Endoscopic dacryocystorhinostomy with an otologic T-type ventilation tube in repeated revision cases

It has been reported that endonasal dacryocystorhinostomies (EN-DCRs) yields good esthetic, functional results and similar success rate to that of the external dacryocystorhinostomies (EX-DCRs) [1‐5]. By simulating EX-DCRs in EN-DCRs (mainly in two aspects: 1 maximizing bone window and complete exposure of the inside wall of larimal sac; 2 Ensuring anatomical overlap of the lacrimal sac and nasal mucosal flap), the EN-DCRs can achieve similar successful rates to EX-DCRs [6, 7]. But for the cases of small sac who have lacrimal sac mucosa fibrosis and mucosa scarring resulting from limited residual sac mucosa caused by previous DCRs, the success rate of revision DCR is significantly reduced regardless of the surgical approaches [8, 9]. One of the most important reasons is that it is difficult to form lacrimal sac mucosal flap and nasal mucosa flap anatsatomosis, which causes stoma scarring and resultant surgical failure. Those patients often require repeated revision DCRs, and the revision DCRs often require special surgical approach such as agger nasi cell mucosal autograft for lacrimal sac reconstruction or mucosal grafting [10, 11]. A recent retrospective review has reported that powered EN-DCR is a suitable option for revising failed DCRs in which yields good long-term results [3]. Inserting silicone tubes through the inferior and the superior puncta significantly improved the successful rate of revision DCR [2, 4]. However, silicone tubes were too thin to induce the closure of newly performed ostia for patients with the risk factors of failure. To improve the current EN-DCR procedure, we have inserted an otologic T-type tube in addition to silicone tube into the lacrimal sac cavity., and we have achieved a higher success rate than conventional EN-DCR with careful treatment after surgery.

Modified revision EN-DCR was performed on 22 patients (7 male and 15 female), and the average age for those patients at the time of revision was 44.5 years (range from 26 to 68 years; SD, 15.1). Of those patients, 15 cases have previously undergone revision EN-DCRs twice, and 2 cases have undergone revision EN-DCRs three times. The patients have previously undergone either EX-DCR or EN-DCR, but the latest surgery of all patients was EN-DCR. The time from last failed DCR to recurrence of epiphora varied from 5 to 37 months with a mean time of 12 months. The main causes of previous failure were membranous scarring over ostium (8/22), ostium closure (5/22), or wall of lacrimal sac thickness and fibrosis (9/22) (Table 1). In parallel, 22 patients who received conventional EN-DCR were included as a control group. There were no significant differences in age or gender distribution between the experimental and control groups (Table 2). Of these patients, 18 cases have previously undergone revision EN-DCRs twice and 1 case has undergone revision EN-DCRs three times. The time from last failed DCR to recurrence of epiphora varied from 6 to 35 months with a mean time of 12.5 months (Table 3).

All patients felt the symptom of epiphora, nasal secretions and foreign body sensation. However, these symptoms disappeared after the T-tubes were removed. In the postoperative follow-up visits, 8 patients reported alleviated epiphora before the tube was removed, and 6 patients reported foreign body sensation in nasal cavity or inner canthus with increased nasal secretion. In referral, T-tube was found prolapsed in 9 patients among whom 5 cases prolapsed once, 3 cases prolapsed twice, 1 case prolapsed three times. The tubes were successfully reset whenever prolapses were observed. The duration of tube intubation varied from 6 to 20 weeks, with a mean time of 12.5 weeks. The follow-up time varied from 2 to 3 years with a mean time of 2.3 years. After removing T-tube, 3 patients reported mild epiphora when they felt fatigue in cold weather, but the symptoms disappeared after eyedrops and lacrimal. In the group who underwent modified EN-DCR, 3 patients had complications with 2 cases with eyelid edema which disappeared 1 week after operation and 1 case with appearance of granulation tissue formation surrounding the T-tubes which were cleared under endoscope. In most patients, crusts were found to be formed surround the T-tubes, which were cleared and cleaned with nasal saline in follow-up visits. Eyelid edema and bleeding spots typically recovered 1 week after surgery without interference. In the patients undergoing conventional DCR, 6 patients suffered from eyelid edema and 8 patients reported bleeding spots, and 4 patients experienced epiphora, significantly higher than modified EN-DCR (90.0% vs 13.6%, P < 0.0001) (Table 4).

Table 5 summarized surgical outcomes 2 years after surgery. Anatomical success was achieved in 100% (22/22) of modified EN-DCR and 40.9% (9/22) of conventional EN-DCRs. Functional success was observed in 90.9.0% (20/22) of modified EN-DCR and 22.7% (5/22) of EN-DCR. The difference in surgical outcomes between these two surgeries was statistically significant with a two-sample test for equality of proportions with continuity correction (P < 0.01). With 44 study patients, the power of this test is 80% with α = 0.05 to detect a clinically significant decrease in the success rate of 8%.

In the present study, we slightly modified the approach where an otologic T-tube combined with silicone tube intubation was applied in EN-DCRs. Compared with silicone tube intubation only, this combined approach significantly improved the successful rate. This improvement may be due to the fact that T-tube acts as a support for residual sac mucosa, and it promotes the anastomosis of nasal and lacrimal sac mucosa, which results in spacious and long-lasting channel formation between the sac and nasal cavity, thus easing the ostium stenosis and closure caused by gradual contraction. No severe complications, such as orbital injury, occurred during operation for any of the patients, the incidence of minor postoperative complications such as eyelid edema and bleeding spots were lower than the control group, who have only undergone conventional EN-DCR by the same surgeon.

The closure of newly created ostium is the common cause of failure. It is generally accepted that the size of the excised lacrimal bone and created ostium is important for the long-term outcomes [5]. If the bone window is large enough, the sac is fully exposed, and the anastomosis of nasal and lacrimal sac mucosa flaps is fully achieved, most revision DCRs do not require additional T-tube implants since it prolongs the operation time and adds on additional expense for patients. Moreover, some patients may feel discomfort in the nose and inner canthus caused by T-tube implants. But for patients with high risk of failure such as those with small sac, lacrimal sac mucosa fibrosis and scarring or limited residual sac mucosa, T-tube combined with silicone tube intubation is a better alternative than conventional EN-DCRs.

Our modified procedure achieved both anatomical success rate and functional success rate with lower occurrence of minor complications. The functional success rate was lower than the anatomical success rate probably due to systolic dysfunction of the remnants of lacrimal sac and the poor siphon function of lacrimal point. While we appreciate the moderate success rate of current EN-DCR, we believe that for the revision EN-DCR, T-tube combined with silicone tube intubation needs to be improved due to the two reasons: firstly, it is needed to establish some objective indicators for the combined intubation; secondly, the otologic T-tube we applied was used for the treatment of secretory otitis media but not specifically for DCR. We believe if we can further optimize the material and structure of T-tube to make it more suitable for the application in the lacrimal sac, there will be more widely usage of T-tube in DCR.