Outcomes of Endoscopic Dacryocystorhinostomy

Endoscopic DCR has been compared to external DCR in various studies. Advocates for the endonasal approach offer the benefits of addressing comorbid sinonasal disease and septal deviation as well as the avoidance of an external scar. A Cochrane review published in 2017 reports no difference in anatomic success between the two groups when mechanical endonasal DCR was compared to the external approach (90% both groups; relative risk 1.00; 95% confidence interval 0.81 to 1.23; 40 participants); however, the endonasal group did not achieve patency as often when the laser-assisted technique was used (63% vs. 91%; relative risk 0.69; 95% confidence interval 0.52 to 0.92; 64 participants).

Two recent retrospective case series report data from 624 and 120 cases of endoscopic, endonasal DCRs. Coumou et al. reported an anatomic success rate in 90.1% and a functional success rate in 90.1% of adults with the use of silicone stenting, removed 3 months after surgery. They had a 3.2% complication rate and average follow-up time between 3 and 21 months. Ciger et al. reported an anatomic success rate of 92.5% and functional success rate of 92.5% without the use of silicone stenting with mean follow-up of 46.5 months. Similarly, they had a 3.3% complication rate. Both of these studies discuss postoperative scarring (fibrosis, granulation, and synechiae) at the rhinostomy site as the primary cause for failure. Further, Ciger et al. reported that the scarring was identified in the 3- to 4-month postoperative time frame.

Outcome data for patients who undergo revision DCR for either failed external or endoscopic DCR are also available. Ali, Psaltis, and Wormald report 91.3% anatomic and 86.9% functional success rates after revision endoscopic DCR with a mean follow-up of 26.4 months. Moreover, they report the causes of primary failure, including cicatricial ostium closure (55%), scarred internal common opening (27%), and organized granuloma over the common canaliculus (16.7%). In addition, 39% of patients underwent adjunctive septoplasty and 5.5% had endoscopic sinus surgery at the time of revision.

The published rate of success for patency in endoscopic DCR ranges from 78% to 100% and is approximately 90% in the studies reviewed previously. These rates are comparable to those of the open approach, but both approaches continue to leave an approximate 10% risk of failure, primarily from the formation of scar tissue. This provides an opportunity for the discovery of an adjunctive therapy or procedure to improve the overall outcome of this procedure.

Nasolacrimal Stent Type

Since the development of nasolacrimal stents, a number of different materials have been used, including organic materials, metals, and synthetic materials. Stents are designed to promote longer retention and prevent inflammation. Stenting materials should be inert, pliable, smooth, and readily available. Silicone and polyethylene meet many of these qualifications and are the material of choice for most modern stents.

The decision regarding whether or not to insert a stent is made, by some, based on the tightness of the common canaliculus. This can be assessed by passing a Bowman probe (Integra LifeSciences, Cincinnati, OH) through the canaliculus, and if significant resistance is met, a stent is placed. Bicanalicular stents, which traverse the superior and inferior canaliculi and meet in the common canaliculus, are the most common type used for DCR. Silicone material is used most often, but alternative stent materials have also been used, including small red rubber catheters and C-flex lacrimal catheters (Consolidated Polymer Technologies, Clearwater, FL). A study by Woog, Metson, and Puliafito et al. used traditional silicone tubing, 10-Fr red rubber urinary catheters, and C-flex catheters for stenting purposes. A segment of either red rubber catheter or C-flex catheter was used at the stoma site to aid with patency. They demonstrated success in all 10 cases when the red rubber catheter was used, with no need for revision. The use of this catheter was abandoned later in the study, however, as placement and securement of this catheter type were found to be tedious. They instead opted for a C-flex catheter made of highly biocompatible thermoplastic for the remainder of the study. This catheter was easier to place, as the silicone stent could be placed through the center of the C-flex catheter rather than next to it, as was done with the red rubber catheter. Patency was maintained in all five cases stented with the C-flex stent, but follow-up in this group was shorter. In comparison, only a 68% success rate was found when the silicone stent was used alone. The authors endorse that the earlier cases, which used silicone stenting alone, may have been subjected to increased failure rates related to the learning curve for using the device.

Studies have reported that the silicone stent itself may cause tissue granulation, predisposing the site to postoperative infection, adhesions, and punctal lacerations, resulting in surgical failure and complications. A meta-analysis by Kang et al. found no significant difference in the surgical success rate between endoscopic DCR with silicone intubation and that without stenting. Complications of DCR without stenting can also include synechiae formation and granulation tissue surrounding the stoma.

Modifications to stent materials have also been proposed to improve the success rate in DCR. Applications of antimicrobial substances can, in theory, prevent biofilm formation and prevent stent occlusion. However, these substances create irregularities on the stent surface, which allows for easier adherence and colonization by disruptive bacteria. Uncoated devices have been found to develop similar surface irregularities after exposure to normal living conditions for 3 months. This may suggest a negligible change in risk of biofilm development owing to the physical characteristics of coating materials. Nevertheless, no effective method to prevent formation of biofilms on nasolacrimal stents has been identified. Future endeavors should seek to design materials with nonstick properties and material coatings that can prevent formation of polysaccharide biofilms.

