Tear Meniscus Volume Changes in Dacryocystorhinostomy Evaluated With Quantitative Measurement Using Anterior Segment Optical Coherence Tomography

Abstract

Purpose.: To evaluate tear meniscus (TM) changes in external dacryocystorhinostomy (ex-DCR) with quantitative measurement of tear meniscus height (TMH), area (TMA), and volume (TMV) using anterior segment optical coherence tomography (AS-OCT).

Methods.: Twenty-five eyes from 21 patients (11 males and 10 females) with primary acquired nasolacrimal duct obstruction (PANDO) who received ex-DCR from May 2010 to April 2011 were evaluated prospectively on their TMH, TMA, and TMV changes by AS-OCT. Measurements were performed before surgery (Pre) and 2 weeks (2W), 2 months (2M), and 6 months (6M) after surgery. Data were analyzed using Kruskal-Wallis test, Wilcoxon signed-rank test with Bonferroni adjustment, and Spearman's rank correlation coefficient.

Results.: All patients had a good clinical course, and there were significant differences in the values of each TM parameter before and after surgery (P < 0.0001). The median values of TMH (mm) throughout the observation period were 0.707 (Pre), 0.334 (2W), 0.278 (2M), and 0.277 (6M). The TMA median values (mm²) were 0.1097 (Pre), 0.0483 (2W), 0.0255 (2M), and 0.0224 (6M). The TMV median values (mm³) were 0.7799 (Pre), 0.1614 (2W), 0.1071 (2M), and 0.1553 (6M). There were significant differences in TMH, TMA, and TMV reduction at each postoperative visit as compared to preoperative values (P < 0.001). In addition, TMH change 6 months after ex-DCR showed a significant positive correlation with age (r = 0.4434, P = 0.0264).

Conclusions.: The perioperative TM changes in ex-DCR can be evaluated noninvasively and quantitatively by using AS-OCT.

Introduction

In treatment for obstruction of the lacrimal drainage system, tear meniscus (TM) is one of the most important clinical indices for therapeutic evaluation, which can be assessed with slit-lamp examination and fluorescein dye disappearance test as a semiquantitative measurement. ¹ However, it is obvious that all lacrimal functions are not of the same flow, and more sophisticated tests are needed to quantitate lacrimal function. Recently, anterior segment optical coherence tomography (AS-OCT) has enabled noninvasive and quantitative estimation of the TM. ^1–3 An advantage of AS-OCT with Fourier domain detection is that a high-resolution image can be obtained with minimal influence of eye movement using data processing technologies. ⁴ Furthermore, AS-OCT can measure not only tear meniscus height (TMH), but also tear meniscus area (TMA) and tear meniscus volume (TMV). Anterior segment OCT has been reported as a useful way to measure the severity of dry eye and to evaluate the efficacy of treatments. However, no recent studies have addressed the importance of TMH and its time course in primary acquired nasolacrimal duct obstruction (PANDO) patients, despite the recent developments in AS-OCT. There have been few reports of assessment of perioperative changes in TM in patients with obstruction of the lacrimal drainage system, ^1,2 and no study has evaluated the time course of TMH change in patients with nasolacrimal duct obstruction using AS-OCT. 

In the present study, we evaluated TM change in external dacryocystorhinostomy (ex-DCR) with quantitative measurement of TMH, TMA, and TMV using AS-OCT. 

Methods

Subjects

This study was approved by the Institutional Review Board of the University of Tokyo Hospital and was conducted in accordance with the tenets of the Declaration of Helsinki. Informed consent was received from all the enrolled patients. The design of the present study was prospective. Patients who underwent ex-DCR from May 2011 to April 2012 at the University of Tokyo Hospital were enrolled. 

Preoperative ophthalmic examination and irrigation of the nasolacrimal duct were performed. To diagnose PANDO, all subjects underwent nasolacrimal probing with the Bangerter lacrimal probe cannula (Geuder AG, Inc., Heidelberg, Germany). The tip was stopped at the lacrimal sac, and regurgitation from upper and lower puncta was confirmed. Exclusion criteria were canalicular obstruction, partial obstruction, history of congenital obstruction, and identified causes for secondary nasolacrimal duct obstruction. We also excluded two eyes of two patients in whom the ex-DCR was not effective, defined as the patient's dissatisfaction, reappearance of epiphora, and occlusion of anastomosis within 6 months after surgery. 

