Abstract
This study examines the rate of revision surgery after dacryocystorhinostomy and evaluates the associated risk factors in a national patient cohort.
Dacryocystorhinostomy (DCR) is the standard treatment for nasolacrimal duct obstruction. Despite limited data from predominantly small single-surgeon studies showing clear benefit for patient outcomes, silicone stents are commonly placed, and we believe endonasal approaches are increasing in use. In this study, we examined the rate of revision after DCR and evaluated associated risk factors in a national patient cohort.
Methods
Data from all patients undergoing DCR from January 1, 2000, to December 31, 2012, were extracted from the Clinformatics Data Mart Database (Optum). This study was exempt from review by the University of Pennsylvania Institutional Review Board owing to the deidentified data. The primary outcome was the rate of revision DCR within 1 year. For inclusion, patients were required to have 12 or more consecutive months of plan enrollment prior to and after the date of initial DCR. Use of this inclusion period reduced the possibility of a second DCR being considered an initial surgery. Exclusion criteria included insufficient laterality documentation. However, if the initial DCR was bilateral, a second DCR was counted as a revision. Multivariate logistic regression was performed using Stata software (version 14; StataCorp LLC). All P values were 2 sided, and statistical significance was set at < .05. All variables with P < .20 in univariate analysis were included in the final multivariate model.
Results
We found that 1215 patients who underwent DCR met inclusion criteria (903 [74.1%] were female and 312 [38.0%] were male; mean [SD] age, 67.3 [15.5] years), and 98 (8.1%) had a revision within 1 year (Table 1). The category of younger age (patients <30 years) was associated with revision (OR, 2.66; 95% CI, 1.15-6.15; P = .02). Other preexisting diagnoses known to predispose to nasolacrimal duct obstruction were not associated with revision surgery.
Table 1. Cohort Characteristics.
| Characteristic | DCR Without Revision, No. (%) (n = 1117) | DCRs With Revision, No. (%) (n = 98) | Total (n = 1215) |
|---|---|---|---|
| Mean age, y (SD) | 67.4 (15.0) | 66.2 (20.2) | 67.3 (15.5) |
| Age by category, y | |||
| <30 | 30 (2.7) | 8 (8.2) | 38 |
| 30-60 | 218 (19.5) | 13 (13.3) | 231 |
| >60 | 869 (77.8) | 77 (78.6) | 946 |
| Sex | |||
| Female | 832 (74.5) | 71 (72.4) | 903 |
| Male | 285 (25.5) | 27 (27.6) | 312 |
| Race | |||
| White | 844 (75.6) | 77 (78.6) | 921 |
| Black | 94 (8.4) | 6 (6.1) | 100 |
| Hispanic | 45 (4.0) | 2 (2.0) | 47 |
| Asian | 40 (3.6) | 3 (3.1) | 43 |
| Unknown | 94 (8.4) | 10 (10.2) | 104 |
| Placement of a stent | |||
| Yes | 576 (51.6) | 58 (59.1) | 634 |
| No | 541 (48.4) | 40 (40.1) | 581 |
| Surgical approacha | |||
| External | 953 (85.3) | 83 (84.7) | 1036 |
| Endonasal | 115 (10.3) | 12 (12.2) | 127 |
| Glaucoma medication use | 50 (4.5) | 3 (3.1) | 53 |
| History of sinus surgery | 40 (3.6) | 3 (3.1) | 43 |
| Leukemia or lymphoma | 27 (2.4) | 0 (0) | 27 |
| Any prior facial fracture | 16 (1.4) | 4 (4.0) | 20 |
| History of I 131 use | 16 (1.4) | 0 (0) | 16 |
| Sarcoidosis | 15 (1.3) | 0 (0) | 15 |
| Granulomatosis with polyangiitis | 5 (0.5) | 0 (0) | 5 |
| History of docetaxel use | 5 (0.5) | 0 (0) | 5 |
| Nasal/septal fracture | 4 (0.4) | 1 (1.0) | 5 |
| History of fluorouracil use | 3 (0.3) | 0 | 3 |
| Lichen planus | 1 (0.1) | 0 | 1 |
| Naso-ethmoid fracture | 0 | 0 | 0 |
Abbreviation: DCR, dacryocystorhinostomy.
