Dr. Reuther summarizes the results of our investigation in his introductory remarks. The possibilities offered by setting up a control group for studies like ours, such as a further strengthening of validity, are clear. However, we know of virtually no (adult) patients that are willing to accept years of work absences and pain as part of randomization.
Dr. Reuther rightly points out that, according to current data, a wait-and-see approach is well worth considering in children. Firstly, the controversy over this wait-and-see strategy does not exist for adult patients, and secondly, children were not the subject of our investigation. Thus, to include the available studies on this latter patient population would have been confusing. We therefore see no conflict with the studies cited by Dr. Reuther. The possibility that the drop-out-rate might have created a bias was the first point made in the “Limitations” section.
We are very grateful to Prof. Guntinas-Lichius for his suggestions. It must be said with regard to the responder comparison that, in the course of preparing the article, we replaced the missing data from non-responders with a sensitivity analysis using multiple imputations (statistical methods to complete the data lacking due to non-response). The analyses proved to be stable under a variety of conditions. We saw only age differences with no visible effect on the results.
Due to the unavailability of drop-out patients, any subsequent investigation of serum would be associated with a bias and hence unhelpful. Moreover, this would require submitting new ethical-approval and clinical-study applications. The authors of future studies on this subject will undoubtedly be inclined to adopt this suggestion from the outset of their deliberations and will be grateful, as are we, for these comments.
We fully endorse the important idea of investigating predictors of treatment outcome. Such an approach could reduce the number of patients that do not benefit from treatment. Our investigation, however, does not contribute to an expansion of knowledge in this area.
Dr. med. Götz Senska
Klinik für Hals-Nasen-Ohren-Heilkunde, Plastische Operationen
Marienhospital Gelsenkirchen GmbH
Conflict of interest statement
The authors of all contributions declare that no conflict of interest exists.
|1.||Senska G, Atay H, Pütter C, Dost P: Long-term results from tonsillectomy in adults. Dtsch Arztebl Int 2015; 112: 849–55 VOLLTEXT|