Optimizing clinical outcomes for endoscopic lacrimal duct recanalization in patients with complete PANDO

The study aims to describe modifications and refinements in the technique and technology of Transcanalicular Endoscopic Lacrimal Duct Recanalization (TELDR) based on anatomical foundations to optimize its clinical outcomes in patients with complete Primary Acquired Nasolacrimal Duct Obstruction (PANDO). The medical records of 115 patients who underwent TELDR procedures from January 2018 to July 2020 were reviewed retrospectively. Of the 115 patients, only those 35 patients with complete PANDO characterized by longstanding epiphora of 3–5 years duration, dense, diffuse fibrous tissue obstruction involving the sac, sac duct junction and the entire length of the nasolacrimal duct were included in the study. Parameters for success were analyzed based on patency on irrigation, functional endoscopic dye test, and improvement of epiphora. Forty-five cases from 35 patients with complete PANDO were included in the study. The mean length of time from the date of operation to silicone stent removal was 8.1 weeks, while the mean length of follow-up starting from the removal of silicone stent to last follow-up was 61.0 weeks. There were 95.6% anatomic patency on canalicular irrigation with saline and 95.6% functional patency based on functional endoscopic dye test. There was significant improvement of epiphora (p value of < 0.0001) post-operatively. The results of modified TELDR improved clinical outcomes and could be a definitive treatment in patients with complete PANDO with longstanding, dense, diffuse, fibrous tissue obstruction. Patients who experience reobstruction, may undergo a repeat of the recanalization approach.

system without altering its structural and physiologic integrity [2].
The advantages of transcanalicular endoscopic recanalization procedure of the lacrimal drainage system are its minimally invasive approach, decreases morbidity such as risk of bleeding and shortened recovery time [3][4][5][6].
To my knowledge, no study has been done on the most appropriate, safe and effective surgical technique of recanalization for definitive treatment of patients with complete longstanding, dense, diffuse fibrous tissue obstruction [20] involving the sac, sac duct junction and entire length of the nasolacrimal duct. Diagnosis as to type and level of complete obstruction was made at the time of operation during the recanalization procedure confirmed by dacryoendoscopic visualization of the entire lacrimal excretory system and not merely by conventional lacrimal examination (lacrimal irrigation) and syringing.
The aim of this study is to describe the modified and refined technique of TELDR to optimize its clinical outcomes in patients with complete PANDO as described above, not only of selected cases of PANDO such as focal stenosis, mucus, stones, granulations and focal fibrotic membranes.

