- Email: [email protected]
Evaluation of treatment modalities and prognostic factors in children with congenital nasolacrimal duct obstruction
Evaluation of treatment modalities and prognostic factors in children with congenital nasolacrimal duct obstruction Yasser H. Al-Faky, MD, FRCS, Nora Al-Sobaie, Ahmad Mousa, PhD, Hessa Al-Odan, MD, Reem Al-Huthail, Essam Osman, MD, FRCS, and Abdul Rahman Al-Mosallam, MD PURPOSE
To evaluate the success rates of probing and silicone intubation and risk factors for failure of these procedures.
METHODS
This retrospective cohort study included patients who were diagnosed with congenital nasolacrimal duct obstruction (NLDO) and underwent probing or intubation as the primary procedure. Demographic data, clinical features, intraoperative, and postoperative data were evaluated. The patients were divided into four subgroups based on age: 12-24 months, 25-36 months, 37-48 months, and .48 months. The procedure was considered successful when all preoperative signs disappeared with normal dye disappearance test and a positive Jones primary dye test. Statistical analysis was performed to determine the risk factors for failure, and P \ 0.05 was statistically significant.
RESULTS
The cohort comprised 350 subjects (162 males [46%]) who underwent a total of 454 nasolacrimal duct surgeries. The mean chronological age at time of surgery was 32.6 22.1 months (range, 12-132 months). The overall success rate was 84.8%. Probing was performed 188 times, with a success rate of 80.3%; intubation was performed 266 times, with a success rate of 88%.
CONCLUSIONS
Intubation was more successful than probing in patients with bilateral congenital nasolacrimal duct obstruction, Down syndrome, history of acute dacryocystitis, lack of preoperative massage, or intraoperative observation of either a “tight” obstruction or obstruction at sites other than at the level of Hasner’s valve. Primary nasolacrimal duct intubation should be considered in these higher risk patients. ( J AAPOS 2012;16:53-57)
T
he two most common surgical approaches for congenital nasolacrimal duct obstruction (NLDO) are probing and silicone intubation. Many surgeons consider probing the most reliable procedure and perform it as the primary procedure for congenital NLDO; however, success rates with probing have been variable.1-4 Intubation is usually reserved for cases where probing has failed or as a primary procedure in older children and in patients with Down syndrome.1,2,4,5 Success rates of intubation published in medical literature have also been disparate.2 Due to the variable success reported with both procedures, the use of probing or intubation as the primary procedure for congenital NLDO remains controversial. Additionally, there is a relative paucity of studies that Author affiliations: Department of Ophthalmology, College of Medicine, King Saud University, Riyadh, Saudi Arabia Submitted January 17, 2011. Revision accepted July 5, 2011. Correspondence: Yasser H. Al-Faky, MD, FRCS, Ophthalmology Department, College of Medicine, King Saud University, King Abdul Aziz Street, Riyadh, 11411, Saudi Arabia (email: [email protected]). Copyright Ó 2012 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/$36.00 doi:10.1016/j.jaapos.2011.07.020
Journal of AAPOS
have evaluated the effect of the preoperative and intraoperative risk factors on the success of these procedures. We evaluated the pre- and intraoperative risk factors and their effect on the success rates of probing or intubation. These data may assist clinicians in determining the appropriate primary procedure for congenital NLDO.
Patients and Methods This retrospective cohort study comprised subjects diagnosed with congenital NLDO who underwent probing or intubation as the primary procedure. All subjects were diagnosed and treated at the Ophthalmology Department of the College of Medicine, King Saud University, between 1998 and 2008. Subjects were excluded if only partial data were available, there were inconsistent clinical findings (such as no epiphora but abnormal dye tests), or previous surgical intervention was performed elsewhere. The procedures were performed by five different surgeons (two oculoplastic and three pediatric ophthalmologists), using the same surgical techniques. Bicanalicular intubation was performed in all intubated nasolacrimal ducts (NLD). Our routine practice was to intervene surgically as warranted when patients were 12 months of age or older. Approval for this study was obtained from King Saud University Institutional Review Board (IRB).