Procedures for Ex-DCR

All ex-DCRs were performed by the same surgeon under local anesthesia using 2% lidocaine containing 0.0125 mg/mL epinephrine (Astra Zeneca K.K., Osaka, Japan). A skin incision was made at 2 mm from the medial canthus. Orbicularis muscle fibers were separated bluntly to expose the periosteum overlying the anterior lacrimal crest and the medial canthal tendon. A periosteum incision was made inferomedial to the insertion of the medial canthal tendon. A square-shaped osteotomy of approximately 10 × 10 mm called a bony window was made by a drill with a diamond-coated 2-mm round tip (Tact Medical, Inc., Tokyo, Japan). The bony window was extended to the nasal mucosa. The nasal mucosa was then removed along the bony window. For accurate lacrimal sac incision, 1% purified methylrosanilinium chloride (Wako Pure Chemical Industries, Ltd., Osaka, Japan) diluted to 0.01% with saline solution was injected into the lacrimal sac through the upper canaliculus for mucosal epithelium staining. In all cases of ex-DCR, a one-flap anastomosis was made; posterior flap using the lacrimal sac was fixed by electrocoagulation, and then a bicanalicular silicone stent (Nunchaku-style tubing [NST]; Kaneka Medix, Corp., Osaka, Japan) was inserted for all of the patients. After confirmation of hemostasis, the orbicularis muscle was closed with 6-0 polyglactin absorbable sutures. Finally, the skin was closed with 6-0 monofilament suture. 0.01% fluorometholone (Santen Pharmaceutical Co., Ltd., Osaka, Japan) and 0.5% tranilast (Kissei Phamaceutical Co., Ltd., Nagano, Japan) were used for suppression of granulation tissue four times per day after surgery, and 0.5% levofloxacin (Santen Pharmaceutical Co., Ltd.) hydrate was also administered four times per day after surgery for antibiotic prophylaxis. Fluorometholone was used for 4 months postoperatively. Tranilast and levofloxacin hydrate were used during the study period. Postoperative follow-up visits were performed after surgery at 2 weeks and monthly. Each visit included evaluation of TM with slit-lamp examination and AS-OCT, and then evaluation of the patency of the created passage by lacrimal irrigation. Nunchaku-style tubing (Kaneka Medix, Corp.) was used for all subjects and removed after examinations at 2 months after surgery. In three eyes of two patients, mild symptoms and findings of dry eye were observed postoperatively, and purified sodium hyaluronate (Santen Pharmaceutical Co., Ltd.) four times per day was used for these eyes. 

Procedure for TM Measurement

In all subjects, TMH, TMA, and TMV were measured with Fourier-domain swept-source AS-OCT (SS-1000; Tomey, Inc., Tokyo, Japan). The system operates at 1310 nm, and the acquisition speed of the instrument is 30,000 A-scans per second. The axial and transverse resolutions are approximately 8 and 30 μm, respectively. The custom scan mode, which we refer to as the TM mode, is composed of 16 equally separated cross-sectional anterior segment images covering an area 16 mm (length) by 16 mm (width). The scanning position of the eye at the beginning was the center of the cornea, and the internal fixation target controlled the same measurement position, which was necessary for quantitative comparison. The measurement was performed 1 second after a blink with spontaneous opening of the eyelids, and the acquisition time for all scans was 0.3 seconds. The inferior TM was measured. 

Procedures for measurement the TM value were performed on 300% magnified images. The TMH value was defined as line distance of fluid surface from the inferior cornea–meniscus junction to the lower eyelid–meniscus junction. The line distance between two points was decided manually using the software calipers on the magnified image and calculated automatically by the software. The TMA value was defined as an inferior triangular TM area formed by the corneal anterior boundary, the anterior boundary of the lower eyelid, and the anterior borderline of the TM. The triangle boundary line was traced manually using the software calipers; the area within the traced lines was calculated automatically by the software. The TMV value was calculated automatically with the integrated TMA values of the consecutive central 11 images by the software. The resulting TMA and TMV values were divided by the refractive index of balanced salt solution (1.343) to correct the refraction at the air–meniscus interface as described in a previous study. ⁴  

Statistics

The statistical analysis was performed using JMP9 software (SAS, Inc., Cary, NC). Numerical data of TM parameters are presented as the median. Kruskal-Wallis test was used to assess the changes in TMH, TMA, and TMV before and after ex-DCR. Multiple comparisons were performed by Wilcoxon signed-rank test with Bonferroni adjustment. The TM change was calculated using the following formula: Change in TM (%) = (preoperative TM value − postoperative TM value) ÷ (preoperative TM value) × 100. Correlation between age and TM change was evaluated by Spearman's rank correlation coefficient. P < 0.05 was accepted as statistically significant. 

