Retrospective analysis of congenital nasolacrimal duct obstruction outcomes in a tertiary referral center

The aim of this study was to profile patients with simple and complex congenital nasolacrimal duct obstruction (CNLDO) and to determine the success rates of interventions at a third-level referral center. In this retrospective and comparative study, medical records of patients with CNLDO were reviewed. Demographic data, types of CNLDO, surgical approach, type of silicone intubation, as well as outcome were recorded up to 3 months after the intervention. Obstruction location was identified by probing and irrigation following the American Academy of Ophthalmology (AAO) approach. Success rates were calculated for treatment approach and age group. Based on the outcome of 130 eyes with CNLDO, the overall success rate was 96%. The mean age of patients was 2.5 ± 2.2 months; 73% of cases were simple and 27% were complex CNLDO. Five children had to undergo reoperation; consequently, there was a lower success rate in the complex CNLDO (91%) than in the simple CNLDO subgroup (97%). There was a male predominance (61.5%), which was also present in the simple and complex CNLDO subgroups. The most frequent type of CNLDO was complete nasolacrimal duct obstruction (NLDO, 55.4%), followed by partial (17.7%) and bony NLDO (11.5%). In 87% of eyes, initial probing was performed followed by dacryocystorhinostomy (10%). Mono- or bi-canalicular lacrimal drainage system intubation was used in 86% of eyes. The AAO-based diagnostic approach followed by an individualized treatment regimen yielded excellent success rates of 96% in unselected patients with CNLDO. These success rates were independent of patient age or gender.

Schlüsselwörter CNLDO · Komplexe CNLDO · Kongenital · Pädiatrisch · Erfolgsraten Introduction Congenital tear duct obstruction and specifically congenital lacrimal duct obstruction (CNLDO), with its main symptom of epiphora, followed by discharge refractory to treatment, is relatively common in newborns [1][2][3]. According to previously published literature, between 1 and 12% of the affected infants become symptomatic [4,5] Studies differentiate between a simple variant, which is considered the most common reason for CNLDO [6], caused by the persistence of Hasner's membrane [7], and more complex types of CNLDO [8,9]. These account for up to 17% of cases in patients with CNLDO [6,9] and are associated with syndromes or embryonic anomalies, comprising a myriad of entities such as aplasia of the punctum, partial or total canaliculus aplasia, combined or isolated canaliculus, lacrimal sac, and ductus nasolacrimalis aplasia as well as malformations of the canaliculus system [8,10]. To date, male gender, older age at presentation, and poorer outcomes have been reported to characterize the complex CNLDO patient cohort [9,11].
This retrospective analysis aimed to profile patients with CNLDO and determine the success rates of interventions at a third-level referral center. In these patients the exact location of the obstruction was identified by probing and irrigation following the diagnostic approach of the American Academy of Ophthalmology (AAO; [12]).

Methods
This retrospective study was undertaken at the Department of Ophthalmology and Optometry, Medical University of Vienna, Austria. This study was approved by the Ethics Committee of the Medical University of Vienna (ClinicalTrials.gov Number: NCT04931186) and followed the tenets of the Declaration of Helsinki. Medical records of patients with CNDLO who were referred to the Department of Ophthalmology and Optometry, Medical University of Vienna, Austria, between 1 January 2013 and 31 December 2017 were reviewed.
Presence of facial malformations, eyelid position disorders, abnormal nasal bone structure, agenesis or ectopic lacrimal puncta, or congenital fistulas of the lacrimal sac were not considered exclusion criteria as this study investigated the whole spectrum of CNLDO.
The operation performed was based on age, clinical symptoms, and probing results. Examining the patients under general anesthesia helps to determine the exact location of the obstruction and consequently allows for differentiation between simple and complex CNLDO, which determines the surgical approach.
Examination of the lacrimal drainage system was conducted using a Bangerter cannula. Probe advancement (soft stop, hard stop) and irrigation were used to locate the site of obstruction, as illustrated by the AAO [12]. Consequently, we differentiated between: 1. Complete canalicular obstruction: The Bangerter cannula is advanced with difficulty, and irrigation fluid refluxes from the same canaliculus 2. Complete common canalicular obstruction: A "soft stop" is encountered at the level of the common canaliculus, and irrigated fluid refluxes through the opposite punctum and sometimes partially from the same canaliculus as well 3. Complete nasolacrimal duct obstruction (NLDO): The cannula is easily advanced to the medial wall of the lacrimal sac, then a "hard stop" is felt, and irrigation fluid refluxes through the opposite punctum. Often, the refluxed fluid contains mucus and/or pus 4. Partial NLDO: The cannula is easily placed, and irrigation fluid passes into the nose as well as refluxing through the opposite punctum [12] 5. Bony obstructions [9,11,13]: The cannula is easily advanced to the medial wall of the lacrimal sac, then a hard stop is felt, and irrigation fluid refluxes through the opposite punctum and sometimes partially from the same canaliculus too. Turning the cannula by 90 degrees results in another hard stop and the cannula cannot be advanced any further. Again, irrigation fluid refluxes through the opposite punctum and sometimes partially from the same canaliculus 6. Atresia of the punctum: No lacrimal punctum is visible upon investigation K CNLDO 167

