Acta Oto-Laryngologica. 2015; 135: 162–168
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Endonasal flap suture-dacryocystorhinostomy (eFS-DCR): a new surgical technique for nasolacrimal duct obstruction (NLDO)
HIROHIKO TACHINO1, MICHIRO FUJISAKA1, CHIHARU FUCHIZAWA2, MASAHITO TSUBOTA1, HIROMASA TAKAKURA1, MASAYUKI ISHIDA1, ATSUSHI HAYASHI2 & HIDEO SHOJAKU1 1
Department of Otorhinolaryngology, Head and Neck Surgery and 2Department of Ophthalmology, University of Toyama, Toyama, Japan
Abstract Conclusion: Endonasal flap suture-dacryocystorhinostomy (eFS- DCR) is a new and valuable technique for the treatment of nasolacrimal duct obstruction (NLDO). This technique could improve the success rate of endonasal DCR in comparison to external DCR. Objective: The standard procedure for NLDO is DCR. DCR can be performed via either an external or endonasal approach. External DCR is now regarded as the gold standard in the treatment of NLDO because of its higher success rate. However, we report a new endonasal surgical technique, eFS-DCR, and assess its efficacy by comparison with the standard endonasal DCR. Methods: We prospectively investigated a series of 62 consecutive patients with NLDO undergoing endonasal DCR. There were two surgical intervention groups: group 1 (24 patients, 28 sites) undergoing endonasal DCR without eFS; and group 2 (38 patients, 47 sites) undergoing eFS-DCR. Study end points were success rate, the ostium diameter (OD) of the lacrimal sac after DCR, and the duration of stent placement. We compared the two groups on these three points. Results: The success rate was 78.6% (22 of 28 sites) for group 1 (DCR without eFS) and 97.9% (46 of 47 sites) for group 2 (eFS-DCR). The ostium size in patients in group 2 was larger than that in group 1. The median time for the duration of stent placement was 42.5 days for group 1 and 31.5 days for group 2. There was a statistical difference (p < 0.01) in all these parameters between the two groups.
Keywords: Lacrimal sac, needle holder, knot pusher, Nunchaku-style silicone tubes
Introduction The dacryocystorhinostomy (DCR) operation is over 100 years old. In 1893, Caldwell [1], who was famous for the Caldwell-Luc operation, suggested using an endonasal approach. Then, in 1904, Toti described the external approach [2]. However, it was very difficult to perform DCR via the intranasal approach, so external DCRs were performed by ophthalmologists. In the 1990s, the development of the endoscope and other surgical equipment created a revolution in endonasal sinus surgery. This revolution led surgeons to go back to the endonasal DCR.
Recently, there have been many reports of success using both external and endonasal DCRs. Leong et al. [3] reported a systematic review of outcomes (4921 DCRs) after DCR. They reported that the success rate after external DCR varied between 65 and 100% compared with 84–94% after endonasal DCR. The success rate of laser-assisted DCR varied widely between 47 and 100%. Which is the best DCR? DCR can be performed via either an external or endoscopic approach. External DCR has disadvantages, including cutaneous scarring and potential injury to the medial canthal structure, but it remains the gold standard for patients with nasolacrimal duct obstruction (NLDO).
Correspondence: Michiro Fujisaka, MD PhD, Department of Otorhinolaryngology, Head and Neck Surgery, University of Toyama, 2630 Sugitani, Toyama 9300194, Japan. Fax: +81 764345038. E-mail: [email protected]
(Received 8 August 2014; accepted 4 September 2014) ISSN 0001-6489 print/ISSN 1651-2251 online 2015 Informa Healthcare DOI: 10.3109/00016489.2014.968803
Endonasal flap suture-dacryocystorhinostomy The endonasal approach has the disadvantage of DCR without a flap suture, because of the difficulty of suturing in the nasal cavity. In this study, we present a new endonasal DCR, eFS-DCR. eFS means that the surgeon sutures the lacrimal sac mucosa to the nasal mucosa using the endonasal approach instead of the external approach. We assessed the efficacy of eFS-DCR by comparison with the standard endonasal DCR.
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Material and methods We enrolled 62 consecutive patients with NLDO undergoing endonasal DCR at the University of Toyama Hospital (Figure 1). The patients were assigned to two groups, Group 1 underwent
163
endonasal DCRs without eFS performed from December 2010 to February 2012 (15 months); group 2 underwent eFS-DCR performed from March 2012 to December 2013 (22 months). We chose to divide the patient sample into two groups by treatment date only. By taking the patients in sequence, we were able to have two groups with no bias for age, gender, or site. All patients preoperatively underwent a comprehensive ophthalmic examination including a lacrimal irrigation test by an ophthalmologist, CT imaging with dacryocystography, and an intranasal examination by an otolaryngologist. All patients were diagnosed as having NLDO. Informed consent was obtained from each patient in accordance with the Declaration of Helsinki.
Study enrollment 62 non-randomized, consecutive patients received a diagnosis of NLDO
62 patients (75 sites) underwent endonasa lDCRs performed from December 2010 to December 2013
Group 1
Group 2
24 patients (28 sites) underwent DCRs without eFS performed from December 2010 to February 2012
38 patients (47 sites) underwent eFS-DCRs performed from March 2012 to December 2013
All 28 sites were included in the analysis of success rate and ostium diameter
8 sites were excluded: 6 obstruction (failure) 2 accidental removal
20 sites were included in the analysis of duration of stent placement
All 47 sites were included in the analysis of success rate and ostium diameter
7 sites were excluded: 1 obstruction (failure) 5 accidental removal 1 revision case
40 sites were included in the analysis of duration of stent placement
Figure 1. Flow chart for study enrolment. DCR, dacryocystorhinostomy; eFS, endonasal flap suture; NLDO, nasolacrimal duct obstruction.
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
164
H. Tachino et al.
All patients had the stent tube inserted into the lacrimal passage during the surgery to keep the canalicular patency. Within 5 days after the surgery, they were discharged from our hospital and visited our outpatient clinic every 2 weeks. They accepted the treatment for cleaning around the ostium of the lacrimal sac and were checked for symptoms, and the patency of the lacrimal passage was tested by the irrigation test and the rhinoscope. Timing of removal of the stent tube was determined by the relief of symptoms such as epiphora, a good result from the lacrimal irrigation test, endonasal video findings such as a fully open ostium, and lacrimal drainage from the common canaliculus opening at blinking, no crust, and no coagulation around the ostium. After removal of the stent tube, we checked the patients after at least 1 month to confirm the patency of the lacrimal passage by both their symptoms and the findings of these examinations. When the patency of the lacrimal passage is confirmed, we usually have finished the patient’s follow-up. If we confirm obstruction of the lacrimal passage after surgery, this is the end point for this study. However we continued the patients’ treatment for their epiphora. There were three study end points: (1) success rate, (2) the diameter of the ostium (OD) of the lacrimal sac after DCR, and (3) the duration of stent placement. These were calculated, measured, and compared between the two groups. Success was defined as anatomic patency of the lacrimal passage as well as relief of symptoms such as epiphora. Anatomic patency was assessed by the lacrimal irrigation test and digital rhinoscopy findings. The OD measurement was conducted at the final check during the follow-up. The OD was measured under the rhinoscope. DCR ostium has a horizontal and vertical diameter. The OD was defined as the smaller one and this is usually the horizontal diameter. Janssen et al. [4] reported that the minimum diameter of the bony nasolacrimal canal was 3.5 mm in a control group and 3.0 mm in the patient group. The OD in this study is not outer diameter (the bony lacrimal canal), but inner diameter (mucosal lacrimal canal). The OD (patient group) must be smaller than the diameter of the bony nasolacrimal canal in a control group. Accordingly, we sorted the OD into four groups, large (> 4 mm), medium (2–4 mm), small (< 2 mm), and obstruction (0 mm). In regard to the OD, we calculated the proportion of the large group to the total, and the proportion of the large plus medium groups to the total. The duration of the stent placement was defined as the duration from stenting at DCR to the removal of the stent tube. Failure, accidental removal, and revision cases (group 1, eight sites; group 2, seven sites) were excluded from these stent placement data (Figure 1).
Statistical analysis was conducted using the GraphPad Prism 6 software for Microsoft Windows (GraphPad Software, Inc., California, USA) to assess the two groups. The success rate and the OD were compared using the Fisher’s exact test. The durations of time that the stents were in place were compared using the Kaplan– Meier method and the log-rank test. In regard to the characteristics of the groups, patients’ ages were compared using the Mann–Whitney U test. The proportion of female sex to the total and the proportion of bilateral affected sides to the total were compared using the Yates 2 2 chisquared test. Surgical technique The operations were performed with the patient under general anesthesia. Standard endonasal DCRs without eFS were performed following the Wormald [5] technique as described previously. We describe our eFS-DCR technique in detail (Figures 2 and 3). A scalpel is used to cut a mucosal flap starting approximately 10 mm above the insertion of the middle turbinate on the lateral nasal wall. Then, two vertical inferior incisions are made. Eventually, these incisions make an inverted U-shaped mucosal flap (Figure 2a). This flap is tucked into the space between the inferior turbinate and the nasal septum. This allows further dissection and manipulation without obstruction of the surgical equipment. The bone, which obscures the lacrimal sac, is removed with a curved diamond DCR burr. After exposure of the entire lacrimal sac, it is incised in an I-shape (Figure 2b). The lacrimal sac membrane is divided into the anterior and the posterior flaps like double doors. When these flaps are turned backward and forward as doors open, the common canaliculus opening is fully visible (Figure 2c, d). After making the lacrimal flaps, they are sutured to the nasal mucosal membrane. At first, the posterior lacrimal flap is sutured to the nasal membrane, because this stitch is the easiest of all (Figure 3a). Two stitches are needed for the posterior flap (Figure 2c, d). Next, the anterior lacrimal flap is sutured to the nasal membrane (Figure 3c). This stitch is the hardest of all, because it is difficult to precisely insert a needle into the lateral nasal wall membrane because of the narrow working space. This is the primary reason that suturing in the nasal cavity seems to be impossible. The nasal cavity sometimes does not have sufficient working space to suture. In these cases, the inverted U-shaped nasal flap, which was made at the beginning of the surgery, is used as the target for suturing the
Endonasal flap suture-dacryocystorhinostomy
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
a
c
b
a
b
c
d
165
d
Figure 2. Schema of the surgical technique for endonasal flap suture-dacryocystorhinostomy (eFS-DCR). (a) Making the inverted U-shaped nasal flap. (b) Incision of the lacrimal sac in an I-shape (laterally H-shape). The lacrimal sac membrane is divided into two (anterior and posterior) flaps like double doors. (c) If the nasal cavity does not have sufficient working space to suture, the inverted U-shaped nasal flap is used as the target for suturing the anterior lacrimal flap. When the anterior lacrimal flap is sutured to the inverted U-shaped nasal flap, this nasal flap naturally goes back to the original position. (d) If the nasal cavity has sufficient working space to suture, we can suture the anterior lacrimal flap to the nasal membrane. Then the inverted U-shaped nasal flap is cut to the appropriate length to cover the bone wall below the lacrimal sac.
Figure 3. Endoscopic findings for the endonasal flap suturedacryocystorhinostomy (eFS-DCR) procedures. (a) It is sutured from the nasal flap to the posterior lacrimal flap. The blue suture is Prolene (Ethicon). Black star ($) indicates the middle turbinate in the right nasal cavity. Circle (*) indicates the inside of the lacrimal sac. Black triangle (~) indicates the exposed bone of the lateral nasal wall covered with bleeding. (b) Ligation by knot pusher. (c) Insertion of a needle into the anterior lacrimal flap. (d) After suture of the anterior lacrimal flap to the inverted U-shaped nasal flap: black square (&). This nasal flap goes back to the original position by suture.
these narrow working spaces. A knot pusher (Hope Denshi Co., Ltd., Chiba, Japan) is useful for suturing in deep spaces (Figure 3b). After suturing, a pair of Nunchaku-style silicone tubes (Lacrifast, Kaneka Medix Corp., Osaka, Japan) are inserted into the lacrimal passage through the common canaliculus opening.
Results anterior lacrimal flap (Figure 2c). It is easier to insert a needle into the free nasal flap than the fixed anterior nasal wall membrane. When the anterior lacrimal flap is sutured to the inverted U-shaped nasal flap, this nasal flap naturally goes back to the original position (Figure 2c). This flap could be trimmed away if necessary. If the nasal cavity has sufficient working space to suture, we can continue to suture the anterior lacrimal flap to the nasal membrane (Figure 2d). Then the inverted U-shaped nasal flap is cut to the appropriate length to cover the bone wall below the lacrimal sac. In addition, specialized surgical equipment makes it possible to do the most difficult stitch in the narrow nasal cavity. The needle holder (Aesculap, Inc., Pennsylvania, USA) helps to manipulate a needle easily in
In regard to the characteristics of the groups, there were no significant differences in patients’ ages, sex distribution, or affected side among the groups (Table I). The success rate was 78.6% (22 of 28 sites) for group 1 (DCR without eFS) and 97.9% (46 of 47 sites) for group 2 (eFS-DCR) (Table II). There was a statistical difference between the two groups (p = 0.00894). In the present study, there was a much higher success rate for eFS-DCR than for endonasal DCR without eFS. The proportion of the large group of the OD to the total number was 10.7% in group 1 and 85.1% in group 2 (Table II). There was a statistical difference between the two groups (p < 0.01). The proportion of the large plus medium groups of the OD to the total
166
H. Tachino et al.
Table I. Demographic data for the two groups. Variable
Group 1 (DCR without eFS)
Group 2 (eFS-DCR)
n = 24 (28)
n = 38 (47)
72.9 ± 11.0 (53–91)
69.4 ± 11.5 (42–88)
0.281
17 (75%)
29 (70%)
0.861
Adult (sides) Age (years), mean ± SD (range) Female sex, n (%)
P value
Side, n (%) 12 (50%)
13 (34%)
Left
8 (33%)
16 (42%)
Bilateral
4 (17%)
9 (24%)
135.1
113.4
Follow-up period (days) 1
2
0.731
3
Success rate, OD 1/total, and OD 1+2/total. DCR, dacryocystorhinostomy; eFS, endonasal flap suture. (i) Mann-Whitney U test (two-tailed); (ii) Yates 2 2 chi-squared test.
Table II. Main outcome measures between the two groups. Outcome
Group 1 (DCR without eFS)
Group 2 (eFS-DCR)
22/28 (78.6%)
46/47 (97.9%)
1, large (> 4 mm)
3
40
2, medium (2–4 mm)
4
5
Success rate
p value 0.0089
OD variable
3, small (< 2 mm) 4, obstruction Total OD 1/total OD 1 + 2/total
15
1
6
1
28
47
3/28 (10.7%)
40/47 (85.1%)
1.34 10–10
7/28 (25%)
45/47 (95.7%)
1.09 10–10
DCR, dacryocystorhinostomy; eFS, endonasal flap suture. p values were calculated by Fisher’s exact test.
number was 25% in group 1 and 95.7% in group 2. There was a statistical difference between the two groups (p < 0.01). This means that the ostium size of patients in group 2 (eFS-DCR) was significantly larger than that of patients in group 1 (DCR without eFS). The duration of stent placement in the two groups is shown in Figure 4. The median time of this duration was 42.5 days in group 1 and 31.5 days in group 2. The log-rank-based assessment at 50% of the patients showed a statistical difference between the two groups (p = 0.0003). The hazard ratio (group 2/ group 1) was 2.127 (95% confidence interval (CI) = 1.796–4.820). The duration of stent placement for patients in group 2 was shorter than that for patients in group 1. This means that the patients that underwent eFS-DCR (group 2) had the stent tube removed sooner than the patients that underwent DCR without eFS (group 1).
Duration of stent placement 100 Patients with stent placement (%)
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Right
p = 0.0003 by log-rank test 50 Group 2 eFS-DCR
Group 1 DCR without eFS
0 0
20
40 60 Postoperative days
80
100
Figure 4. Kaplan–Meier curves for the duration of stent placement between the two groups. The dotted line indicates group 1 (dacryocystorhinostomy, DCR, without endonasal flap suture, eFS) and the solid line indicates group 2 (eFS-DCR). According to the Kaplan–Meier method and log-rank test, the duration of stent placement for patients in group 2 (eFS-DCR) was shorter than that for patients in group 1 (DCR without eFS).
Endonasal flap suture-dacryocystorhinostomy
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Discussion Acquired NLDO is one of the most common problems in the lacrimal drainage system. The majority of acquired NLDO is idiopathic in nature. Other causes of NLDO include acute or chronic inflammation, trauma, facial palsy, drugs, and complications of nasal surgery. Patients generally present with epiphora, eyelid and lacrimal sac swelling, purulent secretion, blurred vision, and facial pain. External DCR is now regarded as the gold standard in the treatment of NLDO. The reported success rates of this approach have varied but are often reported as over 90% in many specialist lacrimal units. However, the procedure leaves a cutaneous scar on the patient’s face and the potential for injury to medial canthal structures, cerebrospinal fluid rhinorrhea, and functional interference with the physiological action of the lacrimal pump. Vagefi et al. [6] reported that among 215 patients and 247 surgeries, 16 individuals (7.4%) were identified who demonstrated abnormalities of eyelid closure in the postoperative period after external DCR. The advantages of endonasal DCR include the absence of cutaneous scarring, maintenance of the pumping mechanism of the orbicularis oculi muscle, and less disruption of the medial canthal anatomy. However, the success rates for endonasal DCR have varied. The disadvantage of endonasal DCR is that it is very difficult to suture the lacrimal membrane to the nasal membrane. Sometimes the lacrimal ostium has closed after endonasal DCR. This is one of the reasons for reduced success rates for endonasal DCR compared with external DCR. Here, we have presented a new endonasal pocedure, eFS-DCR. Our procedure could resolve the disadvantage of endonasal DCR. First, eFS-DCR had a much higher success rate than endonasal DCR without eFS in our study. Second, the ostium size following eFS-DCR was significantly larger than that after endonasal DCR without eFS. Chan and Selva [7] reported that after endoscopic DCR, the final ostium size on average is 35% of the original at 12 months postoperatively. In the present study, the ostium size after eFS-DCR might shrink less because of the suturing. This result directly relates to the higher success rate for eFS-DCR. Third, patients who underwent eFS-DCR were able to have the stent tube removed more quickly than the patients who underwent DCR without eFS. In regard to the wound healing around the lacrimal ostium, it was more rapid after eFS-DCR than after DCR without eFS. Callejas et al. [8] recommended that stents only be inserted when a tight common canaliculus opening is found during surgery. Cannon et al. [9] reported
167
that there were no cases of canalicular closure or stenosis at 12 months in their prospective series of no intubation for primary acquired nasolacrimal duct obstruction (PANDO). They advocated that routine intubation for the purpose of maintaining canalicular patency is not necessary when performing endonasal DCR in PANDO. However, when the lacrimal ostium shrinks almost to the point of closure, insertion of a stent tube will guarantee prevention of obstruction of the ostium. If the size of the lacrimal ostium is large enough in cases such as those undergoing eFS-DCRs, stent intubation might be unnecessary and it would be possible to reduce medical costs. In fact, in five cases that underwent eFS-DCR in which there was an accidental removal of the tube, there was ultimately success in the patency of the lacrimal passage in this study. eFS-DCR has one disadvantage: extra time must be allowed for suturing. It will take about 15 min to suture the lacrimal flap to the nasal flap in the nasal cavity. However, this extra time will be reduced as surgeons gain skill in the procedure for eFS, or by the development of new special surgical tools for eFS. If the eFS technique for endonasal DCR spreads throughout the world and the success rate of endonasal DCR increases to be comparable to external DCR, many patients, especially females with NLDO, who refused the external DCR approach because it leaves a cutaneous scar, would reverse their decision and accept this surgery. Wormald [10] stated that cooperation between the otolaryngologist and the ophthalmologist might be advantageous for the best possible management of the patient with epiphora. When both the ophthalmologist and the otolaryngologist participate in endonasal DCRs, this cooperative work will lead to successful results in endonasal DCRs.
Acknowledgement I would like to thank to Mrs. Carol Macbain for helpful discussion. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Caldwell GW. Two new operations for obstruction of the nasal duct, with preservation of the canaliculi, and with an incidental description of a new lacrymal probe. Am J Ophthalmol 1893;10:189–93.
168
H. Tachino et al.
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
[2] Toti A. Nuovo metodo conservatore dicura radicale delle suppurazione croniche del sacco lacrimale (dacricistorhinostomia). Clin Moderna (Firenza) 1904;10:385. [3] Leong SC, Macewen CJ, White PS. A systematic review of outcomes after dacryocystorhinostomy in adults. Am J Rhinol Allergy 2010;24:81–90. [4] Janssen AG, Mansour K, Bos JJ, Castelijns JA. Diameter of the bony lacrimal canal: normal values and values related to nasolacrimal duct obstruction: assessment with CT. AJNR Am J Neuroradiol 2001;22:845–50. [5] Wormald PJ. Powered endoscopic dacryocystorhinostomy. Laryngoscope 2002;112:69–72. [6] Vagefi MR, Winn BJ, Lin CC, Sires BS, LauKaitis SJ, Anderson RL, et al. Facial nerve injury during
[7] [8]
[9]
[10]
external dacryocystorhinostomy. Ophthalmology 2009;116: 585–90. Chan WO, Selva D. Ostium shrinkage after endoscopic dacryocystorhinostomy. Ophthalmology 2013;120:1693–6. Callejas CA, Tewfik MA, Wormald PJ. Powered endoscopic dacryocystorhinostomy with selective stenting. Laryngoscope 2010;120:1449–52. Cannon PS, Chan WO, Selva D. Incidence of canalicular closure with endonasal dacryocystorhinostomy without intubation in primary nasolacrimal duct obstruction. Ophthalmology 2013;120:1688–92. Wormald PJ. Endoscopic and external dacryocystorhinostomy (DCR) – which is better? Braz J Otorhinolaryngol 2012; 78:2.
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Endonasal flap suture-dacryocystorhinostomy (eFS-DCR): a new surgical technique for nasolacrimal duct obstruction (NLDO)
HIROHIKO TACHINO1, MICHIRO FUJISAKA1, CHIHARU FUCHIZAWA2, MASAHITO TSUBOTA1, HIROMASA TAKAKURA1, MASAYUKI ISHIDA1, ATSUSHI HAYASHI2 & HIDEO SHOJAKU1 1
Department of Otorhinolaryngology, Head and Neck Surgery and 2Department of Ophthalmology, University of Toyama, Toyama, Japan
Abstract Conclusion: Endonasal flap suture-dacryocystorhinostomy (eFS- DCR) is a new and valuable technique for the treatment of nasolacrimal duct obstruction (NLDO). This technique could improve the success rate of endonasal DCR in comparison to external DCR. Objective: The standard procedure for NLDO is DCR. DCR can be performed via either an external or endonasal approach. External DCR is now regarded as the gold standard in the treatment of NLDO because of its higher success rate. However, we report a new endonasal surgical technique, eFS-DCR, and assess its efficacy by comparison with the standard endonasal DCR. Methods: We prospectively investigated a series of 62 consecutive patients with NLDO undergoing endonasal DCR. There were two surgical intervention groups: group 1 (24 patients, 28 sites) undergoing endonasal DCR without eFS; and group 2 (38 patients, 47 sites) undergoing eFS-DCR. Study end points were success rate, the ostium diameter (OD) of the lacrimal sac after DCR, and the duration of stent placement. We compared the two groups on these three points. Results: The success rate was 78.6% (22 of 28 sites) for group 1 (DCR without eFS) and 97.9% (46 of 47 sites) for group 2 (eFS-DCR). The ostium size in patients in group 2 was larger than that in group 1. The median time for the duration of stent placement was 42.5 days for group 1 and 31.5 days for group 2. There was a statistical difference (p < 0.01) in all these parameters between the two groups.
Keywords: Lacrimal sac, needle holder, knot pusher, Nunchaku-style silicone tubes
Introduction The dacryocystorhinostomy (DCR) operation is over 100 years old. In 1893, Caldwell [1], who was famous for the Caldwell-Luc operation, suggested using an endonasal approach. Then, in 1904, Toti described the external approach [2]. However, it was very difficult to perform DCR via the intranasal approach, so external DCRs were performed by ophthalmologists. In the 1990s, the development of the endoscope and other surgical equipment created a revolution in endonasal sinus surgery. This revolution led surgeons to go back to the endonasal DCR.
Recently, there have been many reports of success using both external and endonasal DCRs. Leong et al. [3] reported a systematic review of outcomes (4921 DCRs) after DCR. They reported that the success rate after external DCR varied between 65 and 100% compared with 84–94% after endonasal DCR. The success rate of laser-assisted DCR varied widely between 47 and 100%. Which is the best DCR? DCR can be performed via either an external or endoscopic approach. External DCR has disadvantages, including cutaneous scarring and potential injury to the medial canthal structure, but it remains the gold standard for patients with nasolacrimal duct obstruction (NLDO).
Correspondence: Michiro Fujisaka, MD PhD, Department of Otorhinolaryngology, Head and Neck Surgery, University of Toyama, 2630 Sugitani, Toyama 9300194, Japan. Fax: +81 764345038. E-mail: [email protected]
(Received 8 August 2014; accepted 4 September 2014) ISSN 0001-6489 print/ISSN 1651-2251 online 2015 Informa Healthcare DOI: 10.3109/00016489.2014.968803
Endonasal flap suture-dacryocystorhinostomy The endonasal approach has the disadvantage of DCR without a flap suture, because of the difficulty of suturing in the nasal cavity. In this study, we present a new endonasal DCR, eFS-DCR. eFS means that the surgeon sutures the lacrimal sac mucosa to the nasal mucosa using the endonasal approach instead of the external approach. We assessed the efficacy of eFS-DCR by comparison with the standard endonasal DCR.
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Material and methods We enrolled 62 consecutive patients with NLDO undergoing endonasal DCR at the University of Toyama Hospital (Figure 1). The patients were assigned to two groups, Group 1 underwent
163
endonasal DCRs without eFS performed from December 2010 to February 2012 (15 months); group 2 underwent eFS-DCR performed from March 2012 to December 2013 (22 months). We chose to divide the patient sample into two groups by treatment date only. By taking the patients in sequence, we were able to have two groups with no bias for age, gender, or site. All patients preoperatively underwent a comprehensive ophthalmic examination including a lacrimal irrigation test by an ophthalmologist, CT imaging with dacryocystography, and an intranasal examination by an otolaryngologist. All patients were diagnosed as having NLDO. Informed consent was obtained from each patient in accordance with the Declaration of Helsinki.
Study enrollment 62 non-randomized, consecutive patients received a diagnosis of NLDO
62 patients (75 sites) underwent endonasa lDCRs performed from December 2010 to December 2013
Group 1
Group 2
24 patients (28 sites) underwent DCRs without eFS performed from December 2010 to February 2012
38 patients (47 sites) underwent eFS-DCRs performed from March 2012 to December 2013
All 28 sites were included in the analysis of success rate and ostium diameter
8 sites were excluded: 6 obstruction (failure) 2 accidental removal
20 sites were included in the analysis of duration of stent placement
All 47 sites were included in the analysis of success rate and ostium diameter
7 sites were excluded: 1 obstruction (failure) 5 accidental removal 1 revision case
40 sites were included in the analysis of duration of stent placement
Figure 1. Flow chart for study enrolment. DCR, dacryocystorhinostomy; eFS, endonasal flap suture; NLDO, nasolacrimal duct obstruction.
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
164
H. Tachino et al.
All patients had the stent tube inserted into the lacrimal passage during the surgery to keep the canalicular patency. Within 5 days after the surgery, they were discharged from our hospital and visited our outpatient clinic every 2 weeks. They accepted the treatment for cleaning around the ostium of the lacrimal sac and were checked for symptoms, and the patency of the lacrimal passage was tested by the irrigation test and the rhinoscope. Timing of removal of the stent tube was determined by the relief of symptoms such as epiphora, a good result from the lacrimal irrigation test, endonasal video findings such as a fully open ostium, and lacrimal drainage from the common canaliculus opening at blinking, no crust, and no coagulation around the ostium. After removal of the stent tube, we checked the patients after at least 1 month to confirm the patency of the lacrimal passage by both their symptoms and the findings of these examinations. When the patency of the lacrimal passage is confirmed, we usually have finished the patient’s follow-up. If we confirm obstruction of the lacrimal passage after surgery, this is the end point for this study. However we continued the patients’ treatment for their epiphora. There were three study end points: (1) success rate, (2) the diameter of the ostium (OD) of the lacrimal sac after DCR, and (3) the duration of stent placement. These were calculated, measured, and compared between the two groups. Success was defined as anatomic patency of the lacrimal passage as well as relief of symptoms such as epiphora. Anatomic patency was assessed by the lacrimal irrigation test and digital rhinoscopy findings. The OD measurement was conducted at the final check during the follow-up. The OD was measured under the rhinoscope. DCR ostium has a horizontal and vertical diameter. The OD was defined as the smaller one and this is usually the horizontal diameter. Janssen et al. [4] reported that the minimum diameter of the bony nasolacrimal canal was 3.5 mm in a control group and 3.0 mm in the patient group. The OD in this study is not outer diameter (the bony lacrimal canal), but inner diameter (mucosal lacrimal canal). The OD (patient group) must be smaller than the diameter of the bony nasolacrimal canal in a control group. Accordingly, we sorted the OD into four groups, large (> 4 mm), medium (2–4 mm), small (< 2 mm), and obstruction (0 mm). In regard to the OD, we calculated the proportion of the large group to the total, and the proportion of the large plus medium groups to the total. The duration of the stent placement was defined as the duration from stenting at DCR to the removal of the stent tube. Failure, accidental removal, and revision cases (group 1, eight sites; group 2, seven sites) were excluded from these stent placement data (Figure 1).
Statistical analysis was conducted using the GraphPad Prism 6 software for Microsoft Windows (GraphPad Software, Inc., California, USA) to assess the two groups. The success rate and the OD were compared using the Fisher’s exact test. The durations of time that the stents were in place were compared using the Kaplan– Meier method and the log-rank test. In regard to the characteristics of the groups, patients’ ages were compared using the Mann–Whitney U test. The proportion of female sex to the total and the proportion of bilateral affected sides to the total were compared using the Yates 2 2 chisquared test. Surgical technique The operations were performed with the patient under general anesthesia. Standard endonasal DCRs without eFS were performed following the Wormald [5] technique as described previously. We describe our eFS-DCR technique in detail (Figures 2 and 3). A scalpel is used to cut a mucosal flap starting approximately 10 mm above the insertion of the middle turbinate on the lateral nasal wall. Then, two vertical inferior incisions are made. Eventually, these incisions make an inverted U-shaped mucosal flap (Figure 2a). This flap is tucked into the space between the inferior turbinate and the nasal septum. This allows further dissection and manipulation without obstruction of the surgical equipment. The bone, which obscures the lacrimal sac, is removed with a curved diamond DCR burr. After exposure of the entire lacrimal sac, it is incised in an I-shape (Figure 2b). The lacrimal sac membrane is divided into the anterior and the posterior flaps like double doors. When these flaps are turned backward and forward as doors open, the common canaliculus opening is fully visible (Figure 2c, d). After making the lacrimal flaps, they are sutured to the nasal mucosal membrane. At first, the posterior lacrimal flap is sutured to the nasal membrane, because this stitch is the easiest of all (Figure 3a). Two stitches are needed for the posterior flap (Figure 2c, d). Next, the anterior lacrimal flap is sutured to the nasal membrane (Figure 3c). This stitch is the hardest of all, because it is difficult to precisely insert a needle into the lateral nasal wall membrane because of the narrow working space. This is the primary reason that suturing in the nasal cavity seems to be impossible. The nasal cavity sometimes does not have sufficient working space to suture. In these cases, the inverted U-shaped nasal flap, which was made at the beginning of the surgery, is used as the target for suturing the
Endonasal flap suture-dacryocystorhinostomy
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
a
c
b
a
b
c
d
165
d
Figure 2. Schema of the surgical technique for endonasal flap suture-dacryocystorhinostomy (eFS-DCR). (a) Making the inverted U-shaped nasal flap. (b) Incision of the lacrimal sac in an I-shape (laterally H-shape). The lacrimal sac membrane is divided into two (anterior and posterior) flaps like double doors. (c) If the nasal cavity does not have sufficient working space to suture, the inverted U-shaped nasal flap is used as the target for suturing the anterior lacrimal flap. When the anterior lacrimal flap is sutured to the inverted U-shaped nasal flap, this nasal flap naturally goes back to the original position. (d) If the nasal cavity has sufficient working space to suture, we can suture the anterior lacrimal flap to the nasal membrane. Then the inverted U-shaped nasal flap is cut to the appropriate length to cover the bone wall below the lacrimal sac.
Figure 3. Endoscopic findings for the endonasal flap suturedacryocystorhinostomy (eFS-DCR) procedures. (a) It is sutured from the nasal flap to the posterior lacrimal flap. The blue suture is Prolene (Ethicon). Black star ($) indicates the middle turbinate in the right nasal cavity. Circle (*) indicates the inside of the lacrimal sac. Black triangle (~) indicates the exposed bone of the lateral nasal wall covered with bleeding. (b) Ligation by knot pusher. (c) Insertion of a needle into the anterior lacrimal flap. (d) After suture of the anterior lacrimal flap to the inverted U-shaped nasal flap: black square (&). This nasal flap goes back to the original position by suture.
these narrow working spaces. A knot pusher (Hope Denshi Co., Ltd., Chiba, Japan) is useful for suturing in deep spaces (Figure 3b). After suturing, a pair of Nunchaku-style silicone tubes (Lacrifast, Kaneka Medix Corp., Osaka, Japan) are inserted into the lacrimal passage through the common canaliculus opening.
Results anterior lacrimal flap (Figure 2c). It is easier to insert a needle into the free nasal flap than the fixed anterior nasal wall membrane. When the anterior lacrimal flap is sutured to the inverted U-shaped nasal flap, this nasal flap naturally goes back to the original position (Figure 2c). This flap could be trimmed away if necessary. If the nasal cavity has sufficient working space to suture, we can continue to suture the anterior lacrimal flap to the nasal membrane (Figure 2d). Then the inverted U-shaped nasal flap is cut to the appropriate length to cover the bone wall below the lacrimal sac. In addition, specialized surgical equipment makes it possible to do the most difficult stitch in the narrow nasal cavity. The needle holder (Aesculap, Inc., Pennsylvania, USA) helps to manipulate a needle easily in
In regard to the characteristics of the groups, there were no significant differences in patients’ ages, sex distribution, or affected side among the groups (Table I). The success rate was 78.6% (22 of 28 sites) for group 1 (DCR without eFS) and 97.9% (46 of 47 sites) for group 2 (eFS-DCR) (Table II). There was a statistical difference between the two groups (p = 0.00894). In the present study, there was a much higher success rate for eFS-DCR than for endonasal DCR without eFS. The proportion of the large group of the OD to the total number was 10.7% in group 1 and 85.1% in group 2 (Table II). There was a statistical difference between the two groups (p < 0.01). The proportion of the large plus medium groups of the OD to the total
166
H. Tachino et al.
Table I. Demographic data for the two groups. Variable
Group 1 (DCR without eFS)
Group 2 (eFS-DCR)
n = 24 (28)
n = 38 (47)
72.9 ± 11.0 (53–91)
69.4 ± 11.5 (42–88)
0.281
17 (75%)
29 (70%)
0.861
Adult (sides) Age (years), mean ± SD (range) Female sex, n (%)
P value
Side, n (%) 12 (50%)
13 (34%)
Left
8 (33%)
16 (42%)
Bilateral
4 (17%)
9 (24%)
135.1
113.4
Follow-up period (days) 1
2
0.731
3
Success rate, OD 1/total, and OD 1+2/total. DCR, dacryocystorhinostomy; eFS, endonasal flap suture. (i) Mann-Whitney U test (two-tailed); (ii) Yates 2 2 chi-squared test.
Table II. Main outcome measures between the two groups. Outcome
Group 1 (DCR without eFS)
Group 2 (eFS-DCR)
22/28 (78.6%)
46/47 (97.9%)
1, large (> 4 mm)
3
40
2, medium (2–4 mm)
4
5
Success rate
p value 0.0089
OD variable
3, small (< 2 mm) 4, obstruction Total OD 1/total OD 1 + 2/total
15
1
6
1
28
47
3/28 (10.7%)
40/47 (85.1%)
1.34 10–10
7/28 (25%)
45/47 (95.7%)
1.09 10–10
DCR, dacryocystorhinostomy; eFS, endonasal flap suture. p values were calculated by Fisher’s exact test.
number was 25% in group 1 and 95.7% in group 2. There was a statistical difference between the two groups (p < 0.01). This means that the ostium size of patients in group 2 (eFS-DCR) was significantly larger than that of patients in group 1 (DCR without eFS). The duration of stent placement in the two groups is shown in Figure 4. The median time of this duration was 42.5 days in group 1 and 31.5 days in group 2. The log-rank-based assessment at 50% of the patients showed a statistical difference between the two groups (p = 0.0003). The hazard ratio (group 2/ group 1) was 2.127 (95% confidence interval (CI) = 1.796–4.820). The duration of stent placement for patients in group 2 was shorter than that for patients in group 1. This means that the patients that underwent eFS-DCR (group 2) had the stent tube removed sooner than the patients that underwent DCR without eFS (group 1).
Duration of stent placement 100 Patients with stent placement (%)
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Right
p = 0.0003 by log-rank test 50 Group 2 eFS-DCR
Group 1 DCR without eFS
0 0
20
40 60 Postoperative days
80
100
Figure 4. Kaplan–Meier curves for the duration of stent placement between the two groups. The dotted line indicates group 1 (dacryocystorhinostomy, DCR, without endonasal flap suture, eFS) and the solid line indicates group 2 (eFS-DCR). According to the Kaplan–Meier method and log-rank test, the duration of stent placement for patients in group 2 (eFS-DCR) was shorter than that for patients in group 1 (DCR without eFS).
