Epiphora is defined as an overflow of tears in the presence of normal tear production, produced due to either obstruction in the drainage apparatus, i.e., puncta, canaliculi, sac or nasolacrimal duct or improper tear drainage. In children, the most common cause of epiphora is congenital nasolacrimal duct obstruction (NLDO) occurring in 20%–30% of the newborn. However, only 1%–6% of these children become symptomatic. The other causes are mostly acquired and occur following trauma, surgery, inflammation, etc., leading to NLDO, canalicular obstruction or punctal stenosis and malposition.
Several studies have demonstrated the efficacy of probing in congenital NLDO in children up to 3–4 years of age. Various other studies have evaluated the role dacryocystorhinostomy (DCR) in congenital as well as acquired cases of NLDO. However, none of the studies have taken into consideration the complete range of causes and treatment modalities for pediatric epiphora in children. Therefore, the primary objective of this work was to study the clinical profile and treatment outcome of epiphora in pediatric patients presenting at a tertiary eye care center. The secondary objective was to highlight the differences in demographic and clinical characteristics between the early versus (<3 years) the late-onset group (≥3 years) and to determine the factors predictive of the outcome.
Materials and Methods
A prospective, nonrandomized observational study was conducted on 209 eyes of 167 patients (42 bilateral cases) aged <10 years that presented with true epiphora to our center in the period between July 2011 and July 2013. The Institute Ethical Committee approval was taken, and the study was in accordance with the Declaration of Helsinki. Informed consent was obtained from the parents/guardians. The diagnosis of epiphora was based on the presence of any one of the following – a history of tearing and discharge for >4 weeks, presence of lacus lacrimalis, raised tear meniscus height, presence of discharge, and positive regurgitation test. Patients with hyperlacrimation or pseudoepiphora and those not willing for follow-up were excluded from the study.
Demographic details and clinical history of the patients were taken from the parents or attendants and included the following – age, gender, laterality, socioeconomic status (using modified Kuppuswamy scale), etiology, and previous treatment history. The presenting symptoms and signs, culture and sensitivity reports of conjunctival swabs (in cases where swabs were sent), associated craniofacial and ocular anomalies, treatment given and any other additional investigations such as computed tomography (CT)-dacryocystography (DCG) as advised by the treating clinician were noted. The patients were followed up for a minimum period of 3 months after the treatment to look for the resolution of presenting signs and symptoms.
The analysis was done following compilation of data using SPSS (version 11) (SPSS Inc., Chicago, USA). Descriptive statistics were used for demographic characteristics and the data being presented as percentages, mean, and standard deviation. Chi-square and Fisher's exact test were used to test the difference in the two proportions. P < 0.05 was denoted a statistically significant difference.
A total of 167 patients were studied. The mean age of onset of symptoms was 1.4 ± 2 years (range, 1 month–8 years) and the mean age of presentation was 2.9 ± 2.4 years (range, 2 months–9 years). Ninety-two patients (55%) had an early onset (<3 years of age), and 75 patients (45%) had a late-onset epiphora (≥3 years). The male:female ratio in the early-onset group was 1.4:1, and in the late-onset group, it was 3.2:1, with an overall ratio of 1.9:1. Overall, 125 cases (75%) were unilateral (64% in early-onset group 64% and 88% in late-onset group). Congenital NLDO was the most common cause of pediatric epiphora seen in 76% (126/167) of the cases, followed by trauma (iatrogenic and non-iatrogenic) seen in 18% (30/167). The other causes included – acquired NLDO in 4% (6/167) and punctal causes in 2% (5/167) cases. Table 1 describes the various types of traumatic and punctal cases. The most common cause in the early-onset group was congenital NLDO constituting 99% of the cases, whereas in the late-onset group, both traumatic and acquired causes contributed equally.
Regarding socioeconomic status, 50% of the patients belonged to the upper lower class on assessment of socioeconomic status using modified Kuppuswamy scale. Eighty-eight percent (147/167) cases first presented to an ophthalmologist, and the rest 12% (20) showed first to a general practitioner or a pediatrician, which was followed by a referral to an ophthalmologist. A previous history of treatment was present in 140 eyes (77% in the early-onset cases and 52% in late-onset cases) that included sac massage in 102 cases (73%), a single trial of probing in 30 cases (21%), multiple failed trials of probing in 7 cases (5%), and 1 case (1%) of a failed DCR.
The most common symptom seen in both the groups, early as well as late onset, was discharge (61%), followed by watering (34%). There was one case of congenital fistula associated with NLDO.
Conjunctival swab was sent for culture only in cases with discharge, and culture positivity was seen in 67% cases. The most common bacteria isolated was Streptococcus pneumoniae (51%), followed in frequency by Staphylococcus epidermidis (20%), Staphylococcus aureus (19%), Pseudomonas (5%), and other Gram-negative bacteria (5%). The other Gram-negative bacteria included cases of Acinetobacter, Escherichia, Proteus, and Diphtheroids. There were no fungal isolates.