Materials and methods
The medical records of 115 patients who underwent Transcanalicular Endoscopic Lacrimal Duct Recanalization were reviewed retrospectively. Patients were seen at the Javate Lacrimal, Orbital and Oculofacial Plastic Surgery Clinic. Procedures were conducted from January 2018 to July 2020 at the University of Santo Tomas Hospital. This study is in accordance with the guidelines of the Declaration of Helsinki of 1975, as revised in 2000 and submitted to the Research Ethics Committee of the University of Santo Tomas Hospital, Manila, Philippines.
Of the 115 patients included in the chart review, only those 35 patients (45 cases) with complete PANDO characterized by longstanding epiphora of 3-5 years duration, dense, diffuse fibrous tissue obstruction involving the sac, sac duct junction and entire length of the nasolacrimal duct diagnosed at the time of operation during the recanalization procedure through dacryoendoscopic visualization of the entire lacrimal excretory system as stated in the medical and operative record were included in the study.
Short-distance focal obstructions confined to one anatomic region, as distinguished from long-distance diffuse lesions that covered the entire length of the nasolacrimal duct as described in the operative records were not included in the study [20][21][22]. Cases with complete primary acquired nasolacrimal duct obstruction with concomitant canalicular obstructions were also omitted from the study.
The following information was recorded: Length of time from the date of operation to date of removal of silicone stent, the length of time from the removal of stent up to the last follow-up of the patient. The following parameters for success were also recorded such as symptomatic relief of epiphora indicated by Munk's [23] score pre-operatively and post-operatively, functional patency exhibited by a positive functional endoscopic dye test, and anatomic patency on canalicular irrigation with saline. Munk's score determines the degree of tearing or epiphora of a patient with NLDO based on the frequency of dabbing or drying of tears. Endoscopic visualization of the fluorescein dye coming out from the opening of the nasolacrimal duct indicates functional patency of the recanalized nasolacrimal duct. Complications arising intra-operatively, post-operatively, and during follow-ups were also noted. A single surgeon (RMJ) performed all the operations.
Surgical technique for modified TELDR All TELDR procedures were performed under general anesthesia. The ipsilateral nostril was packed with cotton pledgets soaked in 0.05% oxymetazoline hydrochloride. The inferior turbinate was mobilized medially using a Freer elevator to facilitate access into the inferior meatus and to avoid passage of the trephine through the inferior turbinate which is usually tightly adherent to the lateral nasal wall in most cases.
The lacrimal puncta were expanded using punctum dilators of increasing caliber with the dilating probe directed through the punctum vertically to the base of the ampulla 2-3 mm below. The dilating probe was then turned nearly horizontal to follow the contour of the eyelid margin. The eyelid was drawn temporally to prevent kinking of the canaliculus. The passage of the larger punctum dilator allowed the introduction of the trephine.
The lacrimal trephine (Machida Endoscope Co., Ltd, Japan) ( Fig. 1) slightly bent with a 0.95 mm outside diameter, 60 mm length was then passed vertically 3 mm through the dilated punctum. The lacrimal trephine was then turned to the near horizontal to parallel the margin of the eyelid in its passage to the lacrimal sac.
The microendoscope (Machida Endoscope Co., Ltd, Japan) ( Fig. 2) has the following specifications: 60 mm length, probe outer diameter, 0.7 mm; field of view, 65°; number of image elements, 5000; and observation depth, 5 mm. This was inserted into the lumen of the lacrimal trephine to view the canaliculus. While advancing, the microendoscope was maintained coaxial to the canaliculus, and intermittent fluid infusion using a 10 cc syringe was done to keep the cavities under examination open throughout the procedure.
At the completion of the horizontal pass, the trephine with the microendoscope was rotated to a vertical position, which allowed it to visualize the lacrimal sac lumen and the movement of the lateral wall of the lacrimal sac, which served as a guide to the direction of the trephine. As mentioned, the lacrimal trephine is slightly bent so as not to hit the posterior wall of the lacrimal sac to allow its smooth passage into the sac duct junction or opening of the NLD and to adapt to the relative orientation of the lacrimal fossa to the bony nasolacrimal canal. To avoid creating a false passage, the lateral descent of the nasolacrimal duct also served as a guide. The divergence of the descending course of the nasolacrimal duct was estimated clinically by drawing a line between the tear sac and ala nasi [24]. Individuals with narrow interorbital distances and wide noses will exhibit the greatest lateral divergence, while those with wide interorbital distances and narrow noses will exhibit a more vertical divergence. Care was taken to maintain the position of the microendoscope coaxial to the sac, and to visualize the lumen and the movement of the walls of the lacrimal sac. As the trephine and microendoscope passed through the lacrimal sac, the surgeon palpated for the trephine from the external surface of the sac-duct junction using his index finger. While viewing the video images of the microendoscope traversing from the punctum to the inferior meatus, the surgeon was able to localize fibrous tissue obstructions along the sac, sac duct junction and the entire nasolacrimal duct. At the same time, the trephine was utilized to pierce through and remove dense fibrous tissue obstructions. A rigid 4-mm endoscope (Karl Storz GmbH and Co., Tuttlingen, Germany) was used to view the lacrimal trephine below the vault of the anterior end of the inferior nasal meatus. Next, the microendoscope with the trephine was gradually withdrawn to view the recanalized nasolacrimal duct up to the sac.  The Machida microendoscope (Machida Endoscope Co., Ltd, Japan) with the following specifications: 60 mm length, probe outer diameter, 0.7 mm; field of view, 65°; number of image elements, 5000; and observation depth, 5 mm Then, the microendoscope was connected to a flow-regulated roller pump (Endomat LC, Karl Storz GmbH and Co., Tuttlingen, Germany) which provided continuous and forceful irrigation with sterile water to wash off remaining fibrous tissue obstruction.