Fifty-two patients (4.4%) had both an external and endonasal procedure code for the same eye on the same day.
There were 634 patients (52.2%) who received a lacrimal stent at initial surgery. Fifty-eight revisions (9.1%) were performed in the patients who initially received a stent. Stent placement was not associated with rate of revision in multivariate analysis (OR, 1.30; 95% CI, 0.85-1.98; P = .23) (Table 2).
Table 2. Final Multivariate Logistic Regression Model Resultsa.
| Characteristic | OR (95% CI) | P Value |
|---|---|---|
| Age by category, yb | ||
| <30 | 2.66 (1.15-6.15) | .02 |
| 30-60 | 0.69 (0.38-1.27) | .24 |
| Placement of stent | 1.30 (0.85-1.98) | .23 |
| Endonasal surgical approach | 0.97 (0.69-1.35) | .84 |
A total of 1036 patients (89.1%) underwent external DCR and 127 (10.9%) endonasal. Ninety-five (8.2%) underwent revision, with 83 (8.0%) having had external and 12 (9.5%) endonasal DCR initially. The rate of revision was not associated with surgical approach in multivariate analysis (OR, 0.97; 95% CI, 0.69-1.35; P = .84) (Table 2).
Discussion
Our study from a large North American database supports findings of smaller single-surgeon studies. Stent placement was not associated with revision. Although stents can maintain patency during the postoperative period, at least 1 study has suggested stents may promote ostial granulation. Another study has shown an association between positive Pseudomonas aeruginosa culture on stents and surgical failure. Surgical approach, external or endonasal, was not associated with revision surgery. Patients younger than 30 years had higher odds of revision; however, the small sample size (38 [3.1%]) within this age category limits generalizability about this finding.
The current study has several limitations related to claims data research. First, prior studies have defined DCR success by resolution of epiphora or patency on irrigation. We defined failure through the surrogate outcome of revision surgery, which may underestimate the number of cases with residual mild epiphora or partial obstruction on irrigation that did not require further surgery. Second, we were unable to review medical records to verify billed procedure codes. Third, because physician choice determined stent placement or surgical approach, we cannot determine if severity of canalicular or nasolacrimal disease biased management.
Conclusions
We found that of 1215 patients, 98 (8.1%) had undergone revision DCR within a year after initial surgery. Numerous medical and surgical factors were evaluated, including the insertion of a lacrimal stent or surgical approach, and we found no association with revision surgery. Knowing the rate of revision in the typical clinical setting may assist prognostic counseling and set performance measures for quality reporting in registries.
References
- 1.Fayers T, Dolman PJ. Bicanalicular silicone stents in endonasal dacryocystorhinostomy: results of a randomized clinical trial. Ophthalmology. 2016;123(10):2255-2259. [DOI] [PubMed] [Google Scholar]
- 2.Chong KK, Lai FH, Ho M, Luk A, Wong BW, Young A. Randomized trial on silicone intubation in endoscopic mechanical dacryocystorhinostomy (SEND) for primary nasolacrimal duct obstruction. Ophthalmology. 2013;120(10):2139-2145. [DOI] [PubMed] [Google Scholar]
- 3.Huang J, Malek J, Chin D, et al. Systematic review and meta-analysis on outcomes for endoscopic versus external dacryocystorhinostomy. Orbit. 2014;33(2):81-90. [DOI] [PubMed] [Google Scholar]
- 4.Longari F, Dehgani Mobaraki P, Ricci AL, Lapenna R, Cagini C, Ricci G. Endoscopic dacryocystorhinostomy with and without silicone intubation: 4 years retrospective study. Eur Arch Otorhinolaryngol. 2016;273(8):2079-2084. [DOI] [PubMed] [Google Scholar]
- 5.Sobel RK, Carter KD, Allen RC. Bilateral lacrimal drainage obstruction and its association with secondary causes. Ophthal Plast Reconstr Surg. 2014;30(2):152-156. [DOI] [PubMed] [Google Scholar]
- 6.Kim SE, Lee SJ, Lee SY, Yoon JS. Clinical significance of microbial growth on the surfaces of silicone tubes removed from dacryocystorhinostomy patients. Am J Ophthalmol. 2012;153(2):253-257.e1, e251. [DOI] [PubMed] [Google Scholar]