Sir,

We thank Yeun et al for taking the time to read our article1 in detail and for their valid comments.

Firstly, the authors request clarification on the surgeon's actual experience with regard to the three procedures in order to help interpret the final results. We agree that the learning curve may influence surgical time and success. At the time of commencing the study, the operating surgeon (JMO) had gained sufficient training, both in supervision and independently in order to perform endoscopic surgical and laser dacryocystorhinostomy (DCR) competently and safely. Since completing this study, endosurgical times have decreased marginally with increased experience. We have in fact abandoned endolaser (holmium) because we were disappointed with the poor results in comparison to endosurgical DCR.2

Secondly, Yeun et al ask for information on the possible reasons for the time difference between endosurgical and endolaser DCR with regard to preparation and passing tubes. In the study, the differences between these two small groups, both for preparation time and for passing tubes, were statistically insignificant. The apparent difference in preparation time may have reflected a choice of practice at the time, whereby aqueous povidine iodine was not applied to the skin for endolaser cases. Since the study, we have changed our practice and no longer use povidine iodine for endosurgical cases either.

Thirdly, the case mix for the endosurgical and endolaser groups was similar. No patient had a narrow nasal space requiring septoplasty. In addition, the tubes were knotted within the nose and the position of the knot in relation to the ostium was checked endoscopically at the end of surgery to ensure that they were not too loose and at risk of prolapse.3

Lastly, we agree with Yeun et al that granulation tissue formation may affect surgical success in DCR. During the follow-up period, all patients received a postoperative endoscopic endonasal examination by the senior author (JMO) 1 week after surgery, and then at the time of removal of tubes, which was usually 8 weeks postoperative. Although data were not prospectively recorded with regard to debris and granulation tissue removal, each group had a similar postoperative regimen. We do not give additional visits to endonasal DCR patients for debris removal.

We would once again like to thank Yeun et al for reading our paper so thoroughly and for raising valid questions and agree that they help promote discussion and ultimately better understanding of the issues surrounding the various approaches to DCR.