Reply to the comment of R. Härtl:

Arts MP et al. Tubular Discectomy vs. Conventional Microdiscectomy ... Minim Invas Neurosurg 2010; 53: 96 The fact that tubular discectomy was not superior to conventional surgery was not due to inexperience of surgeons, as H ä rtl and colleagues suggest. Surgeons needed to perform at least 15 procedures before they could participate in our trial [1] . Moreover, we found a tendency of worse results in more experienced surgeons. We strongly disagree with the suggestion to exclude the patients with recurrent disk herniation from primary analysis. First, in an intention-to-treat analysis, by defi nition, the primary endpoint should include all patients with all possible reasons for an unfavourable outcome. Moreover, recurrent disk herniation might very well be an important reason for diff erence in pain scores between the groups. To decide on new guidelines, the fi nal clinical result of both groups, including all its determinants, is crucial. In our opinion, excluding these patients would certainly bias the primary outcome. We agree with the statement that ‘ the lack of benefi t from tubular discectomy over conventional surgery does not mean that tubular surgery would not have a signifi cant advantage when comparing potentially much more invasive proceReply to the comment of R. H ä rtl: