Graefes Arch Clin Exp Ophthalmol DOI 10.1007/s00417-015-2967-5
OCULOPLASTICS AND ORBIT
Association of rhinostomy shape and surgical outcome after endoscopic endonasal dacryocystorhinostomy Joonsik Lee & Sung Won Yang & Hwa Lee & Minwook Chang & Minsoo Park & Sehyun Baek
Received: 5 November 2014 / Revised: 3 February 2015 / Accepted: 5 February 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose The purpose of this study was to investigate surgical outcomes and complications after endoscopic endonasal dacryocystorhinostomy (EDCR) in relation to rhinostomy shape. Methods A retrospective electronic medical record review of patients who underwent EDCR for primary acquired nasolacrimal duct obstruction (PANDO) was performed. Surgical success rates and postoperative complications were compared among three groups of patients in relation to rhinostomy shape (alcove, cavern, or concealed cavern). Results A total of 280 patients (358 eyes) were included in the study. Of the 358 eyes, 194 rhinostomies were alcove-shaped, 157 were cavern-shaped, and 7 were concealed cavernshaped. There were no patients with flat-shaped rhinostomies. The nasal cavity was wider in patients in the alcove group than those in the cavern and concealed cavern groups (p=0.012). The mean time to tube removal was longest in the concealed cavern group (p=0.029). There were no significant differences in anatomical success rates among the three groups (p=0.338). With regard to functional success for patients with anatomically patent DCR, the cavern and concealed cavern groups had significantly poorer results than the alcove group
(p=0.001). Functional success rates were 91.6 %, 84.8 %, and 57.1 % for the alcove, cavern, and concealed cavern groups, respectively. Development of postoperative granuloma was more frequent in the concealed cavern group (85.7 %) than in the alcove (29.3 %) or cavern groups (26.1 %) (p=0.003). Multiple logistic regression models for surgical outcome showed that rates of functional failure after EDCR were influenced by patient age and rhinostomy shape (odds ratio 1.824, p=0.045 for age; odds ratio=9.605, p=0.000 for rhinostomy shape) (Table 5). Conclusions The incidence rate of symptomatic epiphora after EDCR was approximately 12 %, and this result may have been associated with cavernous and concealed rhinostomy shapes. For patients with persistent epiphora and anatomically patent DCR, it is important to identify rhinostomy shape by endoscopy in order to differentiate causes of functional failure. Keywords Rhinostomy shape . Surgical success . Granuloma
Introduction J. Lee : S. W. Yang : H. Lee : S. Baek (*) Department of Ophthalmology, Korea University College of Medicine, 80, Guro-dong, Guro-gu, Seoul, South Korea 152-703 e-mail: [email protected] M. Chang Department of Ophthalmology, Ilsan Hospital, Dongguk University, Goyang, Korea M. Park Department of Ophthalmology, KEPCO Medical Center, Seoul, Korea
The use of an endonasal approach for the treatment of primary acquired nasolacrimal duct obstruction (PANDO) was first described by Caldwell in 1893. Since then, numerous case series in the ophthalmic literature have reported various modifications of the technique and have outlined its advantages and disadvantages [1, 2]. Endoscopic endonasal dacryocystorhinostomy (EDCR) has demonstrated significant advantages, including concurrent correction of intranasal abnormalities that can cause failure due to synechiae formation between the ostium and septum. However, EDCR does involve a number of factors that can result in failure.
Graefes Arch Clin Exp Ophthalmol
Adhesions between the ostium and middle turbinate, synechiae between the ostium and nasal septum, and granuloma formation can cause postoperative osteotomy site obstruction [3, 4]. Postoperative complications have been reported at rates of 0.6 % to 11 % [5, 6]. Numerous modifications have been developed to promote ostial patency, including the creation of nasal mucosal and/or lacrimal sac flaps [7], varying the size of the bony ostium [8], lacrimal sac incision [9], and tube stenting [10]. Recently, some authors have advocated the intraoperative use of mitomycin C or nasal packing to reduce fibrosis formation [11]. All of these techniques focus on large osteotomies to reduce excessive secondary intention healing processes. The prevention of granuloma and synechia formation is a key component in achieving successful surgical results for EDCR. The surgical success rate, defined as anatomical patency, is usually assessed using lacrimal irrigation and fluorescein dye disappearance tests. Functional success is evaluated based on patient symptoms, and several reports have indicated persistent symptoms of epiphora despite an anatomically patent dacryocystorhinostomy [DCR]. Anatomical patency is dependent on postoperative clearance of the rhinostomy site, with direct drainage from the common canaliculus to the nasal cavity resulting from complete removal of the medial lacrimal sac. In this study, rhinostomy shape was classified as flat, alcove, cavern, or concealed cavern, according to textbook descriptions [12]. Some authors have demonstrated a possible relationship between rhinostomy shape and risk of functional failure after external DCR [13]. We investigated the effect of rhinostomy shape on surgical outcome after EDCR and factors associated with surgical complications in relation to rhinostomy shape. We also identified factors that influence rhinostomy shape.
classified the cause of epiphora based on exam results, and we enrolled only patients with primary acquired nasolacrimal duct obstruction and who were over 18 years of age. Exclusion criteria included (1) external DCR, (2) canalicular obstruction, (3) follow-up of less than 3 months, and (4) causes of epiphora other than PANDO. All surgery was performed in the same manner (EDCR) by a single experienced surgeon (one of the authors, S.B.). Surgical technique EDCR was performed as described by Wu et al. [14]. A mixture of 2 ml of 2 % lidocaine and epinephrine was injected into the lateral nasal wall, and Bosmine-soaked gauze was preoperatively packed into the nasal cavity. Under direct visualization using a 0° 4-mm endonasal endoscope (Stryker endoscopic equipment system; Stryker Corporation, Kalamazoo, MI, USA), the lateral nasal mucosa in the area of the lacrimal sac fossa was incised and removed. An osteotomy (8 mm × 10 mm in size) was created using a 45°-angled 2 mm-bite Kerrison punch (CareFusion, San Diego, CA, USA). The lacrimal sac was tented using a probe through the superior canaliculus and incised vertically with a sickle knife to create a large posterior lacrimal sac flap. The medial flap was removed, and the lateral flap was flattened on the lateral nasal mucosal wall. After bicanaliculus intubation (Bika; FCI Ophthalmics, Pembroke, MA, USA) through the upper and lower canaliculus, synthetic polyurethane foam (Nasopore; Polyganics, Rozenburglaan, Groningen, Netherlands) or expandable polyvinyl acetate (Merocel; Medtronic Xomed, Jacksonville, FL, USA) (1 cm × 1 cm × 1 cm) was packed at the anastomosis site. Nasopore and Merocel were then soaked in a gentamycin solution. Postoperative care and follow-up
Materials and methods Patients and selection criteria We performed a retrospective review of the electronic medical records of all patients who underwent EDCR for PANDO at the Korea University Guro Hospital from January 2012 through February 2014. The research protocol adhered to the tenets of the Declaration of Helsinki, and ethics approval was obtained from the institutional review board of Guro Hospital. All patients underwent a comprehensive ophthalmic examination, irrigation of the nasolacrimal drainage system, and intranasal examination. Patient evaluation included slit-lamp examination, dye disappearance test, diagnostic nasolacrimal probing, and irrigation. Other tests included the Schirmer test and tear film break-up time. When the cause was unclear, a lacrimal scan (dacryoscintigraphy) was performed. We
Patients were instructed to apply topical antibiotics (Tarivid solution; Santen Pharmaceutical Co., Ltd, Osaka, Japan) and steroid eye drops (Lotemax ophthalmic solution, 0.5 %; Bausch & Lomb, North Bridgewater, NJ, USA) four times a day. All patients were instructed to use a steroid nasal spray (Nasonex; Schering-Plough Corporation, Kenilworth, NJ, USA). Follow-up visits occurred at 1, 2, 4, 8, and 12 weeks postoperatively. At every visit, nasal wounds were examined by endoscopy and cleaned by suction and forceps to carefully remove blood clots/debris obstructing the ostial site. Granuloma observed at the nasal wound were recorded on the chart, and patients were instructed to use the nasal spray more often. Regular endoscopic nasal examinations were carried out to assess wound healing, with specific reference to the status of the epithelial mucosa and the presence of scars and granulation tissue within 1–2 mm of the ostium, as described by Berlucchi et al. [15] and Xu et al. [16]. Silicone stents were
Graefes Arch Clin Exp Ophthalmol
removed 2 months after the operation. Symptoms of epiphora and purulent discharge were also recorded on the chart using Munk scores. Outcome measurements Postoperative results were evaluated 3 months after surgery. Surgical success or failure was assessed separately for anatomical and functional measures [17]. To assess anatomical success, nasal endoscopy was used to apply gentle lacrimal irrigation without regurgitation. Direct visualization was performed by applying fluorescein dye in the conjunctival fornix through the osteotomy site. Munk scores below 1 were considered to indicate functional success, and scores at or above 2 were considered as indicative of functional failure (Table 1). Rhinostomy shapes were classified as flat, alcove, cavern, or concealed cavern, according to descriptions by Olver [12] (Fig. 1). A flat shape was characterized by a small opening without a clear border between the lacrimal sac and nasal mucosa; an alcove shape described an opening of moderate size without a mark of lacrimal sac; a cavern shape was characterized by a large opening with a hollow made by lacrimal sac flaps; and a concealed cavern shape described a small opening with a laterally concealed lacrimal sac. The size of the ostial opening was determined using a 4-mm-diameter endoscope as a guide and was evaluated according to previous studies [18–20]. A large opening indicated an ostial diameter larger than 3-mm ; a moderate opening, between 1.5 and 3 mm in diameter; and a small opening, an ostial diameter of less than 1.5 mm. Regular endoscopic postoperative follow-up assessed granuloma formation, postoperative bleeding, and signs of infection, and compared rates among the four groups (flat, alcove, cavern, concealed cavern). Functional success was assessed for patients with anatomically patent DCR. Nasal cavity width was recorded by measuring the distance from the nasal septum to the lateral nasal wall during preoperative endoscopy examination. A narrow cavity was defined as one through which a Table 1 Primary outcome measurements of surgical success (anatomical and functional)
Anatomical success Lacrimal syringing Fluorescein dye passage Functional success Assessed by Munk score*
Success
Failure
Passage Visible
Partial or no passage Not visible
0, 1
2, 3, 4, 5
* Munk score is defined as follows: grade 0, no epiphora; grade 1, occasional epiphora requiring dabbing less than twice daily; grade 2, epiphora requiring dabbing 2 to 4 times daily; grade 3, epiphora requiring dabbing 5 to 10 times daily; grade 4, epiphora requiring dabbing more than 10 times daily; and grade 5, constant tearing.
4-mm-diameter endoscope could not pass or could barely pass; a moderate cavity was defined as between two and three times as large as a 4-mm-diameter endoscope; and a wide cavity was defined as at least three times as large as a 4-mm diameter endoscope. Statistical analysis All statistical analysis was performed with SPSS version 20.0 software (SPSS Inc., Chicago, IL, USA), and a p value of less than 0.05 was considered statistically significant. Comparison of surgical results among the four groups in relation to rhinostomy shape was performed using Pearson’s chi-square tests.
Results A total of 280 patients (358 eyes) were included in this study, and all subjects were Korean, with no ethnic variation. Of the 358 eyes, 194 rhinostomies were alcove-shaped, 157 were cavern-shaped, and the remaining 7 had a concealed cavern shape. No patients had a flat-shaped rhinostomy. Patient demographics and clinical data are shown in Table 2. There were no statistically significant differences in sex ratio, preoperative history, or acute dacryocystitis history among the three groups. However, the mean age was significantly higher in the concealed cavern group (60±10.2 years), followed by the cavern (56±13.3 years) and alcove groups (52±13.2 years) (p=0.003 by ANOVA). The cavern group underwent bilateral surgery most often (p=0.022 by Pearson chi-square test). Preoperatively recorded nasal cavity size was wider in the alcove group than in the cavern or concealed cavern groups (p=0.012 by Pearson chi-square test). With respect to nasal packing materials, Nasopore was used more often in the alcove group and Merocel was used more often in the cavern and concealed cavern groups (p=0.022 by Pearson chi-square test). Mean time to tube removal was longest for the concealed cavern group (p=0.029 by ANOVA). With respect to primary outcome measurements, there were no significant differences in anatomical success rates among the three groups (p = 0.338 by Pearson chi-square test) (Table 3). The anatomical success rates were 92.2 %, 88.5 %, and 100 % for the alcove, cavern, and concealed cavern groups, respectively. With regard to functional success rates for patients with anatomically patent DCR, the cavern and concealed cavern groups had significantly poorer results than the alcove group (p=0.001 by Pearson chi-square test). Functional success rates were 91.6 %, 84.8 %, and 57.1 % for the alcove, cavern, and concealed cavern groups, respectively. The association between rhinostomy shape and granuloma formation, excessive scarring around the ostial site, and postoperative bleeding or infection was assessed. Granulomas
Graefes Arch Clin Exp Ophthalmol Fig. 1 Representative cases showing classification of rhinostomy shape after EDCR. (a) alcove-shaped rhinostomy (black arrowhead). (b) cavernshaped rhinostomy (black arrowhead). (c) flat-shaped rhinostomy (black arrowhead)
developed more frequently in the concealed cavern group (85.7 %) than in the alcove (29.3 %) and cavern groups (26.1 %) (p=0.003 using Pearson chi-square test) (Table 4). The secondary outcome measurements of bleeding, synechia, and infection were not identified in the concealed cavern group, and they were not significantly different between the alcove and cavern groups (p=0.103 for bleeding, p=0.796 for synechia, and p=0.284 for infection using Pearson chi-square test). Crust formation and revision surgery were also not significantly different among the three groups (p=0.241 for crust
Table 2
Demographic and clinical data of study subjects
No. of eyes (patients) Age (mean ± SD) Gender (M/F) Eye (OD/OS/OU) Infection (dacryocystitis) Width of nasal cavity‡ (narrow/moderate/wide) Packing (Nasopore/ Merocel) Mean time to tube removal (weeks ) * †
Alcove
Cavern
Concealed p value cavern
194 (159) 52±13.2 26/168 51/73/70 12 19/72/103
157 (115) 56±13.3 31/126 31/42/84 13 27/56/74
7 (6) 60±10.2 0/7 2/3/2 0 3/4/0
0.003* 0.138† 0.022† 0.571† 0.012†
110/84
66/91
4/3
0.022†
8.7±1.4
8.6±1.6
10.2±2.5
0.029*
ANOVA for comparison of age, mean time to tube removal Pearson’s chi-square test was used.
‡
A narrow cavity was defined as one through which a 4-mm-diameter endoscope could not pass or could barely pass; a moderate cavity was defined as between two and three times as large as a 4-mm-diameter endoscope; and a wide cavity was defined as at least three times as large as a 4-mm diameter endoscope. p values < 0.05 were considered statistically significant.
and p=0.512 for revision surgery using Pearson chi-square test). Multiple logistic regression models for surgical outcome showed that patient age and rhinostomy shape affected rates of functional failure after EDCR (odds ratio 1.824, p=0.045 for age; odds ratio=9.605, p=0.000 for rhinostomy shape) (Table 5).
Discussion This study investigated the anatomical and functional surgical success rates for patients with anatomically patent DCR after EDCR in relation to rhinostomy shape. No patient in this study had a flat-shaped rhinostomy. An alcove-shaped rhinostomy was associated with a significantly higher functional, but not anatomical, success rate. All three rhinostomy shapes influenced anatomical success, but the cavern and concealed cavern shapes were associated with poorer functional success results than the alcove shape. These results suggest that rhinostomy shape plays a major role in functional failure after EDCR, even for patients with an anatomically patent DCR. Lee et al. [13] observed a 16 % incidence rate of persistent epiphora among patients with an anatomically patent DCR. The authors also demonstrated a potential relationship between rhinostomy shape and the odds of functional failure after external DCR, reporting a possible association between cavernous-shaped rhinostomy and functional failure. No patients in that study had a rhinostomy with a concealed cavern shape. The healed appearance of rhinostomy varies among patients and is dependent on original rhinostomy size, location, and bone thickness. A flat shape might be associated with a
Graefes Arch Clin Exp Ophthalmol Table 3
Primary outcome measurements (surgical success rate)
Anatomical success (%) (eyes/total) Functional* success (%) (eyes/total) * †
Alcove
Cavern
Concealed cavern
p value†
92.2 % (179/194) 92.2 % (179/194)
88.5 % (139/157) 81.5 % (128/157)
100 % (7/7) 57.1 % (4/7)
0.338 0.001
Functional success rate was assessed for patients with anatomically patent DCR. Pearson’s chi-square test was used.
p values < 0.05 were considered statistically significant.
small osteotomy and sufficient lacrimal sac removal [13]. A cavernous shape is created when the osteotomy is large and the superior part of the lacrimal sac is relatively preserved. Concealed caverns are created when an osteotomy is large and the whole lacrimal sac is not sufficiently removed. An alcove shape results from a large osteotomy and sufficient removal of the lacrimal sac [13]. Based on the abovementioned physiology of rhinostomy shape formation, therefore, cavern and concealed cavern shapes result from insufficient removal of the lacrimal sac, and flat shapes result from insufficient osteotomy around the lacrimal fossa [13]. The mechanism underlying functional failure after anatomically successful DCR is not clearly defined [13]. Rose [21] suggested that retention of the lacrimal sac could cause persistent epiphora after anatomically successful DCR. He divided the lacrimal drainage system into three compartments: the tear lake, lacrimal sac, and nasal space. Elimination of the lacrimal sac to convert the system into two compartments is a crucial component of the symptomatic success of DCR. In the current study, rhinostomy with cavernous and concealed cavern shapes conserved the lacrimal sac, a risk factor for persistent epiphora after EDCR. There were no flat rhinostomies in our study. A flat rhinostomy is caused by a small osteotomy. In the current study, a single experienced surgeon (S.B.), who has performed over 1,000 EDCR surgeries, performed all of the operations. For this reason, stable large osteotomies were possible for all patients enrolled in the study, and this is likely the reason that
Table 4
Secondary outcome measurements (granuloma, bleeding, crust, synechia, infection, revision surgery)
Granuloma (%) (eyes/total) Bleeding (%) (eyes/total) Crust (%) (eyes/total) Synechia (%) (eyes/total) Infection (%) (eyes/total) Revision (%) (eyes/total) *
no patients had a flat-shaped rhinostomy. With regard to patient demographics, the alcove shape was associated with a greater number of open nasal cavities than the other rhinostomy shapes. It is more difficult to handle instruments in a narrow nasal cavity during an osteotomy, increasing the risk of nasal bleeding during the EDCR procedure [22, 23]. Excessive bleeding and mucosal damage can affect the mucosal healing process, causing fibrosis at newly formed rhinostomy sites. This may result in obstruction of ostial patency. Nasopore was associated with an alcove shape and Merocel was associated with a cavern shape. Nasopore is a bioresorbable dressing for the nasal cavity, and is composed of fully synthetic, biodegradable foam that absorbs fluids while supporting and providing pressure to the surrounding tissue. As such, Nasopore can prevent the formation of postoperative adhesions [24, 25]. It can also accelerate wound healing by providing a wet dressing environment. Merocel is a polymer of hydroxylated polyvinyl acetate, a material containing cavities capable of absorbing fluid. However, it is known to cause pain and frequent bleeding upon removal, with an increased risk of adhesion formation [16, 17, 25]. The removal of Merocel and potential bleeding after EDCR may influence rhinostomy shape. The mean patient age was significantly higher in the concealed cavern group, followed by the cavern and the alcove groups. Wound healing ability decreases with age. The results suggest that poor wound healing after EDCR also affects rhinostomy shape.
Alcove
Cavern
Concealed cavern
p value*
29.3 5.67 28.8 3.60 3.09 7.73
26.1 % (41/157) 11.4 % (18/157) 21.6 % (34/157) 4.45 % (7/157) 6.36 % (10/157) 10.1 % (16/157)
85.7 0.00 14.2 0.00 0.00 0.00
0.003 0.103 0.241 0.796 0.284 0.512
% (57/194) % (11/194) % (56/194) % (7/194) % (6/194) % (15/194)
Pearson’s chi-square test was used.
p values < 0.05 were considered statistically significant.
% (6/7) % (0/7) % (1/7) % (0/7) % (0/7) % (0/7)
Graefes Arch Clin Exp Ophthalmol Table 5
Multiple logistic regression models for surgical outcome
Surgical success
OR
Age Anatomical 0.807 Functional 2.924 Gender Anatomical 0.316 Functional 0.545 Mean time to tube removal Anatomical 0.482 Functional 1.046 Nasal packing materials Anatomical 0.906 Functional 1.068 Width of nasal cavity Anatomical 0.906 Functional 1.154 Rhinostomy shape Anatomical 1.626 Functional
9.605
Std err.
p value
95 % CI
0.537 0.417
0.126 0.042
0.832–2.467 2.626–7.588
0.471 0.452
0.925 0.713
0.295–2.785 0.985–3.024
0.317 0.303
0.161 0.105
0.536–1.213 0.897–1.945
0.511 0.495
0.129 0.083
0.532–1.467 1.105–2.413
0.511 0.495
0.096 0.129
0.332–2.467 1.325–3.215
0.303
0.109
2.897–4.945
0.607
0.000
3.925–18.539
OR odds ratio, Std err. standard error, CI confidence interval p values < 0.05 were considered statistically significant.
Mean time for tube removal was shorter in the alcove and cavern groups than the concealed cavern group (8.7 weeks, 8.6 weeks, and 10.2 weeks, respectively). Tube removal should be performed after identifying proper epithelialization of the osteotomy site, and remnants of the lacrimal sac in the concealed cavern group may disturb the mucosal reepithelialization of the osteotomy site. We removed the silicone tube later in the concealed cavern group. Our results suggest that rhinostomy shape and mucosal wound healing interact after EDCR. Granulomas developed more frequently in the concealed cavern group than in the alcove and cavern groups. Rhinostomy shape formation is affected by osteotomy size and lacrimal sac removal, which are factors that influence mucosal wound healing after EDCR. Rhinostomy shape is also affected by the primary wound healing process (e.g., use of nasal packing materials). If complete anatomic obstruction causes DCR failure, a secondary surgical bypass procedure is necessary and may achieve success. However, in cases of failure caused by a nonfunctioning but patent DCR with no obvious punctal or eyelid abnormalities, the situation is less clear. Kim et al. [26] demonstrated that lacrimal silicone intubation was a simple, safe, and effective procedure for patients with epiphora even after anatomically successful DCR. Bicanalicular silicone intubation supports the punctal position to allow effective opposition of the upper and lower puncta during the closure phase of blinking, which enhances the lacrimal pump function.
