressure, relieve pain, and heal acute anal fissures, these authors studied its effect on chronic anal fissures.

* POPULATION STUDIED

Patients were recruited from the emergency surgery and gastroenterology center in Italy that conducted the study. Inclusion criteria were chronic anal fissure and age older than 18 years. Chronic anal fissure was assessed by clinical examination and a history of anal pain on defecation for longer than 2 months that did not resolve with stool softeners and simple anesthetic agents. Exclusion criteria were pregnancy, allergy to nifedipine or lidocaine, and complications warranting surgery.

* STUDY DESIGN AND VALIDITY

This was a prospective, randomized, double-blind study. The control group received 1.5% lidocaine and 1% hydrocortisone acetate, and the treatment group received 1.5% lidocaine and 0.3% nifedipine. The ointments were applied every 12 hours for 6 weeks. The ointments were indistinguishable, and all parties were blinded with proper allocation concealment. Data analysis was by intention-to-treat. The groups were randomly assigned and had similar baseline characteristics.

The ethics committee required the control ointment to have hydrocortisone with lidocaine rather than a true placebo. It is possible that hydrocortisone was detrimental to healing and thus made the nifedipine ointment appear even better than it would have if compared with a true placebo. Also, the authors did not explain why clinical examination was done at 42 days, but manometric examination and clinical pain score was performed at 21 days. Overall this study was methodical and well-executed. Further, no patients were lost to follow-up.

* OUTCOMES MEASURED

Healing of the chronic anal fissure was the primary outcome and was defined by anoscopy when epithelialization or formation of a scar was achieved at 42 days. Patients also subjectively rated pain as absent, modest, or persistent at day 21. Manometric studies were used as a secondary measure of clinical improvement and were measured at baseline and 21 days.

* RESULTS

Of the 55 patients in the nifedipine group, 94.5% healed clinically at 42 days and 87.3% reported no pain at 21 days. Conversely, of the 55 patients in the control group evaluated at the same intervals, 16.4% healed and 10.9% reported no pain (P<.001; number needed to treat=1.3).

In the healed nifedipine group, 3 patients experienced a recurrence of their ulcers, and 2 were treated successfully with a second round of nifedipine. In the control group, 5 of the 9 who initially healed experienced a recurrence of the ulcer and also were treated successfully with nifedipine.

Mean anal resting pressure decreased by 11.0% in the nifedipine group but increased by 4.4% in the control group. After removal of blinding, 46 patients who were not healed in the control group were offered the nifedipine treatment. Of these, 38 healed with nifedipine ointment. No patients in the nifedipine group had any systemic side effects, whereas 1 patient treated with nifedipine and 3 in the control group had slight local hyperemia, which improved when treatment was completed.

Dan Merenstein, MD, and Dan Rosenbaum, MD, Department of Family Medicine, Georgetown University, Arlington, VA. E-mail: dmerenstein@yahoo.com.