We also included patients of pediatric epiphora with associated craniofacial and ocular anomalies which were excluded in most other studies. These systemic and ocular anomalies have been enlisted in Table 2.
CT-DCG was done in 14 out of 30 cases of post-traumatic pediatric epiphora where there was dilemma regarding the site of obstruction. The parents were explained about the risk of radiation exposure during the procedure. The most common site of obstruction seen in 13 cases was the sac-NLDO junction as was evident by the accumulation of the dye in the dilated lacrimal sac. In one case, the obstruction was noted at the proximal nasolacrimal duct.
The various modes of treatment advised included sac massage in 84 cases (40%), probing in 74 cases (35%), intubation DCR in 43 cases (21%), and surgeries for punctal malposition in 7 cases (3.5%). No treatment was advised in one case with Type 4 Tessier's cleft as the patient was referred to a maxillofacial surgeon. The punctal surgeries included punctal dilatation, snip procedure, and skin grafting (for cicatricial ectropion and euryblepharon). The type of treatment administered based on the age of presentation has been shown in Fig. 1. The mean duration of follow-up was 5.4 ± 1.3 months.
The overall success rate of treatment was 81%. The success rates of individual treatment modality were –82% for sac massage, 77% for probing, 79% for intubation DCR, and 100% for punctal surgeries. Some cases showed recurrence following an initial resolution of symptoms that included 7 (8.3%) cases of sac massage, 6 (8.1%) cases of probing, and 2 (4.7%) cases of intubation DCR. These cases were considered as treatment failure. Repeat interventions done in these cases were repeat interventions done in these cases were - seven cases of failed sac massage underwent probing, two cases of failed probing underwent repeat probing, four cases of failed probing underwent intubation DCR, and two cases of failed intubation DCR underwent a repeat surgery. The success rate of intubation DCR for NLDO (congenital and acquired) was higher (92%) as compared to traumatic cases (73%). The association of treatment outcome with demographic factors was studied. These included the age, sex, socioeconomic status, etiology, laterality, symptom severity, presence or absence of infection, and previous treatment history. Increasing age was a significant risk factor associated with the failure of sac massage and probing The success rate of sac massage declined significantly after 12 months of age (P = 0.0004) and that of probing declined after the age of 3 years (P = 0.005). A statistically significant association was also seen between treatment outcome and laterality (P = 0.04), presence of infection (P = 0.032), symptom severity (P = 0.027), and previous history of treatment (P = 0.024) [Table 3]. The success rate of probing in the eyes undergoing a first trial of syringing and probing was 90% (40/44), whereas the success rate in the eyes with a previous failed history of syringing and probing was 57% (17/30) (P < 0.05).
Pediatric epiphora is a common problem in clinical ophthalmology. However, with the exception of congenital NLDO, little has been known and studied about its other causes. We performed a prospective evaluation of pediatric patients with true epiphora due to any cause and compared the data between the early versus the late-onset group. Hence, our study gives a good representation of the entire range of causes of pediatric epiphora.
Congenital NLDO constituted the most common cause of epiphora in our study, followed by traumatic NLDO. Ninety-nine percent cases in the early-onset group were due to congenital NLDO, whereas in late-onset group, both congenital and acquired causes were almost equally prevalent. The most common mode of trauma was motor vehicle accident which resulted in medial canthal injuries. These were mostly avulsion injuries typically associated with a triad of telecanthus, epiphora, and ptosis as also reported by Priel et al. The obstruction in traumatic cases can be anywhere in the lacrimal drainage apparatus. Hence, the location of site of obstruction before intervention is of relevance in traumatic cases. CT-DCG is a useful diagnostic tool in clinically challenging cases of traumatic epiphora. It helps to (l) describe the level of the obstruction; (2) evaluate whether the obstruction is complete or incomplete, intrinsic, or extrinsic to the duct; and (3) determine the cause of obstruction. In our study, CT-DCG was done in 14 out of 30 traumatic cases with the most common site of obstruction being sac-NLDO junction.
On comparing the severity of the symptoms in the two groups, it was found that there was greater symptom severity in the late-onset group with 77% cases presenting with discharge, 19% with watering, and 4% cases with medial canthal swelling. This can be explained by the presence of cases with refractory pathology and recalcitrant cases with prolonged duration of inflammmation, leading to greater symptomatology in the late-onset group.
Bacteriology of pediatric epiphora acquires great significance in view of its bearing on the treatment outcome as seen in our study. In our study, Gram-positive bacteria were the major isolate (89%), with S. pneumoniae contributing to 51% of the cases. The results are very similar to those reported by Bareja and Ghose and Al-Faky et al. Brook and Frazier who studied the bacteriological profile of the dacryocystitis patients in the age group 14–81 years, reported S. aureus followed by S. epidermidis as the common causative organisms. The inclusion of causes other than congenital NLDO did not alter our results much from the results of those studies that solely studied the bacteriology of congenital cases of NLDO.