Results

Table 1 shows demographic data of the patients. No complications occurred during or after surgery, and all patients had surgical success according to patient satisfaction with disappearance of epiphora and lacrimal patency up to 6 months after surgery. Tear menisci in patients with PANDO were clearly visualized preoperation and postoperation on AS-OCT (Fig. 1). 

Figure 1

 

AS-OCT images of a patient with PANDO before (at preoperation [A]) and after surgery (at 2 weeks [B], 2 months [C], and 6 months [D]). Cornea (CO), lower eyelids (LL), and lower tear menisci (LTM) were visualized. In the patient with PANDO, the LTM became significantly lower after surgery (B–D).

Table 1

 

Demographic Data of the Subjects

Number (male/female)	21 (11/10)
Operative eye (right/left)	25 (17/8)
Mean age, y (SD); range	66.2 (11.2); 42–83

 SD, standard deviation.

There were significant differences in the values of each TM parameter before and after surgery (Kruskal-Wallis test, P < 0.0001; Table 2). The median value of TMH and the change from baseline were 0.707 mm and 0% at baseline or preoperatively (Pre), 0.334 mm and 52.3% at 2 weeks postoperatively (2W), 0.278 mm and 52.0% at 2 months (2M), and 0.277 mm and 58.5% at 6 months (6M), respectively. There were significant differences in TMH reduction at each point postoperatively (Wilcoxon signed-rank test with Bonferroni adjustment, P = 0.0003 at 2W, P < 0.0001 at 2M and 6M; Fig. 2). The median value of TMA and the change from baseline were 0.1097 mm² and 0% (Pre), 0.0483 mm² and 51.3% (2W), 0.0255 mm² and 73.9% (2M), and 0.0224 mm² and 76.4% (6M), respectively. There were significant differences in TMA at each point postoperatively (Wilcoxon signed-rank test with Bonferroni adjustment, P < 0.0028 at 2W, P < 0.0001 at 2M and 6M; Fig. 2). The median value of TMV and the change from baseline were 0.7799 mm³ and 0% (Pre), 0.1614 mm³ and 80.0% (2W), 0.1071 mm³ and 81.9% (2M), and 0.1553 mm³ and 80.2% (6M), respectively. There were significant differences in TMV at each time postoperatively (Wilcoxon signed-rank test with Bonferroni adjustment, P = 0.0009 at 2W, P < 0.0001 at 2M and 6M; Fig. 2). 

Figure 2

 

Change in tear menisci after ex-DCR in patients with PANDO. The TMH value was decreased postoperatively ([A] P = 0.0003 at 2 weeks, P < 0.0001 at 2 months and 6 months). The TM change was calculated using the following formula: Change in TM (%) = (preoperative TM value − postoperative TM value) ÷ (preoperative TM value) × 100. The change in TMH is shown in (B). The TMA value was decreased postoperatively ([C] P < 0.0028 at 2 weeks, P < 0.0001 at 2 months and 6 months). The change in TMA is shown in (D). The value of TMV was decreased postoperatively ([E] P < 0.0009 at 2 weeks, P < 0.0001 at 2 months and 6 months). The change in TMV is shown in (F). *Wilcoxon signed-rank test with Bonferroni adjustment; ♦ maximum value or change rate of TMH, TMA, TMV; ▪ median value or change rate of TMH, TMA, TMV; ▴ minimum value or change rate of TMH, TMA, TMV.