Technique in balloon dacryoplasty
After performing recanalization, balloon dilatation of the nasolacrimal duct was done using the Ophtacath ® Lacrimal Duct Balloon Catheter (FCI, 20-22 rue Louis Armand, 75015 Paris, France) which has a length of 15 mm, a diameter of 0.75 mm uninflated and a diameter of 3 mm once inflated in order to achieve reaming or widening of the lumen of the nasolacrimal duct. The catheter tip was lubricated with tobramycin dexamethasone ointment and was introduced into the lacrimal system in an anterograde direction via the lacrimal punctum. After inserting, the catheter was then directed vertically and advanced beyond its 20 mm mark. Nasal endoscopy was then done to confirm the passage of the catheter through the nasolacrimal duct.
The balloon catheter was withdrawn until the 15 mm mark proximal to balloon, aligned with the punctum. Afterward, it was inflated up to a fill pressure of 8 atm for 90 s and then completely deflated (Fig. 3). An additional inflation was done up to 8 atm, this was held for 60 s and deflated once more. The catheter was withdrawn up to the 10 mm mark, after which a second inflation cycle was started to a pressure of 8 atm for 60 s. It was inflated again up to 8 atm for 40 s, then deflated. The catheter was then withdrawn until the tip was visualized.

Technique in silicone intubation
Upon completing recanalization and balloon dacryoplasty, bicanalicular silicone intubation was done using either one of the following: Crawford Bicanalicular silicone tube, Ritleng + Lacrimal Intubation Stent or Nunchaku silicone stent (FCI, 20-22 rue Louis Armand, 75015 Paris, France). For ease of silicone intubation, the author recommends the Nunchaku, a pushed silicone self-retaining bicanaliculus intubation stent (Fig. 4). The metallic guides are located inside the lumen, not as an extension of the stent as in conventional stents and therefore, no nasal retrieval was needed. Since the stability of the stent was guaranteed by the designed silicone tubes, knots and sutures in the nasal fossa were not needed.

Post-operative care
Post-operative medication consisting of steroid-containing antibiotic eye drops (TobraDex®; Alcon Laboratories, Inc., Fort Worth, Texas, USA) was instilled every three hours to the affected eye.
Canalicular irrigation with steroid-laden antibiotic eye drops was started on the seventh post-operative day and then weekly for the next two months in succession. The frequency of irrigation was then reduced to monthly, for the next four months. Finally, it was  Fig. 4 For ease of silicone intubation, the Nunchaku, a pushed silicone self-retaining bicanaliculus intubation stent was used further decreased, conducting irrigation quarterly, every three to four months.
The following parameters were monitored to assess the outcome of surgery: a patent lacrimal flow on irrigation without any reflux, a positive Functional Endoscopic Dye Test (FEDT) and the resolution of infection and epiphora with no recurrence of epiphora during the follow-up periods after removal of the stent.