53
54
Al-Faky et al
A retrospective chart review was conducted and data were collected on demographics, age at presentation, age at intervention (to estimate the effect of waiting time because this hospital is a tertiary referral care hospital with surgical waiting lists), sex, onset of the complaint, unilateral or bilateral disease, previous treatment, previous history of acute dacryocystitis, and the presence of Down syndrome. Additional data included symptoms of congenital NLDO (epiphora, increased tear lake, and discharge) in the absence of upper respiratory tract infection or ocular surface irritation, results of the dye disappearance tests (DDT), and Jones primary dye test (dye retrieval test). Intraoperative data included type of surgery, site of the obstruction (divided into lower end NLDO at the level of Hasner’s valve and multiple sites of NLDO with or without canalicular stenosis), severity of the obstruction (classified as simple if probe 1 could easily pass or tight if a smaller probe was used, more force was needed to bypass the obstruction, or there was a difficulty noted on bypassing the obstruction), difficulties associated with tube placement, and inferior turbinate infracture. It also included the presence or absence of a stent fixation suture, which engaged the nasal mucosa with 5-0 polyglactin 910 to prevent early dislodging of the stent during the first 3-4 weeks. Postoperative data included the presence or absence of epiphora, discharge, increased tear lake, results of the DDT and the Jones primary dye test, the time of silicone tube removal (tube was routinely removed after 3 months), displacement of the stent, and the date of the last postoperative visit. The patients were divided into four subgroups based on age: 12-24 months, 25-36 months, 37-48 months, and .48 months. The procedure was considered successful when all preoperative symptoms disappeared with normal DDT and a positive Jones primary dye test. Statistical analysis was performed using SPSS version 16.0 (SPSS Inc, Chicago, IL), Medcalc 8.0 (MedCalc Software, Mariakerke, Belgium), and Statpac 11.0 (StatPac Inc, Bloomington, MN). Descriptive analyses and inferential analysis were performed with the c2 test to evaluate the relationship among variables and success rates. Selected variables for the analysis were those which were statistically significant by univariate analysis, while other factors were excluded. Logistic regression analysis was performed to evaluate the effects of potential risk factors on failure to control for confounding and intervening variables. Statistical significance was indicated by P \ 0.05.
Results A total of 515 NLD procedures were performed. Of these, 61 were excluded due to incomplete data (n 5 57) or inconsistent clinical findings (no epiphora but abnormal dye tests, n 5 4). The remaining 454 cases occurred in 350 patients (162 males [46%]) between 1998 and 2008. The mean age at the time of intervention was 32.6 22.1 months (range, 12-132 months); 2.8% of the subjects were older than 8 years. Mean follow-up was 15.3 19.5 months (range, 3-108 months). The right and left sides were almost equally affected. Unilateral congenital NLDO was present in 246 patients (54.2%); bilateral NLDO was present in 104 patients (208 NLD; 45.8%).
Volume 16 Number 1 / February 2012 The total success rate for both procedures combined was 84.8% (385 of 454 NLD). Probing was performed 188 times, with an 80.3% success rate; intubation was performed 266 times, with an 88% success rate. The difference between success rates was statistically significant (P 5 0.0341). Most probings (134 NLD) were performed on subjects in their second year of life, with an 87.3% success rate (Figure 1). The number of probings, as well as the success rates, deteriorated markedly after 2 years of age until 4 years of age (Figure 1). The decreased success rate was statistically significant (P \ 0.0001). Success rates increased in subjects older than 4 years of age (Figure 1). The success rate increased by 19.5% for subjects .4 years old (nine cases between 49 months and 72 months) compared to subjects who were \4 years old (P 5 0.638). Waiting time had no statistical effect on the outcome of probing (P 5 0.696). Success rates of intubation were high for subjects in their second year of life and gradually declined in subjects between 2 and 4 years of age (Figure 1) (P 5 0.089). There was a statistically insignificant increase in the success rate of intubation in the eldest age group (.48 months) (P 5 0.4483). Waiting time varied between 1 month and 10 months (probing average, 4.33 months; intubation average, 4.34 months). Waiting time had a statistically significant effect on the outcome of intubation: the longer the waiting time, the lower the success rate (P 5 0.001). Success rates of intubation were always higher than probing in all age groups; however, the difference was only statistically significant for subjects between 12 months and 36 months (Figure 1). The potential risk factors for failure of probing or intubation including the effect of demographic data, preoperative clinical features, intraoperative observations, and postoperative data are presented in Table 1. Subjects affected bilaterally had a slightly higher, statistically insignificant incidence of tight obstruction than unilateral cases (20 [9.6%] vs 22 [8.9%]; P 5 0.9251). Sites of obstruction other than at the level of Hasner’s valve were statistically more frequent in subjects affected bilaterally than those affected unilaterally (56 [26.9%] vs 34 [13.8%]; P 5 0.0007). Subjects with previous episodes of acute dacryocystitis had a statistically significantly higher incidence of tight obstruction compared to subjects with no history of dacryocystitis (11 [44%] vs 31 [7.1%]; P \ 0.0001). Obstruction was not as tight in subjects who underwent massage prior to surgical intervention compared to cases that did not (6 [4.7%] vs 36 [11.1%]); this difference approached statistical significance (P 5 0.0532). This observation is clinically important and may have reached significance with a larger sample size. The success rates decreased with both procedures in bilateral cases, without previous massage, and in subjects with Down syndrome or acute dacryocystitis (Figure 2). Other factors, such as sex, severity of symptoms, inferior turbinate infracture, suture fixation, or early tube dislocation, were not statistically significant.
Journal of AAPOS
Volume 16 Number 1 / February 2012
FIG 1. Success rates of probing and silicone intubation in different age groups with their statistical significance (P values) implies whether intubation is better in various age groups (np, total number of probings; nt, total number of intubations).
The effect of adjusted risk factors for failure is presented in Table 1.
Discussion Preoperative and intraoperative risk factors for failure of probing or intubation have been previously reported in a handful of studies6-13; however, some studies have excluded some risk factors from their selection criteria. The Pediatric Eye Disease Investigator Group (PEDIG) investigated similar variables as our study for probing,6 but intubation was not as thoroughly investigated.13 Our success with probing was 80.3%, which is within the range reported by similar studies.2,4,6-9 We also found that the outcome varied based on the age group, which confirms the observations of previous studies.8,10,11 We found age was a significant risk factor for failure of probing (P 5 0.035); however, waiting time for surgical intervention was not significant (P 5 0.746). The total success rate with intubation was 88%, which is similar to previous reports but lower than that of the PEDIG study.12-14 In our study, chronological age was not a risk factor but waiting time for surgery was a significant risk factor for failure of intubation. Studies evaluating intubation in children over 4 years old are meager.12 To our knowledge, this study reports the largest number of cases older than 4 years. Documentation of our experience in this age group is important for future studies and for clinicians seeking guidance for children over 4 years old with congenital NLDO. In our study, bilateral disease was a statistically significant risk factor for both procedures (P \ 0.0001, all cases). This outcome is similar to a number of previous studies of probing but contradicts Lim and colleagues’12 experience
Journal of AAPOS
Al-Faky et al
55
FIG 2. Success rates of probing and silicone intubation with integrated risk factors (np, total number of probings; nt, total number of intubations).
with intubation.6,8,13 We found the success rate was approximately 15% lower with bilateral disease with both procedures (Table 1). A higher incidence of obstructions at sites other than the level of Hasner’s valve among subjects with bilateral disease may explain these results; however, the severity of obstruction did not differ markedly between bilateral and unilateral cases. Two studies reported similar results, yet they assumed significant anatomical and physiological variations that were not confirmed by observational data.6,8 Down syndrome patients had 20% and 28% lower success rates than the total success rate of probing and intubation, respectively. Down syndrome was a highly significant risk factor for intubation using univariate analysis but not significant with multivariate analysis. Lueder15 reported similar results and our results partially confirm the results of Coats and colleagues16; however, Lim and colleagues12 failed to find any impact of Down syndrome on success rates. A history of acute dacryocystitis was a significant risk factor of failure for both procedures (confirmed with univariate and multivariate analysis). There was a 37% greater incidence of tight NLDOs with a positive history of acute dacryocystitis compared to those negative for acute dacryocystitis. This observation supports the suggestion of more fibrosis due to acute inflammation. To our knowledge, there is only one previous report that highlights the impact of acute dacryocystitis in the management of congenital NLDO, but the success rate of probing was not reported.17 Success rates statistically and significantly increased for both probing and intubation in subjects who had previously undergone massage (P 5 0.024 and P 5 0.01, respectively). The outcome is likely due to continuous elimination of noxious material and progressive thinning and dilation of the obstructed site via hydrostatic pressure.
56
Al-Faky et al
Volume 16 Number 1 / February 2012
Table 1. Detailed effect of risk factors for failure on different procedures Probing Characteristic Bilateralism Unilateral Bilateral Down syndrome Yes No Acute dacryocystitis Yes No Previous treatment modalities (massage) Yes No Site of obstruction Level of Hasner valve Other Severity of obstruction Tight Simple
%
Success/total
89.1% 74.9%
82/92 69/96
60% 81.5%
6/10 145/178
0.0% 84.3%
0/9 151/179
89.9% 66.4%
62/69 79/119
83.8% 52.4%
140/167 11/21
28.6% 82.3%
2/7 149/181
Silastic intubation
P value (multivariate)
0.0195 (0.028) 0.209 (0.756) \0.0001 (0.9) 0.0006 (0.024)
0.006 (0.9) 0.0025
%
Success/total
P value (multivariate)
94.1% 79.5%
145/154 89/112
0.0006 (0.003)
61.1% 89.9%
11/18 223/248
0.001 (0.054)
68.8% 89.2%
11/16 223/250
0.042 (0.037)
94.9% 86%
56/59 178/207
0.014 (0.01)
92.4% 75.4%
182/197 52/69
88.6% 87.9%
31/35 203/231
0.0004 (0.012) 0.873
The percentage represents the success rate.
The lower incidence of tight obstructions in subjects who had undergone massage supports our explanation (4.7% vs 11.1%). This outcome contradicts Honavar and colleagues,8 who reported that failed conservative treatment was a risk factor for failure. Based on our outcomes, we recommend parents and caregivers continue to massage until the child undergoes surgery, regardless of age. Prior to our study, only Kushner18 investigated the site of obstruction as a risk factor for probing failure; however, he did not separately report data on severity of the obstruction. We found that cases with an obstruction at the level of Hasner’s valve had significantly higher success rates with both procedures than other sites of obstruction irrespective of the severity (P 5 0.006 for probing and P 5 0.0004 for intubation). Additionally, cases with multiple sites (upper and lower) of congenital NLDO were more likely to fail with either procedure, with a greater tendency of failure with probing. A partial common canalicular obstruction with NLDO dramatically reduced the success rate to 33% with probing, yet had minimal effect on the success of intubation. Severity of the obstruction was a highly significant risk factor for failure with probing (P 5 0.0025) only. Probing succeeded in only one-third of the cases with tight obstruction, which is similar to two previous studies and differs from a PEDIG study.6,11,18 This observation explains the better outcomes of probing in subjects above 4 years old, all of whom did not have tight obstructions. As the presence of bilateral disease, previous massage, Down syndrome, and history of acute dacryocystitis decreased the success rates for both procedures, the clinical application would be of limited value unless we compared each procedure’s outcome to select the more likely procedure to succeed. In this analysis, intubation was always more successful (Figure 2).
Limitations of this study arise from the retrospective nature of this investigation. These limitations hinder accurate comparison and may result in selection bias because our routine practice is to begin with probing as the first line of treatment between 12 and 24 months. We also failed to find a satisfactory explanation for the better outcome of intubation in subjects older than 4 years. Although comparison of various techniques for treatment of congenital NDLO has been extensively reported, our data provide evidence for predictors of failure during clinical evaluation. This enables the clinician to determine the most successful strategy for a particular patient. This study also highlights the value of reducing referral times and decreasing waiting times. Additionally, the outcomes of our study indicate that surgeons should counsel parents of the possibility of changing treatment intraoperatively to insertion of a stent in cases originally selected for probing only. Prior to this investigation, we routinely told parents that massage is of limited value after the first year; however, our results indicate that parents should continually massage until surgery. Last, we are reconsidering the requirement for suture fixation, which seems unnecessary based on the outcomes of this study.
Literature Search A literature search was performed on PubMed and MEDLINE using the following terms: congenital, nasolacrimal duct obstruction, probing, and intubation. References 1. Nesi FA, Lishman RD, Levine MR. Ophthalmic plastic and reconstructive surgery. Congenital lacrimal disorders. 2nd ed. St. Louis, MO: Mosby-Year Book, Inc; 1998:649-60.
Journal of AAPOS
Volume 16 Number 1 / February 2012 2. Casady DR, Meyer DR, Simon JW, Stasior GO, Zobal-Ratner JL. Stepwise treatment paradigm for congenital nasolacrimal duct obstruction. Ophthal Plast Reconstr Surg 2006;22:243-7. 3. MacEwen CJ, Young JD. Epiphora during the first year of life. Eye 1991;5:596-600. 4. Katowitz JA, Welsh MG. Timing of initial probing and irrigation in congenital nasolacrimal duct obstruction. Ophthalmology 1987;94: 698-705. 5. Pediatric Eye Disease Investigator Group. Primary treatment of nasolacrimal duct obstruction with balloon catheter dilation in children younger than 4 years of age. J AAPOS 2008;12:451-5. 6. Pediatric Eye Disease Investigator Group. Primary treatment of nasolacrimal duct obstruction with probing in children younger than 4 years. Ophthalmology 2008;115:577-84. 7. Robb RM. Success rates of nasolacrimal duct probing at time intervals after 1 year of age. Ophthalmology 1998;105:1307-9. 8. Honavar SG, Prakash VE, Rao GN. Outcome of probing for congenital nasolacrimal duct obstruction in older children. Am J Ophthalmol 2000;130:42-8. 9. Maheshwari R. Success rate and cause of failure for late probing for congenital nasolacrimal duct obstruction. J Pediatr Ophthalmol Strabismus 2008;45:168-71. 10. Mannor GE, Rose GE, Frimpong-Ansah K, Ezra E. Factors affecting the success of nasolacrimal duct probing for congenital nasolacrimal duct obstruction. Am J Ophthalmol 1999;127:616-17.
Journal of AAPOS
Al-Faky et al
57
11. Kashkouli MB, Beigi B, Parvaresh MM, Kassaee A, Tabatabaee Z. Late and very late initial probing for congenital nasolacrimal duct obstruction: What is the cause of failure? Br J Ophthalmol 2003;87: 1151-3. 12. Lim CS, Martin F, Beckenham T, Cumming RG. Nasolacrimal duct obstruction in children: Outcome of intubation. J AAPOS 2004;8: 466-72. 13. Pediatric Eye Disease Investigator Group. Primary treatment of nasolacrimal duct obstruction with nasolacrimal duct intubation in children younger than 4 years of age. J AAPOS 2008;12:445-50. 14. Engel JM, Hichie-Schmidt C, Khammar A, Ostfeld BM, Vyas A, Ticho BH. Monocanalicular silastic intubation for the initial correction of congenital nasolacrimal duct obstruction. J AAPOS 2007;11: 183-6. 15. Lueder GT. Treatment of nasolacrimal duct obstruction in children with trisomy 21. J AAPOS 2000;4:230-32. 16. Coats DK, McCreery KM, Plager DA, Bohra L, Kim DS, Paysse EA. Nasolacrimal outflow drainage anomalies in Down’s syndrome. Ophthalmology 2003;110:1437-41. 17. Campolattaro BN, Lueder GT, Tychsen L. Spectrum of pediatric dacryocystitis: Medical and surgical management of 54 cases. J Pediatr Ophthalmol Strabismus 1997;34:143-53. 18. Kushner BJ. The management of nasolacrimal duct obstruction in children between 18 months and 4 years old. J AAPOS 1998;2: 57-60.