Endonasal flap suture-dacryocystorhinostomy
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
Discussion Acquired NLDO is one of the most common problems in the lacrimal drainage system. The majority of acquired NLDO is idiopathic in nature. Other causes of NLDO include acute or chronic inflammation, trauma, facial palsy, drugs, and complications of nasal surgery. Patients generally present with epiphora, eyelid and lacrimal sac swelling, purulent secretion, blurred vision, and facial pain. External DCR is now regarded as the gold standard in the treatment of NLDO. The reported success rates of this approach have varied but are often reported as over 90% in many specialist lacrimal units. However, the procedure leaves a cutaneous scar on the patient’s face and the potential for injury to medial canthal structures, cerebrospinal fluid rhinorrhea, and functional interference with the physiological action of the lacrimal pump. Vagefi et al. [6] reported that among 215 patients and 247 surgeries, 16 individuals (7.4%) were identified who demonstrated abnormalities of eyelid closure in the postoperative period after external DCR. The advantages of endonasal DCR include the absence of cutaneous scarring, maintenance of the pumping mechanism of the orbicularis oculi muscle, and less disruption of the medial canthal anatomy. However, the success rates for endonasal DCR have varied. The disadvantage of endonasal DCR is that it is very difficult to suture the lacrimal membrane to the nasal membrane. Sometimes the lacrimal ostium has closed after endonasal DCR. This is one of the reasons for reduced success rates for endonasal DCR compared with external DCR. Here, we have presented a new endonasal pocedure, eFS-DCR. Our procedure could resolve the disadvantage of endonasal DCR. First, eFS-DCR had a much higher success rate than endonasal DCR without eFS in our study. Second, the ostium size following eFS-DCR was significantly larger than that after endonasal DCR without eFS. Chan and Selva [7] reported that after endoscopic DCR, the final ostium size on average is 35% of the original at 12 months postoperatively. In the present study, the ostium size after eFS-DCR might shrink less because of the suturing. This result directly relates to the higher success rate for eFS-DCR. Third, patients who underwent eFS-DCR were able to have the stent tube removed more quickly than the patients who underwent DCR without eFS. In regard to the wound healing around the lacrimal ostium, it was more rapid after eFS-DCR than after DCR without eFS. Callejas et al. [8] recommended that stents only be inserted when a tight common canaliculus opening is found during surgery. Cannon et al. [9] reported
167
that there were no cases of canalicular closure or stenosis at 12 months in their prospective series of no intubation for primary acquired nasolacrimal duct obstruction (PANDO). They advocated that routine intubation for the purpose of maintaining canalicular patency is not necessary when performing endonasal DCR in PANDO. However, when the lacrimal ostium shrinks almost to the point of closure, insertion of a stent tube will guarantee prevention of obstruction of the ostium. If the size of the lacrimal ostium is large enough in cases such as those undergoing eFS-DCRs, stent intubation might be unnecessary and it would be possible to reduce medical costs. In fact, in five cases that underwent eFS-DCR in which there was an accidental removal of the tube, there was ultimately success in the patency of the lacrimal passage in this study. eFS-DCR has one disadvantage: extra time must be allowed for suturing. It will take about 15 min to suture the lacrimal flap to the nasal flap in the nasal cavity. However, this extra time will be reduced as surgeons gain skill in the procedure for eFS, or by the development of new special surgical tools for eFS. If the eFS technique for endonasal DCR spreads throughout the world and the success rate of endonasal DCR increases to be comparable to external DCR, many patients, especially females with NLDO, who refused the external DCR approach because it leaves a cutaneous scar, would reverse their decision and accept this surgery. Wormald [10] stated that cooperation between the otolaryngologist and the ophthalmologist might be advantageous for the best possible management of the patient with epiphora. When both the ophthalmologist and the otolaryngologist participate in endonasal DCRs, this cooperative work will lead to successful results in endonasal DCRs.
Acknowledgement I would like to thank to Mrs. Carol Macbain for helpful discussion. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Caldwell GW. Two new operations for obstruction of the nasal duct, with preservation of the canaliculi, and with an incidental description of a new lacrymal probe. Am J Ophthalmol 1893;10:189–93.
168
H. Tachino et al.
Acta Otolaryngol Downloaded from informahealthcare.com by Washington University Library on 01/10/15 For personal use only.
[2] Toti A. Nuovo metodo conservatore dicura radicale delle suppurazione croniche del sacco lacrimale (dacricistorhinostomia). Clin Moderna (Firenza) 1904;10:385. [3] Leong SC, Macewen CJ, White PS. A systematic review of outcomes after dacryocystorhinostomy in adults. Am J Rhinol Allergy 2010;24:81–90. [4] Janssen AG, Mansour K, Bos JJ, Castelijns JA. Diameter of the bony lacrimal canal: normal values and values related to nasolacrimal duct obstruction: assessment with CT. AJNR Am J Neuroradiol 2001;22:845–50. [5] Wormald PJ. Powered endoscopic dacryocystorhinostomy. Laryngoscope 2002;112:69–72. [6] Vagefi MR, Winn BJ, Lin CC, Sires BS, LauKaitis SJ, Anderson RL, et al. Facial nerve injury during
[7] [8]
[9]
[10]
external dacryocystorhinostomy. Ophthalmology 2009;116: 585–90. Chan WO, Selva D. Ostium shrinkage after endoscopic dacryocystorhinostomy. Ophthalmology 2013;120:1693–6. Callejas CA, Tewfik MA, Wormald PJ. Powered endoscopic dacryocystorhinostomy with selective stenting. Laryngoscope 2010;120:1449–52. Cannon PS, Chan WO, Selva D. Incidence of canalicular closure with endonasal dacryocystorhinostomy without intubation in primary nasolacrimal duct obstruction. Ophthalmology 2013;120:1688–92. Wormald PJ. Endoscopic and external dacryocystorhinostomy (DCR) – which is better? Braz J Otorhinolaryngol 2012; 78:2.