Results
Forty-five cases were included in the chart review. Thirty were female and five were male with a mean age of 51 years old ( Table 1). The mean length of time from the date of operation to silicone stent removal was 8.1 (6.7-11.7) weeks, while the mean length of follow-up starting from the removal of silicone stent to the last follow-up was 61.0 (4.0-95.1) weeks.
There were three parameters for success used in this study. First, there was anatomic patency on canalicular irrigation with steroid-laden antibiotic eye drops. Results showed that there was a prevalence of 95.6% anatomic patency on canalicular irrigation after TELDR with balloon dacryoplasty and silicone intubation. Functional patency was evaluated using FEDT and results also indicated a 95.6% success rate. Munk's score was used in determining the improvement of epiphora. Results showed a significant improvement of epiphora post-operatively, with a p value of < 0.0001 on all comparisons between pre-operative values versus one-week post-op, one week after stent removal and last visit, thus showing a clinically significant difference (Table 2).
In this study, complications such as hematoma and edema of the surrounding soft tissues did not occur as observed in the previous study [18]. In addition, there was no damage to the fibroelastic punctal ring and cheese wiring.
Modifications and refinements in the technique and technology of TELDR start with meticulous endoscopic technique based on the anatomy of the lacrimal excretory system and the surrounding soft and bony tissues. The previous design of the lacrimal trephine, measuring 1.1 mm and 1.3 mm outside diameter, can cause damage to the fibroelastic punctal ring and cheese wiring of the horizontal canaliculus to a small sized punctum (< 0.4 mm) [27]. Damage to the common canaliculus junction resulting in  1 (1-2) z-score, p value 6.1, < 0.001 1 week after stent removal 0 (0-1) z-score, p value 5.9, < 0.001 Last visit 0 z-score, p value 6.0, < 0.001 extravasation of fluid and edema in the surrounding soft tissues can also occur [18]. Thus, a modified slightly bent, smaller outside diameter lacrimal trephine of 0.95 mm was utilized. A study by Carter et al. [28], demonstrated variations in lacrimal punctal sizes. The upper puncta showed a mean area of 0.264mm 2 , while the lower puncta showed a mean area of 0.321 mm 2 . Tucker et al. [29], published an article on the Anatomy of the Common Canaliculus, the canaliculi bend at the canaliculus junction at an angle of 118°, before passing anteriorly to enter the lacrimal sac at an acute angle of 58°.
This size variation of the lacrimal punctum and anatomy of the common canaliculus are important to consider when doing recanalization of the nasolacrimal system.
During descent of the trephine microendoscope into the sac duct junction, nasolacrimal duct, it will hit the posterior wall of the lacrimal sac especially in patients with prominent frontal bone. A slightly bent trephine with the microendoscope was used to allow its smooth passage into the sac duct junction or opening of the NLD without hitting the posterior wall of the lacrimal sac (Fig. 5a, b).
The newly designed trephine and microendoscope allowed the fibrous tissue obstruction to be pierced through and removed in the sac, sac duct junction and the nasolacrimal duct (Fig. 6a, b). The complete dense fibrous tissue obstructions of the NLD require a stiff instrument such as the trephine which can be advanced forcefully [12]. The use of a trephine, unlike sheath-guided dacryoplasty, avoids technical failure during recanalization of dense, diffuse, fibrous tissue obstruction. A procedure is defined as a technical failure when it fails to recanalize or pierce through and remove the obstruction [30]. Plastic sheath guides will have difficulty in removing epithelial and subepithelial fibrosis. According to Linberg and McCormick [22], the nasolacrimal duct with longstanding obstruction is obliterated by dense fibrous tissue. As a result, probing or intubation would be unlikely to produce a patent duct.
Lee and Lew [19] using sheath-guided dacryoplasty, showed a success rate of 74% in the diffused pattern of obstruction in their study on Transcanalicular Endoscopic Dacryoplasty in patients with primary acquired Nasolacrimal Duct Obstruction. By contrast, the success rate of this study was 95.6% using trephine for cases of complete PANDO as described above.
As mentioned in the surgical technique, with the trephine entering the osseous nasolacrimal canal, the lateral divergence of the descending course of the nasolacrimal duct can serve as a guide to avoid false passage or false route during recanalization. It can be estimated by drawing a line between the tear sac and the ala nasi. Individuals with narrow interorbital distances and wide noses will show the greatest lateral divergence along the descending course of the nasolacrimal duct, identified as the "outward type." Alternatively, the "inward type" is characterized by those with wide interorbital distances and narrow noses will exhibit a more vertical divergence of the descending course of the nasolacrimal duct [24].
When the sac is not completely obliterated or obstructed, one can see the lateral wall of the lacrimal sac changes flexibly with pressure [31]. Under   Fig. 5 a, b A slightly bent trephine with the microendoscope in its lumen allowed its smooth passage into the sac duct junction or opening of the NLD without hitting the posterior wall of the sac positive pressure, the lateral wall moves outward, which can serve as a guide to the direction of the trephine during recanalization to avoid false passage.
A flow-regulated roller pump (Endomat LC) was used to provide continuous and forceful irrigation with sterile water to completely wash off any remaining fibrous tissue obstruction [32,33]. Any remnant of fibrous tissue obstruction can possibly again initiate inflammation and malfunctions in the cavernous body with reactive hyperemia, swelling of the mucous membrane, and temporary occlusion of the lacrimal passage which can lead to reobstruction [20,22].
Balloon Dacryoplasty (DCP) is based on transluminal angioplasty. With the addition of balloon dacryoplasty, the application of enough force with the lacrimal balloon duct catheter along the nasolacrimal duct in a radial and longitudinal manner directly will lead to greater dilatation as compared to probing or intubation, a geometric effect of widening the NLD lumen, producing the so-called reaming of the NLD [23,24].