original article
We pre-specified that complete and partial NLDO are considered simple CNLDO, all others are considered complex CNLDO. All four surgeons performing the operations were trained ophthalmologists undergoing specialized surgical training for oculoplastics including pathologies of the lacrimal drainage system.
Data extracted from medical records comprising demographic data, data on the time between initial diagnosis and intervention, surgical approach, type of silicone tube utilized (mono-or bicanalicular silicone tube), silicone tube loss, clinical symptoms 1 week and 3 months after surgical treatment, as well as preand postsurgical medication were recorded. Also, rereferrals were recorded, as patients were instructed to report back to our center if symptoms reoccurred.
Treatment success was defined as the absence of clinical signs of lacrimal drainage system obstruction (epiphora, increased tear leak, mucous discharge).
Indication for re-treatment was defined as the presence of clinical signs of lacrimal drainage system obstruction, closure of DCR anastomosis, as well as wound dehiscence.
Success rates were calculated for the following approaches: Initial probing: insertion of a modified Bangerter lacrimal cannula to mechanically rupture Hasner's valve or by additionally applying high-pressure irrigation DCR [14]: creation of a new tear drain between the lacrimal sac and the nose through the bone to the nasal cavity Canalicular curettage [15]: Using a curettage, the canaliculus is freed from debris Punctoplasty: A 1-to 3-snip procedure [16] was used Results are reported for different predefined age groups, which are based on studies and Kaplan-Maier analysis reporting success rates in different age groups: patients aged between 2 and 6 months (group A); patients aged between 6 and 12 months (group B); patients older than 1 year (group C); and patients aged over 2 years (group D; [17][18][19]).
As the data of this subgroup analysis are part of an analysis including all age groups (see: Clinical-

Results
A total of 130 eyes of 130 children with CNLDO were evaluated in this retrospective analysis; 50 were female (38.5%) and 80 were male (61.5%).
The demographic profile of all four age groups A-D is summarized in Table 1.

Fig. 2
Detailed summary of surgical interventions performed in children with congenital nasolacrimal duct obstruction. This figure also displays the type of intubation used. DCR dacryocystorhinostomy, BSI bicanalicular silicone intubation, Monoka monocanalicular silicone intubation, numbers in boxes represent n There were statistically significant differences between age groups regarding the time until intervention (p < 0.0001, ANOVA). Patients aged between two and 6 months (group A) at initial presentation were scheduled for surgical intervention after a median time of 270 (CI: 178-452) days. Patients aged between 6 and 12 months (group B), older than 1 year (group C), and patients aged over 2 years (group D) had a similar waiting time for surgical intervention of 122 (68-135) days, 135 (97-176) days, and 161 (127-215) days, respectively. Statistically significant differences between groups were only found when comparing group A with all other groups (all p < 0.0001).
A detailed summary of treatment prescribed postoperatively is given in Table 2. The most frequent postoperative medication used was a combination of dexamethasone and gentamicin with naphazoline eye drops (36.9%), antibiotic ointment (29.2%), or dexamethasone and gentamicin eye drops (16.2%).
Of the eight children scheduled for re-operation, six were planned to undergo balloon dacryoplasty with BSI and two received an appointment for conjunctivodacryocystorhinostomy (CDCR). Of these eight children, only five (two for CDCR and three for balloon dacryoplasty with BSI) appeared for reoperation; in three, the problems resolved with no need for further treatment.
Consequently, we calculated an overall success rate of 96% based on the evaluation of 130 eyes. Table 3 summarizes the success rates for groups A-D. There is no statistically significant difference in success rates between simple and complex NLDO (p = 0.13). There is also no statistically significant difference in success rates between surgeons (p = 0.72).