The success rate of probing with or without silastic intubation reported in various studies range from 55% to 96%. A comparatively lower success rate of probing alone (77%) in our study could be due to several reasons. First, the inclusion of cases with a complex form of obstruction like previously failed treatment and syndromic patients with associated craniofacial and ocular anomalies which were excluded in most other studies is the most important reason. In fact, a comparison of the treatment outcome of probing in cases undergoing a first trial of probing with the cases who had a previous history of ≥1 failed probing showed that the success rate was significantly lower in the latter group (90% vs. 57%). Second, additional procedures such as inferior turbinate fractures repair, etc., were not done during probing in our study. Third, a referral bias cannot be excluded as being a tertiary care center; there were obvious referrals of recalcitrant cases.
In our study, all the patients posted for DCR underwent an intubation DCR. Consistent with our results, a lower success rate of intubation DCR in traumatic cases (73%) as compared to the primary cases (92%) has also been reported in other studies and is explained by the distorted anatomy of the medial canthal area and a tendency toward vigorous repair process. Various other studies that reviewed the surgical outcome in purely primary NLDO cases have reported a higher success rate.
We noted a higher treatment success rate in the early-onset group (86% [106/124]) than in the late-onset group (73% [61/84]) with a statistically significant difference (P < 0.0221). A lower success rate in the late-onset group is expected as a significant percentage of cases have a refractory pathology in the late-onset group (post-traumatic NLDO, post-inflammatory NLDO, etc.).
In our study, we also included the syndromic patients and patients with associated craniofacial and ocular anomalies (36 patients) which were excluded in most other studies.
We analyzed the impact of several clinical factors on the outcome and found a statistically significant association between treatment outcome and laterality (P = 0.04), presence of infection (P = 0.032), symptom severity (P = 0.027), and previous treatment history (P = 0.024). Honavar et al. reported the factors predictive of failure of probing such as age >36 months, bilateral affection, failed conservative therapy, failed earlier probing, dilated lacrimal sac, and firm obstruction. Mannor et al. found a significant association between success of probing with age and symptom severity but not with a previous treatment history while Kashkouli et al. found no association with laterality or presence of infection. Repka et al. found an association with laterality and symptom severity.
NLDO is a common condition during the 1st year of life. Most cases resolve spontaneously or after lacrimal sac massage. For those children in whom the blockage does not resolve, probing of the nasolacrimal duct is a widely used surgical treatment. The success of these treatment modalities has been found to decrease with increasing age, but there are no universally accepted age limits. Several studies have found a significant increase in the failure rate in those patients undergoing probing after the age of 12 months, which has led to the recommendation of early probing in children. Other authors, however, have found no correlation of recurrence rate with the age at probing. Table 4 shows the success rate of probing reported in several studies. Although Robb reported a higher success rate of probing even in older children, his definition of success included even those cases with partial resolution of signs and symptoms. Our definition was stricter as we considered only cases of complete resolution of signs and symptoms as successful. In our series, increasing age was found to be an important factor affecting the success rate of the treatment modalities – sac massage and probing. A significant reduction in the success rate of sac massage was noted in children >12 months of age and of probing in children >3 years of age. It is thought that probing could be less successful in older children, perhaps because of prolonged inflammation of the nasolacrimal system resulting in scarring or because of accumulation of more severe obstructions with time as less severe ones clear spontaneously or with conservative management. Resolution of symptoms can also be attributed to spontaneous resolution in children <1 year. Hence, it is not possible to differentiate whether the success in this age group was due to spontaneous resolution or solely due to intervention in the absence of a control group.
There are a number of strengths of our study. First, we utilized prospective data collection. Second, we successfully recruited a large number of patients with a low rate of loss to follow-up. Third, we included the complete range of causes of pediatric epiphora and studied the outcome of various treatment modalities that has not been the case in the majority of studies so far. In addition, to the best of our knowledge, a comparative analysis between early and late-onset groups has not been conducted.
However, there were a few limitations of this study. First, the number of patients enrolled in the group other than NLDO was few that reduced the precision of the estimates of success for these groups. Second, the differentiation of type of obstruction (simple and complex) was not noted at the time of probing in our study. In addition, additional procedures such as inferior turbinate fracture, etc., were not done that could have influenced the outcome. Third, we could not include the objective tests for assessment such as fluorescein dye disappearance test, and all possible management options of pediatric epiphora, such as balloon dacryoplasty, and which is not commonly done at our center.