Table 2

 

Change in TMH, TMA, and TMV After Ex-DCR

	Pre	2W	2M	6M	P
TMH, mm (range)	0.707 (0.287–0.945)	0.334 (0–0.897)	0.278 (0–0.745)	0.277 (0–0.716)	<0.0001*
TMA, mm2 (range)	0.1097 (0.0196–0.2391)	0.0483 (0–0.2440)	0.0255 (0–0.1241)	0.0224 (0–0.1205)	<0.0001*
TMV, mm3 (range)	0.7799 (0.0975–1.8681)	0.1614 (0.0246–1.6392)	0.1071 (0.0103–0.9195)	0.1553 (0.0355–0.7633)	<0.0001*

 Each value represents a median. Range values represent the values of each tear meniscus parameter from minimum to maximum.

*  Kruskal-Wallis test.

In addition, we evaluated the association between TM change by ex-DCR for PANDO and age. The TMH change at 6M after ex-DCR showed a significant positive correlation with age (Spearman's rank correlation coefficient, r = 0.4434, P = 0.0264; Fig. 3). 

Figure 3

 

The change in TMH at 6 months after ex-DCR showed a significant positive correlation with age (Spearman's rank correlation coefficient, r = 0.4434, P = 0.0264).

Discussion

In the present study, AS-OCT was used to evaluate the time course of changes in TM in patients with PANDO who received ex-DCR. Anterior segment OCT demonstrates good accuracy and repeatability of height, area, and volume, and also demonstrates correlation with scale reading on the slit-lamp examination. ^{3 –5} Previous studies have evaluated TM parameters using AS-OCT in healthy volunteers ^1,3–13 and in patients with dry eye, ^3,5 punctal occlusion by plugs, ⁶ punctual stenosis, ² and PANDO. ^1,7,14 To date, however, the precise time course of changes in TM parameters in patients with PANDO who received ex-DCR has not been clear. 

Previous studies have reported that mean values of TMH in normal eyes varied between 0.23 and 0.40 mm ^1,3–10 whereas those in eyes with PANDO varied between 0.54 and 0.62 mm. ^1,7,14 In this study, the preoperative median value of TMH was 0.707 mm, which was higher than in previous studies, while 0.277 mm at 6M postoperatively was in the range for normal eyes. Mean values of TMA in normal eyes vary between 0.0219 and 0.0456 mm². ^4,7,10–12 Whereas those in eyes with PANDO are 0.1010 mm². ⁷ In our study, the preoperative median value of TMA was 0.1097 mm², which was similar to findings in previous studies, and 0.0224 mm² at 6M postoperatively, in the range for normal eyes. The mean value of TMV in normal eyes was reportedly 0.1338 mm³ (Ref. 4); the TMV value of PANDO patients or its time course after ex-DCR has never been reported. In our study, the preoperative median value of TMV was 0.7799 mm³, while 0.1553 mm³ at 6M after operation was similar to the value for normal eyes in previous studies. 

After ex-DCR, there was a significant change in TMH, TMA, and TMV values in patients with PANDO over time, compared to preoperative values (Fig. 2). To be exact, significant reductions in TM parameters were achieved in the early postoperative period, and the reduction persisted throughout the observation period up to 6M postoperatively. It is of note that the reduction in TM persisted despite removal of NST at 2M postoperatively and tapering of postoperative medications. 

The TMH change showed a significant positive correlation with age at 6M after ex-DCR (Fig. 3). Older patients had a greater reduction in TMH. Reasons for this result are as follows. First, lacrimal hyposecretion associated with aging may have been masked by PANDO before surgery, and it appeared postoperatively. Second, conjunctivochalasis with aging may result in a higher TM because of the shallow space of the conjunctival sac. Collectively, the results suggest that dry eye with excessive TM reduction could occur more easily in older patients with PANDO. 

Limitations of our study include the fact that it was conducted in a single hospital and also the sample size. Because many of the subjects were elderly, it may be difficult to compare the findings to those of previous studies ^3–5 on young volunteers due to different anterior segment status with aging. 

In conclusion, we evaluated TM change in the perioperative course of ex-DCR using AS-OCT. Postoperative TMH, TMA, and TMV were reduced significantly compared with the preoperative values. The TMH change at 6M postoperatively was significantly correlated with age. Perioperative TM changes after ex-DCR can be evaluated noninvasively and quantitatively with use of AS-OCT. 

Acknowledgments

Disclosure: K. Ohtomo, None; T. Ueta, None; R. Fukuda, None; T. Usui, None; T. Miyai, None; R. Shirakawa, None; S. Amano, None; M. Nagahara, None