To evaluate the success rates of probing and silicone intubation and risk factors for failure of these procedures.
METHODS
This retrospective cohort study included patients who were diagnosed with congenital nasolacrimal duct obstruction (NLDO) and underwent probing or intubation as the primary procedure. Demographic data, clinical features, intraoperative, and postoperative data were evaluated. The patients were divided into four subgroups based on age: 12-24 months, 25-36 months, 37-48 months, and .48 months. The procedure was considered successful when all preoperative signs disappeared with normal dye disappearance test and a positive Jones primary dye test. Statistical analysis was performed to determine the risk factors for failure, and P \ 0.05 was statistically significant.
RESULTS
The cohort comprised 350 subjects (162 males [46%]) who underwent a total of 454 nasolacrimal duct surgeries. The mean chronological age at time of surgery was 32.6 22.1 months (range, 12-132 months). The overall success rate was 84.8%. Probing was performed 188 times, with a success rate of 80.3%; intubation was performed 266 times, with a success rate of 88%.
CONCLUSIONS
Intubation was more successful than probing in patients with bilateral congenital nasolacrimal duct obstruction, Down syndrome, history of acute dacryocystitis, lack of preoperative massage, or intraoperative observation of either a “tight” obstruction or obstruction at sites other than at the level of Hasner’s valve. Primary nasolacrimal duct intubation should be considered in these higher risk patients. ( J AAPOS 2012;16:53-57)
T
he two most common surgical approaches for congenital nasolacrimal duct obstruction (NLDO) are probing and silicone intubation. Many surgeons consider probing the most reliable procedure and perform it as the primary procedure for congenital NLDO; however, success rates with probing have been variable.1-4 Intubation is usually reserved for cases where probing has failed or as a primary procedure in older children and in patients with Down syndrome.1,2,4,5 Success rates of intubation published in medical literature have also been disparate.2 Due to the variable success reported with both procedures, the use of probing or intubation as the primary procedure for congenital NLDO remains controversial. Additionally, there is a relative paucity of studies that Author affiliations: Department of Ophthalmology, College of Medicine, King Saud University, Riyadh, Saudi Arabia Submitted January 17, 2011. Revision accepted July 5, 2011. Correspondence: Yasser H. Al-Faky, MD, FRCS, Ophthalmology Department, College of Medicine, King Saud University, King Abdul Aziz Street, Riyadh, 11411, Saudi Arabia (email: [email protected]). Copyright Ó 2012 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/$36.00 doi:10.1016/j.jaapos.2011.07.020
Journal of AAPOS
have evaluated the effect of the preoperative and intraoperative risk factors on the success of these procedures. We evaluated the pre- and intraoperative risk factors and their effect on the success rates of probing or intubation. These data may assist clinicians in determining the appropriate primary procedure for congenital NLDO.
Patients and Methods This retrospective cohort study comprised subjects diagnosed with congenital NLDO who underwent probing or intubation as the primary procedure. All subjects were diagnosed and treated at the Ophthalmology Department of the College of Medicine, King Saud University, between 1998 and 2008. Subjects were excluded if only partial data were available, there were inconsistent clinical findings (such as no epiphora but abnormal dye tests), or previous surgical intervention was performed elsewhere. The procedures were performed by five different surgeons (two oculoplastic and three pediatric ophthalmologists), using the same surgical techniques. Bicanalicular intubation was performed in all intubated nasolacrimal ducts (NLD). Our routine practice was to intervene surgically as warranted when patients were 12 months of age or older. Approval for this study was obtained from King Saud University Institutional Review Board (IRB).
53
54
Al-Faky et al
A retrospective chart review was conducted and data were collected on demographics, age at presentation, age at intervention (to estimate the effect of waiting time because this hospital is a tertiary referral care hospital with surgical waiting lists), sex, onset of the complaint, unilateral or bilateral disease, previous treatment, previous history of acute dacryocystitis, and the presence of Down syndrome. Additional data included symptoms of congenital NLDO (epiphora, increased tear lake, and discharge) in the absence of upper respiratory tract infection or ocular surface irritation, results of the dye disappearance tests (DDT), and Jones primary dye test (dye retrieval test). Intraoperative data included type of surgery, site of the obstruction (divided into lower end NLDO at the level of Hasner’s valve and multiple sites of NLDO with or without canalicular stenosis), severity of the obstruction (classified as simple if probe 1 could easily pass or tight if a smaller probe was used, more force was needed to bypass the obstruction, or there was a difficulty noted on bypassing the obstruction), difficulties associated with tube placement, and inferior turbinate infracture. It also included the presence or absence of a stent fixation suture, which engaged the nasal mucosa with 5-0 polyglactin 910 to prevent early dislodging of the stent during the first 3-4 weeks. Postoperative data included the presence or absence of epiphora, discharge, increased tear lake, results of the DDT and the Jones primary dye test, the time of silicone tube removal (tube was routinely removed after 3 months), displacement of the stent, and the date of the last postoperative visit. The patients were divided into four subgroups based on age: 12-24 months, 25-36 months, 37-48 months, and .48 months. The procedure was considered successful when all preoperative symptoms disappeared with normal DDT and a positive Jones primary dye test. Statistical analysis was performed using SPSS version 16.0 (SPSS Inc, Chicago, IL), Medcalc 8.0 (MedCalc Software, Mariakerke, Belgium), and Statpac 11.0 (StatPac Inc, Bloomington, MN). Descriptive analyses and inferential analysis were performed with the c2 test to evaluate the relationship among variables and success rates. Selected variables for the analysis were those which were statistically significant by univariate analysis, while other factors were excluded. Logistic regression analysis was performed to evaluate the effects of potential risk factors on failure to control for confounding and intervening variables. Statistical significance was indicated by P \ 0.05.