Congenital NLDO remains the most common cause of pediatric epiphora in the younger age group while traumatic and other secondary causes contribute significantly to the etiology of the older age group. Gram-positive bacteria is the most common infecting organism in pediatric epiphora, with S. pneumoniae being the most common of all. In congenital NLDO, sac massage is an important treatment modality in children up to 1 year of age and probing remains an effective treatment approach in children up to 3 years of age. Increasing age decreases the success rate of sac massage and probing. The factors predictive of treatment failure other than age are laterality, increasing symptom severity, presence of infection, and previous history of treatment.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
1. Piest KL, Katowitz JA. Treatment of congenital nasolacrimal duct obstruction Ophthalmol Clin North Am. 1991;4:201–9
2. Guerry D 3rd, Kendig EL Jr. Congenital impatency of the nasolacrimal duct Arch Ophthal. 1948;39:193–204
3. Duke Elder S System of Ophthalmology Embryology. Part 1. 1963;3 London Henry Kimpton:241–5
4. 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
5. Repka MX, Chandler DL, Beck RW, Crouch ER 3rd, Donahue S, et alPediatric Eye Disease Investigator Group. Primary treatment of nasolacrimal duct obstruction with probing in children younger than 4 years Ophthalmology. 2008;115:577–84.e3
6. 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;127:616–7
7. Honavar SG, Prakash VE, Rao GN. Outcome of probing for congenital nasolacrimal duct obstruction in older children Am J Ophthalmol. 2000;130:42–8
8. Maheshwari R. Results of probing for congenital nasolacrimal duct obstruction in children older than 13 months of age Indian J Ophthalmol. 2005;53:49–51
9. Barnes EA, Abou-Rayyah Y, Rose GE. Pediatric dacryocystorhinostomy for nasolacrimal duct obstruction Ophthalmology. 2001;108:1562–4
10. Marr JE, Drake-Lee A, Willshaw HE. Management of childhood epiphora Br J Ophthalmol. 2005;89:1123–6
11. Nemet AY, Fung A, Martin PA, Benger R, Kourt G, Danks JJ, et al Lacrimal drainage obstruction and dacryocystorhinostomy in children Eye (Lond). 2008;22:918–24
12. Priel A, Leelapatranurak K, Oh SR, Korn BS, Kikkawa DO. Medial canthal degloving injuries: The triad of telecanthus, ptosis, and lacrimal trauma Plast Reconstr Surg. 2011;128:300e–5e
13. Udhay P, Noronha OV, Mohan RE. Helical computed tomographic dacryocystography and its role in the diagnosis and management of lacrimal drainage system blocks and medial canthal masses Indian J Ophthalmol. 2008;56:31–7
14. Bareja U, Ghose S. Clinicobacteriological correlates of congenital dacryocystitis Indian J Ophthalmol. 1990;38:66–9
15. Al-Faky YH, Naeem T, Al-Sobaie N, Al-Huthail R, Al-Odan H, Osman EA, et al Value of microbiology study in congenital nasolacrimal duct obstruction Saudi J Ophthalmol. 2012;26:223–8
16. Brook I, Frazier EH. Aerobic and anaerobic microbiology of dacryocystitis Am J Ophthalmol. 1998;125:552–4
17. Robb RM. Success rates of nasolacrimal duct probing at time intervals after 1 year of age Ophthalmology. 1998;105:1307–9
18. 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
19. Lee DH, Fudemberg SJ, Davitt BV, Cruz OA. Success of simple probing and irrigation in patients with nasolacrimal duct obstruction and otitis media J AAPOS. 2005;9:192–4
20. Hakin KN, Sullivan TJ, Sharma A, Welham RA. Paediatric dacryocystorhinostomy Aust N Z J Ophthalmol. 1994;22:231–5
21. Nowinski TS, Flanagan JC, Mauriello J. Pediatric dacryocystorhinostomy Arch Ophthalmol. 1985;103:1226–8
22. MacEwen CJ, Young JD. Epiphora during the first year of life Eye (Lond). 1991;5:596–600
23. Paul TO, Shepherd R. Congenital nasolacrimal duct obstruction: Natural history and the timing of optimal intervention J Pediatr Ophthalmol Strabismus. 1994;31:362–7
24. Kashkouli MB, Kassaee A, Tabatabaee Z. Initial nasolacrimal duct probing in children under age 5: Cure rate and factors affecting success J AAPOS. 2002;6:360–3
25. Katowitz JA, Welsh MG. Timing of initial probing and irrigation in congenital nasolacrimal duct obstruction Ophthalmology. 1987;94:698–705
26. Zwaan J. Treatment of congenital nasolacrimal duct obstruction before and after the age of 1 year Ophthalmic Surg Lasers. 1997;28:932–6
27. Baker JD. Treatment of congenital nasolacrimal system obstruction J Pediatr Ophthalmol Strabismus. 1985;22:34–6