We would like to thank Drs Carifi and Carifi for their comments on our paper and wish to clarify their concerns about our study.

First, regarding surgical technique, the external and endoscopic dacryocystorhinostomy (DCR) were performed in a standardized fashion, as described in detail elsewhere. In external DCR, the rhinostomy extended from the fundus of the sac superiorly and the posterior lacrimal crest posteriorly. It went inferiorly, including the upper part of the medial nasolacrimal duct and the hamular process. The anteroposterior diameter of bony ostium was at least 12 mm. In endoscopic DCR, the rhinostomy was a little smaller, but patent rhinostomy sites with positive functional endoscopic dye test were noted at the last follow-up.

As for patient selection, we performed endoscopic DCR in patients with broad nasal space without septal deviation or middle turbinate hypertrophy, and without canalicular obstruction. In the other cases, we performed external DCR.

Although the revision DCR cases carry a high risk of failure and various factors might cause the rise of failure rate, there was no report regarding the clinical bacteriology of revision DCR. We found that Pseudomonas -isolated cases had histories of endoscopic revision associated with prolonged intubation and suggested that pseudomonal infection could be a factor affecting surgical outcome of revision DCR.

We employed topical quinolones and topical fluorometholone eyedrops postoperatively for 1 week, and nasal steroids were also used for 2 to 3 weeks in endonasal DCR. However, we believe that the use of topical antibiotics and steroid would not influence the bacteriologic results, which were assessed at least a few months after the surgery.

We are also aware of recent reports suggesting that the success rate without silicone tubes seems to be equivalent to that with tubes. On the other hand, some reports announced higher success rates with silicone intubation. Since it is not definitively determined whether to employ silicone tube, it is arguable to denounce our approach using silicone intubation. We are aware that the Lacrimal Intubation Trial, a multicenter, randomized, prospective interventional trial to determine the treatment effect of silicone intubation in external DCR for primary acquired nasolacrimal duct obstruction, is currently enrolling patients. It is only via a large clinical trial of this type that one will be able to definitively determine whether silicone intubation is beneficial.

Though there is a report suggesting that the routine use of systemic antibiotic prophylaxis in external DCR may not be justified, the report was dependent on a retrospective review of case notes and was not a randomized controlled study, indeed not confirmatory. Also, a large study would need to be conducted to determine whether the prophylactic use of antibiotics improves success rates. Our study does not indicate that we have achieved successful clinical outcomes through adequate antibiotic prophylaxis. The point of our study is that in cases of less successful surgery or of longer intubation periods, pseudomonal infection may occur, which may influence surgical outcomes; and that in these cases, early removal of the tube coupled with culture studies is recommended, followed by proper antibiotics application in cases where cultures reveal pseudomonal infections.