Delaney and Khooshabeh [27] reported anatomical patency of 92.8 % (39/42 eyes) and symptomatic success of 74.3 % (29/39 eyes) for DCR. The current study showed similar anatomical results (90.7 %, 325/358) and better functional (88.0 %, 286/325) success rates. However, the evaluation criteria for functional failure in these studies were not identical. In conclusion, symptomatic epiphora after anatomically patent DCR was not an uncommon phenomenon, with an incidence rate of about 12 %. A cavernous and concealed-shape rhinostomy may have been associated with this result. Rhinostomy shape formation is dependent on many individual characteristics such as age and general physiologic conditions. Initial osteotomy size and removal of the lacrimal sac are known to be major factors [12, 13], and our study identified patient age, width of the nasal cavity, and nasal packing materials as additional determinants of rhinostomy shape. These factors influence the mucosal wound healing process and epithelialization, which determine surgical success rates (primarily functional) and postoperative complications. For patients with persistent epiphora and anatomically patent DCR, it is important to determine the rhinostomy shape using endoscopy in order to differentiate the causes of functional failure. In the case of functional failure in patients with alcove-shaped rhinostomies, therefore, we recommend attention to other factors that cause epiphora such as lid laxity, dry eye, or conjunctivitis. However, in the case of functional failure with a cavernous or flat-shaped rhinostomy, revision surgery should be considered for additional removal of the lacrimal sac or enlarging the osteotomy site. We believe that our study provides valuable insight for determining the proper treatment approach for patients with functional failure. A limitation of the study is that we cannot conclude whether rhinostomy shape is influenced by the surgeon performing the procedure (large osteotomy, enough removal of lacrimal sac). Further prospective studies may be needed to determine this effect. Acknowledgements Funding/Support The authors state that they received no financial support and have no conflicts of interest. Financial disclosures We have no financial disclosures. We also uploaded the Modified ICMJE Financial Disclosure Form. Author contributions are as follows: study concept and design (J.L., H.L., S.Y., M.C., M.P., S.B.); conduct of the study (J.L.,S.Y.,M.C.,M.P., S.B.); data collection and management (J.L., H.L., S.Y., M.C., S.B.); data analysis (J.L., H.L., S.Y., M.P., S.B.); data interpretation (J.L., H.L., M.C., M.P., S.B.); preparation, review, and approval of the manuscript (J.L., M.C., H.L., M.P., S.B.).
Graefes Arch Clin Exp Ophthalmol The study was approved by the Institutional Review Board of Korea University Guro Hospital.
References 1. Whittet HB, Shun-Shin GA, Awdry P (1993) Functional endoscopic transnasal dacryocystorhinostomy. Eye 7(Pt 4):545–549 2. Bartley GB (1994) The pros and cons of laser dacryocystorhinostomy. Am J Ophthalmol 117:103–106 3. Allen K, Berlin AJ (1989) Dacryocystorhinostomy failure: association with nasolacrimal silicone intubation. Ophthalmic Surg 20:486– 489 4. Rosen N, Sharir M, Moverman DC, Rosner M (1989) Dacryocystorhinostomy with silicone tubes: evaluation of 253 cases. Ophthalmic Surg 20:115–119 5. Fayet B, Racy E, Assouline M (2004) Complications of standardized endonasal dacryocystorhinostomy with unciformectomy. Ophthalmology (Elsevier) 111:837–845 6. Ben Simon GJ, Joseph J, Lee S, Schwarcz RM, McCann JD, Goldberg RA (2005) External versus endoscopic dacryocystorhinostomy for acquired nasolacrimal duct obstruction in a tertiary referral center. Ophthalmology (Elsevier) 112:1463– 1468 7. Tsirbas A, Davis G, Wormald PJ (2004) Mechanical endonasal dacryocystorhinostomy versus external dacryocystorhinostomy. Ophthal Plast Reconstr Surg 20:50–56 8. Sham CL, van Hasselt CA (2000) Endoscopic terminal dacryocystorhinostomy. Laryngoscope (LWW) 110:1045–1049 9. Apaydin KC, Fisenk F, Karayalcin B, Akar Y, Saka O (2004) Endoscopic transnasal dacryocystorhinostomy and bicanalicular silicone tube intubation. Ophthalmologica 218:306–311 10. Tabatabaie SZ, Heirati A, Rajabi MT, Kasaee A (2007) Silicone intubation with intraoperative mitomycin C for nasolacrimal duct obstruction in adults: a prospective, randomized, double-masked study. Ophthalmic Plast Reconstr Surg 23:455–458 11. Selig YK, Biesman BS, Rebeiz EE (2000) Topical application of mitomycin-C in endoscopic dacryocystorhinostomy. Am J Rhinol 14:205–207 12. Olver J (2002) Colour Atlas of Lacrimal Surgery. Butterworth Heinemann, Oxford, p 135 13. Lee MJ, Khwarg SI, Choung HK, Kim N (2014) Associated factors of functional failure of external dacryocystorhinostomy. Can J Ophthalmol 49:40–44
14. Wu W, Yan W, MacCallum JK, Tu Y, Jiang AC, Yang Y, Xiao T, Li J, Wang Q, Qu J (2009) Primary treatment of acute dacryocystitis by endoscopic dacryocystorhinostomy with silicone intubation guided by a soft probe. Ophthalmology (Elsevier) 116:116–122 15. Berlucchi M, Castelnuovo P, Vincenzi A, Morra B, Pasquini E (2009) Endoscopic outcomes of resorbable nasal packing after functional endoscopic sinus surgery: a multicenter prospective randomized controlled study. Eur Arch Oto-Rhino-Laryngol 266:839–845 16. Xu G, Chen HX, Wen WP, Shi JB, Li Y (2003) Clinical evaluation of local application of Merogel after endoscopic sinus surgery. Chin J Otorhinolaryngol 38:95–97 17. Jang SY, Lee KH, Lee SY, Yoon JS (2013) Effects of nasopore packing on dacryocystorhinostomy. Korean J Ophthalmol 27:73–80 18. Yazici B, Yazici Z (2003) Final nasolacrimal ostium after external dacryocystorhinostomy. Arch Ophthalmol 121(1):76–80 19. Zilelioğlu G, Uğurbaş SH, Anadolu Y, Akiner M, Aktürk T (1998) Adjunctive use of mitomycin C on endoscopic lacrimal surgery. Br J Ophthalmol 82(1):63–66 20. Ezra E, Restori M, Mannor GE, Rose GE (1998) Ultrasonic assessment of rhinostomy size following external dacryocystorhinostomy. Br J Ophthalmol 82(7):786–789 21. Rose GE (2004) The lacrimal paradox: toward a greater understanding of success in lacrimal surgery. Ophthalmic Plast Reconstr Surg 20:262–265 22. Anderson RL (1977) Gelfoam packing after dacryocystorhinostomy. Arch Ophthalmol 95:520 23. Durrani OM, Fernando AI, Reuser TQ (2007) Use of a novel topical hemostatic sealant in lacrimal surgery: a prospective, comparative study. Ophthalmic Plast Reconstr Surg 23:25–27 24. Weber R, Keerl R, Hochapfel F, Draf W, Toffel PH (2001) Packing in endonasal surgery. Am J Otolaryngol 22:306–320 25. Wang YP, Wang MC, Chen YC, Leu YS, Lin HC, Lee KS (2011) The effects of Vaseline gauze strip, Merocel, and Nasopore on the formation of synechiae and excessive granulation tissue in the middle meatus and the incidence of major postoperative bleeding after endoscopic sinus surgery. J Chin Med Assoc 74:16–21 26. Kim NJ, Kim JH, Hwang SW, Choung HK, Lee YJ, Khwarg SI (2007) Lacrimal silicone intubation for anatomically successful but functionally failed external dacryocystorhinostomy. Korean J Ophthalmol 21:70–73 27. Delaney YM, Khooshabeh R (2002) Fluorescein transit test time and symptomatic outcomes after external dacryocystorhinostomy. Ophthalmic Plast Reconstr Surg 18:281–284
OCULOPLASTICS AND ORBIT
Association of rhinostomy shape and surgical outcome after endoscopic endonasal dacryocystorhinostomy Joonsik Lee & Sung Won Yang & Hwa Lee & Minwook Chang & Minsoo Park & Sehyun Baek