Nasolacrimal Stent Timing

The duration of stenting remains a topic of debate. Some clinicians recommend leaving stents in place for 6 to 9 months if the common canaliculus is markedly stenosed. The logic behind this strategy is that mature scar formation is not complete until several months postoperatively. The recent literature suggests that the majority of mucosal healing is complete at 4 to 6 weeks and a longer duration of stenting may promote granuloma formation and subsequent stenosis. Literature directly comparing duration of stenting with outcomes is lacking. Studies using stenting for an average of 4 to 6 weeks have shown success in 85% to 94% of cases, whereas stenting for longer durations demonstrates success rates between 84% and 96%. A survey conducted in 2016 of members of the American Rhinologic Society demonstrates that nasolacrimal stents are most commonly left in place for 6 to 8 weeks.

Topical Mitomycin C

Antimetabolite medications are also considered by some surgeons as an adjuvant therapy. The medication is applied topically to the rhinostomy site with a goal of scar prevention. These medications have a comparatively well-established role in treatment of tracheal and esophageal stenosis. They are also commonly used after strabismus surgery to prevent fibrosis. The most common antimetabolite used in DCR is mitomycin C. This medication, derived from the broth of Streptomyces caespitosus, creates single-strand breaks in DNA. Mitomycin C preferentially affects rapidly dividing cells, resulting in inhibition of fibroblast and modulation of postoperative fibrosis. This is thought to prevent excessive shrinkage of the ostium and canalicular obstruction. Mitomycin C is applied to the rhinostomy site in a concentration of 0.1 to 0.5 mg/mL with a cotton-tipped applicator. This is applied from 2 to 30 minutes before being flushed with saline solution.

Mitomycin C application has not been shown to provide consistent significant symptomatic benefit, and the majority of surgeons do not routinely use this in DCR. There may be a role, however, in cases of chronic dacryocystitis with external lacrimal fistula, repeated acute dacryocystitis, scar prone situations, and failed DCR cases. Sweeney et al. suggest adjuvant use of mitomycin C in cases of obvious intranasal inflammation at the time of surgery or in cases with a high likelihood of failure, such as in granulomatosis with polyangiitis. Meta-analysis by Xu, Mellington, and Norris demonstrated nearly half the failure rate in revision cases when mitomycin was used compared with cases in which it was not used. The side effects of antimetabolites include maculopathy, wound infection, scleral ulceration, and corneal epithelial defects. Currently the use of topical antimetabolites has not been adopted for routine use in DCR.

Anti-Inflammatory Medications

Topical corticosteroid medications have a variety of applications and are commonly used for their innate anti-inflammatory properties. Corticosteroids are known to suppress inflammation and fibroblast recruitment, resulting in reduced scar formation. The role of steroids in DCR is to prevent abnormal wound healing and prevent ostium granuloma formation. The most common application in DCR is in the form of corticosteroid nasal sprays. The downfall of this delivery method is that nasal sprays distribute the drug mainly on the anterior edges of the inferior and middle turbinates, with only a small amount reaching the rhinostomy site. A few studies advocate steroid injection into the nasolacrimal sac wall for more direct application to the nasal mucosa and theoretically providing a stronger and longer therapeutic effect than topical medications. Injection of 1-mg betamethasone into the nasolacrimal sac intraoperatively has shown good results with improved success rates without introducing new complications. Jo et al. found good response to injection of 0.3 mL of 40 mg/mL of triamcinolone acetonide into granulomas identified in the postoperative period. In fact, the granulomas regressed after steroid injection and the final outcomes were similar to cases without granuloma formation. Complications related to intranasal steroid injection are uncommon, but precautions should be used to prevent catastrophic events such as blindness. Measures taken to mitigate the risk of retinal embolization include thorough mixing to prevent particle clumping, application of topical vasoconstrictors, and slow, gradual injection.

Conclusion

DCR is a safe and effective method for treatment of NLDO. A variety of surgical techniques have been adopted without a clear consensus on the most effective method. The most common reasons for the failure of endoscopic DCR include adhesions, restenosis, and obstruction of the common canaliculus. Stenting is a strategy used to circumvent these complications that is used in most revision cases. However, the role of stenting in primary DCR has been evaluated more recently, and surgical outcomes do not differ in the stented and nonstented cohorts. Additional modifications to DCR include application of antimetabolites and anti-inflammatory medications. These treatments have been shown to have minimal side effects and potentially reduce failure rates, especially in revision cases. Future studies should seek to better elucidate the effectiveness of these medications and standardize the manner in which they are applied.