Discussion
This study presents a comparative profile of patients referred to a tertiary eye center for CNLDO. The location of the obstruction was identified by probing and irrigation following the AAO approach [12]. We retrospectively analyzed charts aiming at reporting success rates for simple (73.1%) and complex (26.9%) cases of CNLDO as well as characterizing the patients' profiles in a cohort of 130 patients, with a mean age of 2.5 ± 2.2 years. As previously described [20][21][22], we found an overall male predominance (61.5%), which is also true when grouping for simple (62.0%) and complex (60.0%) CNLDO.
The predominant type of CNLDO was complete NLDO (55.4%, see Fig. 1), which is classified as a reflux of fluid through the opposite punctum after the advancement of the cannula ends with a "hard stop" [12].
The overall success rate was 96% in this patient's collective. In accordance with previous studies [21,22] as well as a study with a similarly large patient cohort (138 eyes of 101 patients; [20]), lower success rates were found in complex versus simple CNLDO, which also applies to our study (91% and 97%, respectively). However, the difference in success rate did not reach the level of significance. Caution is advised when comparing results, as the included patients represent a very heterogeneous group.
We could not confirm the low success rates of initial probing in age groups C and D (1-2 years, > 2 years, respectively) published in the literature [7,19,21], ranging between 94 and 98% in our patient collective. Also, Valcheva et al. and Lin et al., both could not show an association of success rates with age [23,24]. Yet, it should be mentioned that most of our patients undergoing initial probing in age groups C and D received additional lacrimal drainage system intubation (81% and 86%, respectively) in order to increase success rates.
It is well documented that there are several confounders influencing success rates in complex CNLDO; these are older age (> 36 months), failed conservative treatment, failed initial probing, or bony obstructions, e.g., in craniofacial anomalies [20][21][22]. In our patient cohort we did not find a statistically significant difference between simple and complex CNLDO groups regarding age (p = 0.08); however, there was a tendency toward older age in the group with complex CNLDO. Further, we cannot confirm results referred to earlier that unsuccessful conservative pretreatment lowers success rates. Of 27 eyes receiving unsuccessful conservative pretreatment, only one patient (3.8%) was scheduled for re-treatment after initial surgical intervention, whereas seven (7.3%) out of 103 eyes without pretreatment were scheduled for retreatment (p = 0.55). Further, we could not show that bony NLDO is associated with lower success rates. This may be due to the fact that out of 15 eyes with bony NLDO, we treated 13 (86.7%) with primary DCR combined with silicone intubation. This approach is supported by previous studies. First, a meta-analysis, which shows that DCR success rates improve by 5% as soon as silicone intubation is performed. Second, a study showing DCR outcome and complication rates to be the same between pediatric and adult patients [25], ranging between 77 and 95% [25,26]. For all the patients undergoing primary DCR included in this study, an underlying condition was recorded, involving different types of syndromic craniosynostosis [27] and one patient with Goldenhar syndrome [28].
In 86% of eyes, lacrimal drainage system intubation was used. This was most commonly performed by BSI or monocanalicular intubation (see Fig. 2). Consequently, 23 patients did not receive lacrimal drainage system intubation. The data show that the decision not to use any intubation means was clearly based on age. In only three patients in the 0-12 months age group, but in 20 patients aged > 1 years or older, intubation was omitted.
Patients aged between 2 and 6 months had to wait statistically significantly longer (median 312 days) than all other groups. This is due to the fact that Kaplan-Meier analysis shows spontaneous resolution plateaus of between 40 and 50% resolution after 9 months of age [19]. However, there were no spontaneous resolutions recorded in this patient collective. The overall median time until intervention was 142 days (92-192), which is different from previous studies in which the time ranges up to 600 days [29]. However, few data are reported in this regard.
Three months after intervention, the most common symptoms reported were epiphora (9.8%) and NLDO (8.7%). Yet, not all of these patients needed re-treatment, leading to an overall success rate of 96% (see Table 3).

Limitations
Our study has some limitations that warrant further comment; it is of retrospective nature and only a 3month follow-up is available. This is due to the fact that after the 3-month follow-up, patients are discharged to the referring ophthalmologist with the premise to report back to our institution in the case of symptom worsening. Furthermore, the CNLDO classification is based on the AAO probing approach [12], which supports precise identification of the obstruction when performed by experienced surgeons.

Conclusion
This retrospective analysis of patients referred to a tertiary center shows that the American Academy of Ophthalmology-based diagnostic approach followed by an individualized treatment regimen yields excellent success rates of 96% in unselected patients with congenital nasolacrimal duct obstruction. These success rates are independent of patient age or gender. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.