Received: 5 November 2014 / Revised: 3 February 2015 / Accepted: 5 February 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose The purpose of this study was to investigate surgical outcomes and complications after endoscopic endonasal dacryocystorhinostomy (EDCR) in relation to rhinostomy shape. Methods A retrospective electronic medical record review of patients who underwent EDCR for primary acquired nasolacrimal duct obstruction (PANDO) was performed. Surgical success rates and postoperative complications were compared among three groups of patients in relation to rhinostomy shape (alcove, cavern, or concealed cavern). Results A total of 280 patients (358 eyes) were included in the study. Of the 358 eyes, 194 rhinostomies were alcove-shaped, 157 were cavern-shaped, and 7 were concealed cavernshaped. There were no patients with flat-shaped rhinostomies. The nasal cavity was wider in patients in the alcove group than those in the cavern and concealed cavern groups (p=0.012). The mean time to tube removal was longest in the concealed cavern group (p=0.029). There were no significant differences in anatomical success rates among the three groups (p=0.338). With regard to functional success for patients with anatomically patent DCR, the cavern and concealed cavern groups had significantly poorer results than the alcove group
(p=0.001). Functional success rates were 91.6 %, 84.8 %, and 57.1 % for the alcove, cavern, and concealed cavern groups, respectively. Development of postoperative granuloma was more frequent in the concealed cavern group (85.7 %) than in the alcove (29.3 %) or cavern groups (26.1 %) (p=0.003). Multiple logistic regression models for surgical outcome showed that rates of functional failure after EDCR were influenced by patient age and rhinostomy shape (odds ratio 1.824, p=0.045 for age; odds ratio=9.605, p=0.000 for rhinostomy shape) (Table 5). Conclusions The incidence rate of symptomatic epiphora after EDCR was approximately 12 %, and this result may have been associated with cavernous and concealed rhinostomy shapes. For patients with persistent epiphora and anatomically patent DCR, it is important to identify rhinostomy shape by endoscopy in order to differentiate causes of functional failure. Keywords Rhinostomy shape . Surgical success . Granuloma
Introduction J. Lee : S. W. Yang : H. Lee : S. Baek (*) Department of Ophthalmology, Korea University College of Medicine, 80, Guro-dong, Guro-gu, Seoul, South Korea 152-703 e-mail: [email protected] M. Chang Department of Ophthalmology, Ilsan Hospital, Dongguk University, Goyang, Korea M. Park Department of Ophthalmology, KEPCO Medical Center, Seoul, Korea
The use of an endonasal approach for the treatment of primary acquired nasolacrimal duct obstruction (PANDO) was first described by Caldwell in 1893. Since then, numerous case series in the ophthalmic literature have reported various modifications of the technique and have outlined its advantages and disadvantages [1, 2]. Endoscopic endonasal dacryocystorhinostomy (EDCR) has demonstrated significant advantages, including concurrent correction of intranasal abnormalities that can cause failure due to synechiae formation between the ostium and septum. However, EDCR does involve a number of factors that can result in failure.
Graefes Arch Clin Exp Ophthalmol
Adhesions between the ostium and middle turbinate, synechiae between the ostium and nasal septum, and granuloma formation can cause postoperative osteotomy site obstruction [3, 4]. Postoperative complications have been reported at rates of 0.6 % to 11 % [5, 6]. Numerous modifications have been developed to promote ostial patency, including the creation of nasal mucosal and/or lacrimal sac flaps [7], varying the size of the bony ostium [8], lacrimal sac incision [9], and tube stenting [10]. Recently, some authors have advocated the intraoperative use of mitomycin C or nasal packing to reduce fibrosis formation [11]. All of these techniques focus on large osteotomies to reduce excessive secondary intention healing processes. The prevention of granuloma and synechia formation is a key component in achieving successful surgical results for EDCR. The surgical success rate, defined as anatomical patency, is usually assessed using lacrimal irrigation and fluorescein dye disappearance tests. Functional success is evaluated based on patient symptoms, and several reports have indicated persistent symptoms of epiphora despite an anatomically patent dacryocystorhinostomy [DCR]. Anatomical patency is dependent on postoperative clearance of the rhinostomy site, with direct drainage from the common canaliculus to the nasal cavity resulting from complete removal of the medial lacrimal sac. In this study, rhinostomy shape was classified as flat, alcove, cavern, or concealed cavern, according to textbook descriptions [12]. Some authors have demonstrated a possible relationship between rhinostomy shape and risk of functional failure after external DCR [13]. We investigated the effect of rhinostomy shape on surgical outcome after EDCR and factors associated with surgical complications in relation to rhinostomy shape. We also identified factors that influence rhinostomy shape.
classified the cause of epiphora based on exam results, and we enrolled only patients with primary acquired nasolacrimal duct obstruction and who were over 18 years of age. Exclusion criteria included (1) external DCR, (2) canalicular obstruction, (3) follow-up of less than 3 months, and (4) causes of epiphora other than PANDO. All surgery was performed in the same manner (EDCR) by a single experienced surgeon (one of the authors, S.B.). Surgical technique EDCR was performed as described by Wu et al. [14]. A mixture of 2 ml of 2 % lidocaine and epinephrine was injected into the lateral nasal wall, and Bosmine-soaked gauze was preoperatively packed into the nasal cavity. Under direct visualization using a 0° 4-mm endonasal endoscope (Stryker endoscopic equipment system; Stryker Corporation, Kalamazoo, MI, USA), the lateral nasal mucosa in the area of the lacrimal sac fossa was incised and removed. An osteotomy (8 mm × 10 mm in size) was created using a 45°-angled 2 mm-bite Kerrison punch (CareFusion, San Diego, CA, USA). The lacrimal sac was tented using a probe through the superior canaliculus and incised vertically with a sickle knife to create a large posterior lacrimal sac flap. The medial flap was removed, and the lateral flap was flattened on the lateral nasal mucosal wall. After bicanaliculus intubation (Bika; FCI Ophthalmics, Pembroke, MA, USA) through the upper and lower canaliculus, synthetic polyurethane foam (Nasopore; Polyganics, Rozenburglaan, Groningen, Netherlands) or expandable polyvinyl acetate (Merocel; Medtronic Xomed, Jacksonville, FL, USA) (1 cm × 1 cm × 1 cm) was packed at the anastomosis site. Nasopore and Merocel were then soaked in a gentamycin solution. Postoperative care and follow-up
Materials and methods Patients and selection criteria We performed a retrospective review of the electronic medical records of all patients who underwent EDCR for PANDO at the Korea University Guro Hospital from January 2012 through February 2014. The research protocol adhered to the tenets of the Declaration of Helsinki, and ethics approval was obtained from the institutional review board of Guro Hospital. All patients underwent a comprehensive ophthalmic examination, irrigation of the nasolacrimal drainage system, and intranasal examination. Patient evaluation included slit-lamp examination, dye disappearance test, diagnostic nasolacrimal probing, and irrigation. Other tests included the Schirmer test and tear film break-up time. When the cause was unclear, a lacrimal scan (dacryoscintigraphy) was performed. We
Patients were instructed to apply topical antibiotics (Tarivid solution; Santen Pharmaceutical Co., Ltd, Osaka, Japan) and steroid eye drops (Lotemax ophthalmic solution, 0.5 %; Bausch & Lomb, North Bridgewater, NJ, USA) four times a day. All patients were instructed to use a steroid nasal spray (Nasonex; Schering-Plough Corporation, Kenilworth, NJ, USA). Follow-up visits occurred at 1, 2, 4, 8, and 12 weeks postoperatively. At every visit, nasal wounds were examined by endoscopy and cleaned by suction and forceps to carefully remove blood clots/debris obstructing the ostial site. Granuloma observed at the nasal wound were recorded on the chart, and patients were instructed to use the nasal spray more often. Regular endoscopic nasal examinations were carried out to assess wound healing, with specific reference to the status of the epithelial mucosa and the presence of scars and granulation tissue within 1–2 mm of the ostium, as described by Berlucchi et al. [15] and Xu et al. [16]. Silicone stents were