Despite achieving initial patency after trephination, the previously obstructed portions of the nasolacrimal duct may narrow subsequently. Balloon Duct Catheter will not only dilate stenotic areas but will negotiate membranes, webs and granulomas along the duct that the initial trephination may not be able to remove. These areas may serve as foci for stasis, infection, inflammation, and fibrosis, which may ultimately lead to reobstruction. To further improve long-term outcomes, balloon dacryocystoplasty [12,17,35,36] was integrated with the TELDR technique.
The integration of balloon dacryoplasty using the Ophtacath Lacrimal Duct Balloon Catheter (FCI, 20-22 rue Louis Armand, 75,015 Paris, France) improved clinical outcomes of treatment in the modified TELDR. In the author's experience, dilation or widening of the lumen of the nasolacrimal duct after balloon dacryoplasty permits easier silicone stent intubation intraoperatively, smooth canalicular irrigation (decrease flow resistance), increased flow volume and absence of regurgitation on irrigation on followup. Thus, there will no longer be a need for assisted patency, which was done in the previous study [18].
According to Perry et al. [17] in his paper on Balloon Catheter Dilation for Treatment of Adults with Partial Nasolacrimal Duct Obstruction: A Preliminary Report, by mechanically restoring patency, balloon catheter dilation may break this cycle of stasis and inflammation, allowing the soft tissue to regain its normal anatomy.
In a study by Goldstein et al. [37], the effectiveness of balloon dacryoplasty procedure in humans supports this procedure in treating adults and children with nasolacrimal duct obstruction.
The addition of silicone intubation after TELDR avoided the need for coagulation of the "raw" bleeding areas which prevented the adhesion of the nasolacrimal duct mucosa as described in the previous study [18]. Silicone intubation acted as temporary stent maintaining patency while the surrounding tissues heal after recanalization [38]. According to Moscato et al. [39], silicone stent placement allows dilation of the soft tissue (mucosal) portion of the lacrimal outflow system, thereby, reducing flow resistance and potentially increasing flow volume. The stent may also serve to straighten the canaliculi, common canaliculus and other parts of the system, thereby, allowing increased flow. It may lyse mucosal synechiae that could contribute to a narrowing of the nasolacrimal duct. It is also possible that the stent allows increased flow in unrecognized areas of valvular stenosis or fibrosis. This study showed a 95.6% success rate both for anatomic and functional patency postoperatively with two cases of reobstruction.
The first case involved a patient presenting with epiphora, 3 months and 17 days s/p TELDR. Dacryoendoscopy done revealed focal stenosis in the nasolacrimal duct. Canalicular irrigation was done to achieve patency with resolution of infection and epiphora during subsequent follow-up periods. It was postulated that the focal stenosis was a result of previous rhinitis that the patient had. Paulsen described his insights into the pathophysiology of primary acquired dacryostenosis [20]. According to him, ascending inflammation from the region of the nose (perhaps by way of inflammation of "ectopical" nasal epithelial cells in the nasolacrimal duct) may initiate malfunctions in the cavernous body with reactive hyperemia, swelling of the mucous membrane, and temporary occlusion of the lacrimal passage. Ascending infection of these atypical cells during nasal inflammation could be the starting point of dacryostenosis . With this occlusion of the lacrimal passage there will be repeated isolated episodes of dacryocystitis.
The second case involved a patient, 1 month and 8 days s/p TELDR, with chronic sinusitis, right maxilla. The patient presented epiphora with whitish mucoid discharge upon irrigation. Revision done under general anesthesia as per patient's request revealed focal stenosis in the sac duct junction and nasolacrimal duct. This further supports the theory of possible ascending infection from the nose (sinusitis) which was managed by doing a repeat recanalization and medical treatment of the sinusitis. A two-month follow-up after removal of the stent and subsequent follow-up periods showed patent nasolacrimal duct upon irrigation with no reflux, positive Functional Endoscopic Dye Test (FEDT), resolution of infection and epiphora with resolution of symptoms.
In summary, the initial results of this study are encouraging and provide a good baseline/foundation for this procedure, which is technical and practical. This minimally invasive approach of modified TELDR with balloon dacryoplasty and silicone intubation in patients with complete PANDO could spare the patient of resorting to conventional lacrimal surgery such as External or Endonasal DCR with its associated risks and complications. If reobstruction occurs in the future, the patient may just undergo a repeat of the recanalization approach.

Conclusion
The results of this study showed a high success rate of 95.6% in both functional and anatomic patency with significant improvement of epiphora. Modified TELDR could be a definitive treatment for patients with complete Primary Acquired Nasolacrimal Duct Obstruction with longstanding, dense, diffuse fibrous tissue obstruction involving the sac, sac duct junction and the entire length of the nasolacrimal duct, and not only for circumscribed focal stenosis or selected cases of obstruction. Patients who experience reobstruction may undergo a repeat of the recanalization approach.
The procedure could be used both for diagnostic and therapeutic aspects and is cost-efficient for the patient thus, requesting additional ancillary procedures such as dacryocystography, dacryoscintigraphy, CT scan DCG and dacryoendoscopy to determine the appropriate surgical procedure will no longer be necessary.

Recommendations
While the study has a number of strengths, there are a few limitations to address. One of which is that it has a small sample size. However, it is recommended that future research can also examine the procedural success rates of larger groups of patients with longer periods of follow-up.