Results A total of 515 NLD procedures were performed. Of these, 61 were excluded due to incomplete data (n 5 57) or inconsistent clinical findings (no epiphora but abnormal dye tests, n 5 4). The remaining 454 cases occurred in 350 patients (162 males [46%]) between 1998 and 2008. The mean age at the time of intervention was 32.6 22.1 months (range, 12-132 months); 2.8% of the subjects were older than 8 years. Mean follow-up was 15.3 19.5 months (range, 3-108 months). The right and left sides were almost equally affected. Unilateral congenital NLDO was present in 246 patients (54.2%); bilateral NLDO was present in 104 patients (208 NLD; 45.8%).
Volume 16 Number 1 / February 2012 The total success rate for both procedures combined was 84.8% (385 of 454 NLD). Probing was performed 188 times, with an 80.3% success rate; intubation was performed 266 times, with an 88% success rate. The difference between success rates was statistically significant (P 5 0.0341). Most probings (134 NLD) were performed on subjects in their second year of life, with an 87.3% success rate (Figure 1). The number of probings, as well as the success rates, deteriorated markedly after 2 years of age until 4 years of age (Figure 1). The decreased success rate was statistically significant (P \ 0.0001). Success rates increased in subjects older than 4 years of age (Figure 1). The success rate increased by 19.5% for subjects .4 years old (nine cases between 49 months and 72 months) compared to subjects who were \4 years old (P 5 0.638). Waiting time had no statistical effect on the outcome of probing (P 5 0.696). Success rates of intubation were high for subjects in their second year of life and gradually declined in subjects between 2 and 4 years of age (Figure 1) (P 5 0.089). There was a statistically insignificant increase in the success rate of intubation in the eldest age group (.48 months) (P 5 0.4483). Waiting time varied between 1 month and 10 months (probing average, 4.33 months; intubation average, 4.34 months). Waiting time had a statistically significant effect on the outcome of intubation: the longer the waiting time, the lower the success rate (P 5 0.001). Success rates of intubation were always higher than probing in all age groups; however, the difference was only statistically significant for subjects between 12 months and 36 months (Figure 1). The potential risk factors for failure of probing or intubation including the effect of demographic data, preoperative clinical features, intraoperative observations, and postoperative data are presented in Table 1. Subjects affected bilaterally had a slightly higher, statistically insignificant incidence of tight obstruction than unilateral cases (20 [9.6%] vs 22 [8.9%]; P 5 0.9251). Sites of obstruction other than at the level of Hasner’s valve were statistically more frequent in subjects affected bilaterally than those affected unilaterally (56 [26.9%] vs 34 [13.8%]; P 5 0.0007). Subjects with previous episodes of acute dacryocystitis had a statistically significantly higher incidence of tight obstruction compared to subjects with no history of dacryocystitis (11 [44%] vs 31 [7.1%]; P \ 0.0001). Obstruction was not as tight in subjects who underwent massage prior to surgical intervention compared to cases that did not (6 [4.7%] vs 36 [11.1%]); this difference approached statistical significance (P 5 0.0532). This observation is clinically important and may have reached significance with a larger sample size. The success rates decreased with both procedures in bilateral cases, without previous massage, and in subjects with Down syndrome or acute dacryocystitis (Figure 2). Other factors, such as sex, severity of symptoms, inferior turbinate infracture, suture fixation, or early tube dislocation, were not statistically significant.
Journal of AAPOS
Volume 16 Number 1 / February 2012
FIG 1. Success rates of probing and silicone intubation in different age groups with their statistical significance (P values) implies whether intubation is better in various age groups (np, total number of probings; nt, total number of intubations).
The effect of adjusted risk factors for failure is presented in Table 1.