Graefes Arch Clin Exp Ophthalmol
removed 2 months after the operation. Symptoms of epiphora and purulent discharge were also recorded on the chart using Munk scores. Outcome measurements Postoperative results were evaluated 3 months after surgery. Surgical success or failure was assessed separately for anatomical and functional measures [17]. To assess anatomical success, nasal endoscopy was used to apply gentle lacrimal irrigation without regurgitation. Direct visualization was performed by applying fluorescein dye in the conjunctival fornix through the osteotomy site. Munk scores below 1 were considered to indicate functional success, and scores at or above 2 were considered as indicative of functional failure (Table 1). Rhinostomy shapes were classified as flat, alcove, cavern, or concealed cavern, according to descriptions by Olver [12] (Fig. 1). A flat shape was characterized by a small opening without a clear border between the lacrimal sac and nasal mucosa; an alcove shape described an opening of moderate size without a mark of lacrimal sac; a cavern shape was characterized by a large opening with a hollow made by lacrimal sac flaps; and a concealed cavern shape described a small opening with a laterally concealed lacrimal sac. The size of the ostial opening was determined using a 4-mm-diameter endoscope as a guide and was evaluated according to previous studies [18–20]. A large opening indicated an ostial diameter larger than 3-mm ; a moderate opening, between 1.5 and 3 mm in diameter; and a small opening, an ostial diameter of less than 1.5 mm. Regular endoscopic postoperative follow-up assessed granuloma formation, postoperative bleeding, and signs of infection, and compared rates among the four groups (flat, alcove, cavern, concealed cavern). Functional success was assessed for patients with anatomically patent DCR. Nasal cavity width was recorded by measuring the distance from the nasal septum to the lateral nasal wall during preoperative endoscopy examination. A narrow cavity was defined as one through which a Table 1 Primary outcome measurements of surgical success (anatomical and functional)
Anatomical success Lacrimal syringing Fluorescein dye passage Functional success Assessed by Munk score*
Success
Failure
Passage Visible
Partial or no passage Not visible
0, 1
2, 3, 4, 5
* Munk score is defined as follows: grade 0, no epiphora; grade 1, occasional epiphora requiring dabbing less than twice daily; grade 2, epiphora requiring dabbing 2 to 4 times daily; grade 3, epiphora requiring dabbing 5 to 10 times daily; grade 4, epiphora requiring dabbing more than 10 times daily; and grade 5, constant tearing.
4-mm-diameter endoscope could not pass or could barely pass; a moderate cavity was defined as between two and three times as large as a 4-mm-diameter endoscope; and a wide cavity was defined as at least three times as large as a 4-mm diameter endoscope. Statistical analysis All statistical analysis was performed with SPSS version 20.0 software (SPSS Inc., Chicago, IL, USA), and a p value of less than 0.05 was considered statistically significant. Comparison of surgical results among the four groups in relation to rhinostomy shape was performed using Pearson’s chi-square tests.
Results A total of 280 patients (358 eyes) were included in this study, and all subjects were Korean, with no ethnic variation. Of the 358 eyes, 194 rhinostomies were alcove-shaped, 157 were cavern-shaped, and the remaining 7 had a concealed cavern shape. No patients had a flat-shaped rhinostomy. Patient demographics and clinical data are shown in Table 2. There were no statistically significant differences in sex ratio, preoperative history, or acute dacryocystitis history among the three groups. However, the mean age was significantly higher in the concealed cavern group (60±10.2 years), followed by the cavern (56±13.3 years) and alcove groups (52±13.2 years) (p=0.003 by ANOVA). The cavern group underwent bilateral surgery most often (p=0.022 by Pearson chi-square test). Preoperatively recorded nasal cavity size was wider in the alcove group than in the cavern or concealed cavern groups (p=0.012 by Pearson chi-square test). With respect to nasal packing materials, Nasopore was used more often in the alcove group and Merocel was used more often in the cavern and concealed cavern groups (p=0.022 by Pearson chi-square test). Mean time to tube removal was longest for the concealed cavern group (p=0.029 by ANOVA). With respect to primary outcome measurements, there were no significant differences in anatomical success rates among the three groups (p = 0.338 by Pearson chi-square test) (Table 3). The anatomical success rates were 92.2 %, 88.5 %, and 100 % for the alcove, cavern, and concealed cavern groups, respectively. With regard to functional success rates for patients with anatomically patent DCR, the cavern and concealed cavern groups had significantly poorer results than the alcove group (p=0.001 by Pearson chi-square test). Functional success rates were 91.6 %, 84.8 %, and 57.1 % for the alcove, cavern, and concealed cavern groups, respectively. The association between rhinostomy shape and granuloma formation, excessive scarring around the ostial site, and postoperative bleeding or infection was assessed. Granulomas
Graefes Arch Clin Exp Ophthalmol Fig. 1 Representative cases showing classification of rhinostomy shape after EDCR. (a) alcove-shaped rhinostomy (black arrowhead). (b) cavernshaped rhinostomy (black arrowhead). (c) flat-shaped rhinostomy (black arrowhead)
developed more frequently in the concealed cavern group (85.7 %) than in the alcove (29.3 %) and cavern groups (26.1 %) (p=0.003 using Pearson chi-square test) (Table 4). The secondary outcome measurements of bleeding, synechia, and infection were not identified in the concealed cavern group, and they were not significantly different between the alcove and cavern groups (p=0.103 for bleeding, p=0.796 for synechia, and p=0.284 for infection using Pearson chi-square test). Crust formation and revision surgery were also not significantly different among the three groups (p=0.241 for crust
Table 2
Demographic and clinical data of study subjects
No. of eyes (patients) Age (mean ± SD) Gender (M/F) Eye (OD/OS/OU) Infection (dacryocystitis) Width of nasal cavity‡ (narrow/moderate/wide) Packing (Nasopore/ Merocel) Mean time to tube removal (weeks ) * †
Alcove
Cavern
Concealed p value cavern
194 (159) 52±13.2 26/168 51/73/70 12 19/72/103
157 (115) 56±13.3 31/126 31/42/84 13 27/56/74
7 (6) 60±10.2 0/7 2/3/2 0 3/4/0
0.003* 0.138† 0.022† 0.571† 0.012†
110/84
66/91
4/3
0.022†
8.7±1.4
8.6±1.6
10.2±2.5
0.029*
ANOVA for comparison of age, mean time to tube removal Pearson’s chi-square test was used.
‡
A narrow cavity was defined as one through which a 4-mm-diameter endoscope could not pass or could barely pass; a moderate cavity was defined as between two and three times as large as a 4-mm-diameter endoscope; and a wide cavity was defined as at least three times as large as a 4-mm diameter endoscope. p values < 0.05 were considered statistically significant.
and p=0.512 for revision surgery using Pearson chi-square test). Multiple logistic regression models for surgical outcome showed that patient age and rhinostomy shape affected rates of functional failure after EDCR (odds ratio 1.824, p=0.045 for age; odds ratio=9.605, p=0.000 for rhinostomy shape) (Table 5).
