Sirs, Madalinski is not convinced by the results of our placebo-controlled, randomized, double-blind study which evaluated the efficacy and safety of botulinum A toxin in the treatment of chronic anal fissure.1 We are afraid we cannot do better. He tries to explain the lack of difference between the two groups1 by an unusually high rate of healing of chronic anal fissure or symptomatic improvement in the placebo group,2 a possible imbalance between hypothetical treatments associated with the tested drug,3 an inadequate injection site and/or4 an insufficient dose of toxin. We invite him to read carefully the discussion section of our paper, in which most of these hypotheses were already discussed and rejected. Chronic anal fissure may heal when patients use laxatives and ointments. In our study the fissure was healed in 23% and 32% of the patients (these percentages being exactly the same in the two groups of randomization) after 4 and 8 weeks of follow-up, respectively.1 These results are in total agreement with a recent meta-analysis of non-surgical therapy for anal fissure in which the healing rate under placebo, which was fairly uniform across studies, was found to be 35%.2 Careful control of associated drugs was an important option within our study in order to avoid biases in the evaluation of pain and healing.1Nitrates and drugs that modify anal tone could not be used once the patients were included. In contrast, analgesic tablets, laxatives and ointments were given in both groups. Thus, drug consumption was both carefully controlled and quantified, and there was no imbalance between the treatments that was associated with the tested drug. It is true that an inadequate injection site and/or inadequate dosage could, in theory, explain the lack of efficacy of botulinum toxin. However, Madalinski's remark is quite surprising, since he claimed in another letter that botulinum toxin has a remarkably high diffusion around the injection point and therefore that the injection site does not matter.3 Moreover, the internal anal sphincter is often visible (fissure depth) and palpable (hypertonia) in a patient suffering from anal fissure. Considering the dosage, bioequivalent doses between Dysport and Botox botulinum toxins were clearly established before the beginning of our study4 and, as recognized by Madalinski himself, the dose of Dysport used in our study was equal to the dose of Botox applied by Maria et al.5 Thus, it is unlikely that an inadequate injection site and/or an insufficient dose of toxin could explain our results. Remissions and relapses are the main characteristics of the natural history of chronic anal fissure. These features and the healing rate under placebo are of paramount importance when considering conflicting results.6 In sensitivity analyses that excluded studies with placebo success rates below 10% (two standard deviations below the mean), the statistical evidence for the efficacy of nitrates disappeared.2 In another recent meta-analysis of a series of independent comparative clinical trials performed world-wide, the efficacy of topical nitrates to heal anal fissure could not be demonstrated.7 In particular, all multicentric trials failed to show a nitroglycerin-mediated therapeutic effect in comparison with placebo (42% vs. 45%, respectively, in 888 patients).7 Finally, in Nelson's meta-analysis, comparisons of nitroglycerin ointment with botulinum toxin injection or calcium channel blockers showed no significant difference in efficacy between the three drugs.2 Thus, knowing the natural history of anal fissure, the healing rate observed under placebo and the lack of a demonstrative effect of different types of drugs in meta-analyses, we could speculate a poor (if any) benefit from chemical sphincterotomy in the treatment of chronic anal fissure.