Discussion Preoperative and intraoperative risk factors for failure of probing or intubation have been previously reported in a handful of studies6-13; however, some studies have excluded some risk factors from their selection criteria. The Pediatric Eye Disease Investigator Group (PEDIG) investigated similar variables as our study for probing,6 but intubation was not as thoroughly investigated.13 Our success with probing was 80.3%, which is within the range reported by similar studies.2,4,6-9 We also found that the outcome varied based on the age group, which confirms the observations of previous studies.8,10,11 We found age was a significant risk factor for failure of probing (P 5 0.035); however, waiting time for surgical intervention was not significant (P 5 0.746). The total success rate with intubation was 88%, which is similar to previous reports but lower than that of the PEDIG study.12-14 In our study, chronological age was not a risk factor but waiting time for surgery was a significant risk factor for failure of intubation. Studies evaluating intubation in children over 4 years old are meager.12 To our knowledge, this study reports the largest number of cases older than 4 years. Documentation of our experience in this age group is important for future studies and for clinicians seeking guidance for children over 4 years old with congenital NLDO. In our study, bilateral disease was a statistically significant risk factor for both procedures (P \ 0.0001, all cases). This outcome is similar to a number of previous studies of probing but contradicts Lim and colleagues’12 experience
Journal of AAPOS
Al-Faky et al
55
FIG 2. Success rates of probing and silicone intubation with integrated risk factors (np, total number of probings; nt, total number of intubations).
with intubation.6,8,13 We found the success rate was approximately 15% lower with bilateral disease with both procedures (Table 1). A higher incidence of obstructions at sites other than the level of Hasner’s valve among subjects with bilateral disease may explain these results; however, the severity of obstruction did not differ markedly between bilateral and unilateral cases. Two studies reported similar results, yet they assumed significant anatomical and physiological variations that were not confirmed by observational data.6,8 Down syndrome patients had 20% and 28% lower success rates than the total success rate of probing and intubation, respectively. Down syndrome was a highly significant risk factor for intubation using univariate analysis but not significant with multivariate analysis. Lueder15 reported similar results and our results partially confirm the results of Coats and colleagues16; however, Lim and colleagues12 failed to find any impact of Down syndrome on success rates. A history of acute dacryocystitis was a significant risk factor of failure for both procedures (confirmed with univariate and multivariate analysis). There was a 37% greater incidence of tight NLDOs with a positive history of acute dacryocystitis compared to those negative for acute dacryocystitis. This observation supports the suggestion of more fibrosis due to acute inflammation. To our knowledge, there is only one previous report that highlights the impact of acute dacryocystitis in the management of congenital NLDO, but the success rate of probing was not reported.17 Success rates statistically and significantly increased for both probing and intubation in subjects who had previously undergone massage (P 5 0.024 and P 5 0.01, respectively). The outcome is likely due to continuous elimination of noxious material and progressive thinning and dilation of the obstructed site via hydrostatic pressure.
56
Al-Faky et al
Volume 16 Number 1 / February 2012
Table 1. Detailed effect of risk factors for failure on different procedures Probing Characteristic Bilateralism Unilateral Bilateral Down syndrome Yes No Acute dacryocystitis Yes No Previous treatment modalities (massage) Yes No Site of obstruction Level of Hasner valve Other Severity of obstruction Tight Simple
%
Success/total
89.1% 74.9%
82/92 69/96
60% 81.5%
6/10 145/178
0.0% 84.3%
0/9 151/179
89.9% 66.4%
62/69 79/119
83.8% 52.4%
140/167 11/21
28.6% 82.3%
2/7 149/181
Silastic intubation
P value (multivariate)
0.0195 (0.028) 0.209 (0.756) \0.0001 (0.9) 0.0006 (0.024)
0.006 (0.9) 0.0025
%
Success/total
P value (multivariate)
94.1% 79.5%
145/154 89/112
0.0006 (0.003)
61.1% 89.9%
11/18 223/248
0.001 (0.054)
68.8% 89.2%
11/16 223/250
0.042 (0.037)
94.9% 86%
56/59 178/207
0.014 (0.01)
92.4% 75.4%
182/197 52/69
88.6% 87.9%
31/35 203/231
0.0004 (0.012) 0.873
The percentage represents the success rate.
The lower incidence of tight obstructions in subjects who had undergone massage supports our explanation (4.7% vs 11.1%). This outcome contradicts Honavar and colleagues,8 who reported that failed conservative treatment was a risk factor for failure. Based on our outcomes, we recommend parents and caregivers continue to massage until the child undergoes surgery, regardless of age. Prior to our study, only Kushner18 investigated the site of obstruction as a risk factor for probing failure; however, he did not separately report data on severity of the obstruction. We found that cases with an obstruction at the level of Hasner’s valve had significantly higher success rates with both procedures than other sites of obstruction irrespective of the severity (P 5 0.006 for probing and P 5 0.0004 for intubation). Additionally, cases with multiple sites (upper and lower) of congenital NLDO were more likely to fail with either procedure, with a greater tendency of failure with probing. A partial common canalicular obstruction with NLDO dramatically reduced the success rate to 33% with probing, yet had minimal effect on the success of intubation. Severity of the obstruction was a highly significant risk factor for failure with probing (P 5 0.0025) only. Probing succeeded in only one-third of the cases with tight obstruction, which is similar to two previous studies and differs from a PEDIG study.6,11,18 This observation explains the better outcomes of probing in subjects above 4 years old, all of whom did not have tight obstructions. As the presence of bilateral disease, previous massage, Down syndrome, and history of acute dacryocystitis decreased the success rates for both procedures, the clinical application would be of limited value unless we compared each procedure’s outcome to select the more likely procedure to succeed. In this analysis, intubation was always more successful (Figure 2).