Discussion This study investigated the anatomical and functional surgical success rates for patients with anatomically patent DCR after EDCR in relation to rhinostomy shape. No patient in this study had a flat-shaped rhinostomy. An alcove-shaped rhinostomy was associated with a significantly higher functional, but not anatomical, success rate. All three rhinostomy shapes influenced anatomical success, but the cavern and concealed cavern shapes were associated with poorer functional success results than the alcove shape. These results suggest that rhinostomy shape plays a major role in functional failure after EDCR, even for patients with an anatomically patent DCR. Lee et al. [13] observed a 16 % incidence rate of persistent epiphora among patients with an anatomically patent DCR. The authors also demonstrated a potential relationship between rhinostomy shape and the odds of functional failure after external DCR, reporting a possible association between cavernous-shaped rhinostomy and functional failure. No patients in that study had a rhinostomy with a concealed cavern shape. The healed appearance of rhinostomy varies among patients and is dependent on original rhinostomy size, location, and bone thickness. A flat shape might be associated with a
Graefes Arch Clin Exp Ophthalmol Table 3
Primary outcome measurements (surgical success rate)
Anatomical success (%) (eyes/total) Functional* success (%) (eyes/total) * †
Alcove
Cavern
Concealed cavern
p value†
92.2 % (179/194) 92.2 % (179/194)
88.5 % (139/157) 81.5 % (128/157)
100 % (7/7) 57.1 % (4/7)
0.338 0.001
Functional success rate was assessed for patients with anatomically patent DCR. Pearson’s chi-square test was used.
p values < 0.05 were considered statistically significant.
small osteotomy and sufficient lacrimal sac removal [13]. A cavernous shape is created when the osteotomy is large and the superior part of the lacrimal sac is relatively preserved. Concealed caverns are created when an osteotomy is large and the whole lacrimal sac is not sufficiently removed. An alcove shape results from a large osteotomy and sufficient removal of the lacrimal sac [13]. Based on the abovementioned physiology of rhinostomy shape formation, therefore, cavern and concealed cavern shapes result from insufficient removal of the lacrimal sac, and flat shapes result from insufficient osteotomy around the lacrimal fossa [13]. The mechanism underlying functional failure after anatomically successful DCR is not clearly defined [13]. Rose [21] suggested that retention of the lacrimal sac could cause persistent epiphora after anatomically successful DCR. He divided the lacrimal drainage system into three compartments: the tear lake, lacrimal sac, and nasal space. Elimination of the lacrimal sac to convert the system into two compartments is a crucial component of the symptomatic success of DCR. In the current study, rhinostomy with cavernous and concealed cavern shapes conserved the lacrimal sac, a risk factor for persistent epiphora after EDCR. There were no flat rhinostomies in our study. A flat rhinostomy is caused by a small osteotomy. In the current study, a single experienced surgeon (S.B.), who has performed over 1,000 EDCR surgeries, performed all of the operations. For this reason, stable large osteotomies were possible for all patients enrolled in the study, and this is likely the reason that
Table 4
Secondary outcome measurements (granuloma, bleeding, crust, synechia, infection, revision surgery)
Granuloma (%) (eyes/total) Bleeding (%) (eyes/total) Crust (%) (eyes/total) Synechia (%) (eyes/total) Infection (%) (eyes/total) Revision (%) (eyes/total) *
no patients had a flat-shaped rhinostomy. With regard to patient demographics, the alcove shape was associated with a greater number of open nasal cavities than the other rhinostomy shapes. It is more difficult to handle instruments in a narrow nasal cavity during an osteotomy, increasing the risk of nasal bleeding during the EDCR procedure [22, 23]. Excessive bleeding and mucosal damage can affect the mucosal healing process, causing fibrosis at newly formed rhinostomy sites. This may result in obstruction of ostial patency. Nasopore was associated with an alcove shape and Merocel was associated with a cavern shape. Nasopore is a bioresorbable dressing for the nasal cavity, and is composed of fully synthetic, biodegradable foam that absorbs fluids while supporting and providing pressure to the surrounding tissue. As such, Nasopore can prevent the formation of postoperative adhesions [24, 25]. It can also accelerate wound healing by providing a wet dressing environment. Merocel is a polymer of hydroxylated polyvinyl acetate, a material containing cavities capable of absorbing fluid. However, it is known to cause pain and frequent bleeding upon removal, with an increased risk of adhesion formation [16, 17, 25]. The removal of Merocel and potential bleeding after EDCR may influence rhinostomy shape. The mean patient age was significantly higher in the concealed cavern group, followed by the cavern and the alcove groups. Wound healing ability decreases with age. The results suggest that poor wound healing after EDCR also affects rhinostomy shape.
Alcove
Cavern
Concealed cavern
p value*
29.3 5.67 28.8 3.60 3.09 7.73
26.1 % (41/157) 11.4 % (18/157) 21.6 % (34/157) 4.45 % (7/157) 6.36 % (10/157) 10.1 % (16/157)
85.7 0.00 14.2 0.00 0.00 0.00
0.003 0.103 0.241 0.796 0.284 0.512
% (57/194) % (11/194) % (56/194) % (7/194) % (6/194) % (15/194)
Pearson’s chi-square test was used.
p values < 0.05 were considered statistically significant.
% (6/7) % (0/7) % (1/7) % (0/7) % (0/7) % (0/7)
Graefes Arch Clin Exp Ophthalmol Table 5
Multiple logistic regression models for surgical outcome
Surgical success
OR
Age Anatomical 0.807 Functional 2.924 Gender Anatomical 0.316 Functional 0.545 Mean time to tube removal Anatomical 0.482 Functional 1.046 Nasal packing materials Anatomical 0.906 Functional 1.068 Width of nasal cavity Anatomical 0.906 Functional 1.154 Rhinostomy shape Anatomical 1.626 Functional
9.605
Std err.
p value
95 % CI
0.537 0.417
0.126 0.042
0.832–2.467 2.626–7.588
0.471 0.452
0.925 0.713
0.295–2.785 0.985–3.024
0.317 0.303
0.161 0.105
0.536–1.213 0.897–1.945
0.511 0.495
0.129 0.083
0.532–1.467 1.105–2.413
0.511 0.495
0.096 0.129
0.332–2.467 1.325–3.215
0.303
0.109
2.897–4.945
0.607
0.000
3.925–18.539
OR odds ratio, Std err. standard error, CI confidence interval p values < 0.05 were considered statistically significant.
Mean time for tube removal was shorter in the alcove and cavern groups than the concealed cavern group (8.7 weeks, 8.6 weeks, and 10.2 weeks, respectively). Tube removal should be performed after identifying proper epithelialization of the osteotomy site, and remnants of the lacrimal sac in the concealed cavern group may disturb the mucosal reepithelialization of the osteotomy site. We removed the silicone tube later in the concealed cavern group. Our results suggest that rhinostomy shape and mucosal wound healing interact after EDCR. Granulomas developed more frequently in the concealed cavern group than in the alcove and cavern groups. Rhinostomy shape formation is affected by osteotomy size and lacrimal sac removal, which are factors that influence mucosal wound healing after EDCR. Rhinostomy shape is also affected by the primary wound healing process (e.g., use of nasal packing materials). If complete anatomic obstruction causes DCR failure, a secondary surgical bypass procedure is necessary and may achieve success. However, in cases of failure caused by a nonfunctioning but patent DCR with no obvious punctal or eyelid abnormalities, the situation is less clear. Kim et al. [26] demonstrated that lacrimal silicone intubation was a simple, safe, and effective procedure for patients with epiphora even after anatomically successful DCR. Bicanalicular silicone intubation supports the punctal position to allow effective opposition of the upper and lower puncta during the closure phase of blinking, which enhances the lacrimal pump function.