Limitations of this study arise from the retrospective nature of this investigation. These limitations hinder accurate comparison and may result in selection bias because our routine practice is to begin with probing as the first line of treatment between 12 and 24 months. We also failed to find a satisfactory explanation for the better outcome of intubation in subjects older than 4 years. Although comparison of various techniques for treatment of congenital NDLO has been extensively reported, our data provide evidence for predictors of failure during clinical evaluation. This enables the clinician to determine the most successful strategy for a particular patient. This study also highlights the value of reducing referral times and decreasing waiting times. Additionally, the outcomes of our study indicate that surgeons should counsel parents of the possibility of changing treatment intraoperatively to insertion of a stent in cases originally selected for probing only. Prior to this investigation, we routinely told parents that massage is of limited value after the first year; however, our results indicate that parents should continually massage until surgery. Last, we are reconsidering the requirement for suture fixation, which seems unnecessary based on the outcomes of this study.
Literature Search A literature search was performed on PubMed and MEDLINE using the following terms: congenital, nasolacrimal duct obstruction, probing, and intubation. References 1. Nesi FA, Lishman RD, Levine MR. Ophthalmic plastic and reconstructive surgery. Congenital lacrimal disorders. 2nd ed. St. Louis, MO: Mosby-Year Book, Inc; 1998:649-60.
Journal of AAPOS
Volume 16 Number 1 / February 2012 2. Casady DR, Meyer DR, Simon JW, Stasior GO, Zobal-Ratner JL. Stepwise treatment paradigm for congenital nasolacrimal duct obstruction. Ophthal Plast Reconstr Surg 2006;22:243-7. 3. MacEwen CJ, Young JD. Epiphora during the first year of life. Eye 1991;5:596-600. 4. Katowitz JA, Welsh MG. Timing of initial probing and irrigation in congenital nasolacrimal duct obstruction. Ophthalmology 1987;94: 698-705. 5. Pediatric Eye Disease Investigator Group. Primary treatment of nasolacrimal duct obstruction with balloon catheter dilation in children younger than 4 years of age. J AAPOS 2008;12:451-5. 6. Pediatric Eye Disease Investigator Group. Primary treatment of nasolacrimal duct obstruction with probing in children younger than 4 years. Ophthalmology 2008;115:577-84. 7. Robb RM. Success rates of nasolacrimal duct probing at time intervals after 1 year of age. Ophthalmology 1998;105:1307-9. 8. Honavar SG, Prakash VE, Rao GN. Outcome of probing for congenital nasolacrimal duct obstruction in older children. Am J Ophthalmol 2000;130:42-8. 9. Maheshwari R. Success rate and cause of failure for late probing for congenital nasolacrimal duct obstruction. J Pediatr Ophthalmol Strabismus 2008;45:168-71. 10. Mannor GE, Rose GE, Frimpong-Ansah K, Ezra E. Factors affecting the success of nasolacrimal duct probing for congenital nasolacrimal duct obstruction. Am J Ophthalmol 1999;127:616-17.
Journal of AAPOS
Al-Faky et al
57
11. Kashkouli MB, Beigi B, Parvaresh MM, Kassaee A, Tabatabaee Z. Late and very late initial probing for congenital nasolacrimal duct obstruction: What is the cause of failure? Br J Ophthalmol 2003;87: 1151-3. 12. Lim CS, Martin F, Beckenham T, Cumming RG. Nasolacrimal duct obstruction in children: Outcome of intubation. J AAPOS 2004;8: 466-72. 13. Pediatric Eye Disease Investigator Group. Primary treatment of nasolacrimal duct obstruction with nasolacrimal duct intubation in children younger than 4 years of age. J AAPOS 2008;12:445-50. 14. Engel JM, Hichie-Schmidt C, Khammar A, Ostfeld BM, Vyas A, Ticho BH. Monocanalicular silastic intubation for the initial correction of congenital nasolacrimal duct obstruction. J AAPOS 2007;11: 183-6. 15. Lueder GT. Treatment of nasolacrimal duct obstruction in children with trisomy 21. J AAPOS 2000;4:230-32. 16. Coats DK, McCreery KM, Plager DA, Bohra L, Kim DS, Paysse EA. Nasolacrimal outflow drainage anomalies in Down’s syndrome. Ophthalmology 2003;110:1437-41. 17. Campolattaro BN, Lueder GT, Tychsen L. Spectrum of pediatric dacryocystitis: Medical and surgical management of 54 cases. J Pediatr Ophthalmol Strabismus 1997;34:143-53. 18. Kushner BJ. The management of nasolacrimal duct obstruction in children between 18 months and 4 years old. J AAPOS 1998;2: 57-60.