Delaney and Khooshabeh [27] reported anatomical patency of 92.8 % (39/42 eyes) and symptomatic success of 74.3 % (29/39 eyes) for DCR. The current study showed similar anatomical results (90.7 %, 325/358) and better functional (88.0 %, 286/325) success rates. However, the evaluation criteria for functional failure in these studies were not identical. In conclusion, symptomatic epiphora after anatomically patent DCR was not an uncommon phenomenon, with an incidence rate of about 12 %. A cavernous and concealed-shape rhinostomy may have been associated with this result. Rhinostomy shape formation is dependent on many individual characteristics such as age and general physiologic conditions. Initial osteotomy size and removal of the lacrimal sac are known to be major factors [12, 13], and our study identified patient age, width of the nasal cavity, and nasal packing materials as additional determinants of rhinostomy shape. These factors influence the mucosal wound healing process and epithelialization, which determine surgical success rates (primarily functional) and postoperative complications. For patients with persistent epiphora and anatomically patent DCR, it is important to determine the rhinostomy shape using endoscopy in order to differentiate the causes of functional failure. In the case of functional failure in patients with alcove-shaped rhinostomies, therefore, we recommend attention to other factors that cause epiphora such as lid laxity, dry eye, or conjunctivitis. However, in the case of functional failure with a cavernous or flat-shaped rhinostomy, revision surgery should be considered for additional removal of the lacrimal sac or enlarging the osteotomy site. We believe that our study provides valuable insight for determining the proper treatment approach for patients with functional failure. A limitation of the study is that we cannot conclude whether rhinostomy shape is influenced by the surgeon performing the procedure (large osteotomy, enough removal of lacrimal sac). Further prospective studies may be needed to determine this effect. Acknowledgements Funding/Support The authors state that they received no financial support and have no conflicts of interest. Financial disclosures We have no financial disclosures. We also uploaded the Modified ICMJE Financial Disclosure Form. Author contributions are as follows: study concept and design (J.L., H.L., S.Y., M.C., M.P., S.B.); conduct of the study (J.L.,S.Y.,M.C.,M.P., S.B.); data collection and management (J.L., H.L., S.Y., M.C., S.B.); data analysis (J.L., H.L., S.Y., M.P., S.B.); data interpretation (J.L., H.L., M.C., M.P., S.B.); preparation, review, and approval of the manuscript (J.L., M.C., H.L., M.P., S.B.).
Graefes Arch Clin Exp Ophthalmol The study was approved by the Institutional Review Board of Korea University Guro Hospital.
References 1. Whittet HB, Shun-Shin GA, Awdry P (1993) Functional endoscopic transnasal dacryocystorhinostomy. Eye 7(Pt 4):545–549 2. Bartley GB (1994) The pros and cons of laser dacryocystorhinostomy. Am J Ophthalmol 117:103–106 3. Allen K, Berlin AJ (1989) Dacryocystorhinostomy failure: association with nasolacrimal silicone intubation. Ophthalmic Surg 20:486– 489 4. Rosen N, Sharir M, Moverman DC, Rosner M (1989) Dacryocystorhinostomy with silicone tubes: evaluation of 253 cases. Ophthalmic Surg 20:115–119 5. Fayet B, Racy E, Assouline M (2004) Complications of standardized endonasal dacryocystorhinostomy with unciformectomy. Ophthalmology (Elsevier) 111:837–845 6. Ben Simon GJ, Joseph J, Lee S, Schwarcz RM, McCann JD, Goldberg RA (2005) External versus endoscopic dacryocystorhinostomy for acquired nasolacrimal duct obstruction in a tertiary referral center. Ophthalmology (Elsevier) 112:1463– 1468 7. Tsirbas A, Davis G, Wormald PJ (2004) Mechanical endonasal dacryocystorhinostomy versus external dacryocystorhinostomy. Ophthal Plast Reconstr Surg 20:50–56 8. Sham CL, van Hasselt CA (2000) Endoscopic terminal dacryocystorhinostomy. Laryngoscope (LWW) 110:1045–1049 9. Apaydin KC, Fisenk F, Karayalcin B, Akar Y, Saka O (2004) Endoscopic transnasal dacryocystorhinostomy and bicanalicular silicone tube intubation. Ophthalmologica 218:306–311 10. Tabatabaie SZ, Heirati A, Rajabi MT, Kasaee A (2007) Silicone intubation with intraoperative mitomycin C for nasolacrimal duct obstruction in adults: a prospective, randomized, double-masked study. Ophthalmic Plast Reconstr Surg 23:455–458 11. Selig YK, Biesman BS, Rebeiz EE (2000) Topical application of mitomycin-C in endoscopic dacryocystorhinostomy. Am J Rhinol 14:205–207 12. Olver J (2002) Colour Atlas of Lacrimal Surgery. Butterworth Heinemann, Oxford, p 135 13. Lee MJ, Khwarg SI, Choung HK, Kim N (2014) Associated factors of functional failure of external dacryocystorhinostomy. Can J Ophthalmol 49:40–44
14. Wu W, Yan W, MacCallum JK, Tu Y, Jiang AC, Yang Y, Xiao T, Li J, Wang Q, Qu J (2009) Primary treatment of acute dacryocystitis by endoscopic dacryocystorhinostomy with silicone intubation guided by a soft probe. Ophthalmology (Elsevier) 116:116–122 15. Berlucchi M, Castelnuovo P, Vincenzi A, Morra B, Pasquini E (2009) Endoscopic outcomes of resorbable nasal packing after functional endoscopic sinus surgery: a multicenter prospective randomized controlled study. Eur Arch Oto-Rhino-Laryngol 266:839–845 16. Xu G, Chen HX, Wen WP, Shi JB, Li Y (2003) Clinical evaluation of local application of Merogel after endoscopic sinus surgery. Chin J Otorhinolaryngol 38:95–97 17. Jang SY, Lee KH, Lee SY, Yoon JS (2013) Effects of nasopore packing on dacryocystorhinostomy. Korean J Ophthalmol 27:73–80 18. Yazici B, Yazici Z (2003) Final nasolacrimal ostium after external dacryocystorhinostomy. Arch Ophthalmol 121(1):76–80 19. Zilelioğlu G, Uğurbaş SH, Anadolu Y, Akiner M, Aktürk T (1998) Adjunctive use of mitomycin C on endoscopic lacrimal surgery. Br J Ophthalmol 82(1):63–66 20. Ezra E, Restori M, Mannor GE, Rose GE (1998) Ultrasonic assessment of rhinostomy size following external dacryocystorhinostomy. Br J Ophthalmol 82(7):786–789 21. Rose GE (2004) The lacrimal paradox: toward a greater understanding of success in lacrimal surgery. Ophthalmic Plast Reconstr Surg 20:262–265 22. Anderson RL (1977) Gelfoam packing after dacryocystorhinostomy. Arch Ophthalmol 95:520 23. Durrani OM, Fernando AI, Reuser TQ (2007) Use of a novel topical hemostatic sealant in lacrimal surgery: a prospective, comparative study. Ophthalmic Plast Reconstr Surg 23:25–27 24. Weber R, Keerl R, Hochapfel F, Draf W, Toffel PH (2001) Packing in endonasal surgery. Am J Otolaryngol 22:306–320 25. Wang YP, Wang MC, Chen YC, Leu YS, Lin HC, Lee KS (2011) The effects of Vaseline gauze strip, Merocel, and Nasopore on the formation of synechiae and excessive granulation tissue in the middle meatus and the incidence of major postoperative bleeding after endoscopic sinus surgery. J Chin Med Assoc 74:16–21 26. Kim NJ, Kim JH, Hwang SW, Choung HK, Lee YJ, Khwarg SI (2007) Lacrimal silicone intubation for anatomically successful but functionally failed external dacryocystorhinostomy. Korean J Ophthalmol 21:70–73 27. Delaney YM, Khooshabeh R (2002) Fluorescein transit test time and symptomatic outcomes after external dacryocystorhinostomy. Ophthalmic Plast Reconstr Surg 18:281–284
