Variation of Lamina Cribrosa Depth Following Trabeculectomy

Abstract

Purpose.: To investigate the long-term changes in lamina cribrosa (LC) depth following trabeculectomy.

Methods.: Serial horizontal B-scan images of the optic nerve head were obtained using spectral-domain optical coherence tomography from 28 primary open-angle glaucoma patients who underwent trabeculectomy and were followed up for at least 2 years. Approximately 65 B-scans covering the optic discs were obtained before surgery and at 6 months and ≥2 years postoperatively. The pre- and postoperative LC depth (the distance from the opening plane of Bruch's membrane to the level of the anterior LC surface) was determined on seven selected B-scan images from each eye and averaged (mean LC depth).

Results.: The intraocular pressure (IOP) decreased from 27.4 ± 9.0 (mean ± SD) to 9.7 ± 3.1 mm Hg at postoperative 6 months (P < 0.001) and subsequently increased to 12.7 ± 5.1 mm Hg at a mean final follow-up of 27.1 ± 3.3 months (P = 0.001). The mean LC depth was reduced from 625.6 ± 186.3 to 499.6 ± 140.6 μm at postoperative 6 months (P < 0.001). A subsequent slight, but nonsignificant, increase in the LC depth was noted at final follow-up. The degree of LC depth increase after 6 months was significantly associated with younger age, higher IOP at final follow-up, greater IOP fluctuation, and higher mean follow-up IOP from 6 months to final follow-up (all P < 0.05).

Conclusions.: The postoperative reduction in the LC depth that was observed 6 months after surgery was not maintained in some eyes. The redisplacement of the LC after postoperative 6 months appeared more likely to occur in patients who were younger and who had higher IOP and IOP fluctuation during the postoperative follow-up.

Introduction

Glaucoma is a multifactorial disease, and lowering the intraocular pressure (IOP) has been the mainstay treatment. According to the recently suggested biomechanical paradigm, IOP plays a central role in the pathogenesis of optic nerve damage in glaucoma. ¹ The IOP-related stress, which can be substantial even at low levels of IOP, ² may generate strain on the optic nerve. The resulting tensile strain within the lamina cribrosa (LC) would provide shearing stress on the axons passing through the laminar pores. In addition, the LC deformation may be relevant with optic nerve ischemia because the strain within the LC can compress the laminar capillaries, which are contained in the laminar trabeculae.¹ Thus, imaging the LC and determining factors associated with LC deformation may not only expand our understanding of the pathogenesis of glaucomatous damage but also help to develop better strategies in treating glaucoma. 

Until now, however, little observation has been performed with regard to the biomechanical behavior of the LC. This may be largely due to the lack of imaging technology that can capture the LC in patients. With the emergence of spectral-domain optical coherence tomography, it became possible to obtain images of the LC, especially with enhanced depth imaging technique (EDI). Using this technique, we have recently demonstrated that displacement and compression of the LC may be reversed after trabeculectomy in eyes with primary open-angle glaucoma (POAG). ³ This finding provides an explanation for how IOP lowering can be beneficial in preventing or slowing glaucoma progression from a biomechanical perspective. Since deformation and compression of the LC may induce shearing stress in the axons passing through the laminar pores and possible occlusion of the laminar capillaries, ^4–8 it can be postulated that reversal of the LC deformation would provide relief to the compressed nerve fibers or laminar capillaries. ^3,9  

In a previous study ³ we described changes in the LC depth up to 6 months after trabeculectomy. The reversal of displacement of the LC occurred rather rapidly and then continued gradually up to 6 months, and the amount of reversal was associated with percent degree of IOP lowering. The purpose of the current study was to elucidate the longer-term changes in the LC depth (i.e., later than 6 months after surgery). As it is not uncommon for IOP to re-elevate after 2 years in trabeculectomized eyes, long-term follow-up observation would provide a window of opportunity to examine the effect of varying IOP on LC configuration. 

Methods

The study included POAG patients who were followed up for at least 2 years after trabeculectomy. The study was approved by the Seoul National University Bundang Hospital Institutional Review Board and conformed to tenets of the Declaration of Helsinki. Informed written consent to participate was obtained from all subjects. 

The preoperative examinations and criteria for inclusion and exclusion have been described in detail elsewhere. ³ In brief, all patients underwent a complete ophthalmic examination including visual acuity assessment, refraction test, slit-lamp biomicroscopy, gonioscopy, Goldmann applanation tonometry, and dilated stereoscopic examination of the optic disc. They also underwent central corneal thickness measurement (Orbscan II; Bausch & Lomb Surgical, Rochester, NY), spectral-domain optical coherence tomography (SD-OCT), and standard automated perimetry (24–2 Swedish interactive threshold algorithm; Humphrey Field Analyzer II 750; Carl Zeiss Meditec, Dublin, CA). 

The inclusion criteria were a diagnosis of POAG, a best corrected visual acuity of ≥20/40, a spherical refraction of −8.0 to +5.0 diopters, and cylinder correction within ±3.0 diopters. POAG was defined as the presence of glaucomatous optic nerve damage and associated visual field defects without ocular disease or conditions that may elevate the IOP. Indications for trabeculectomy were IOP deemed to be associated with a high risk for progression, or glaucomatous progression of the visual field or optic disc in spite of maximally tolerated medications. 

Eyes that had undergone previous intraocular surgery or had coexisting retinal or neurological diseases that could have affected the visual field were excluded from this study. Eyes were also excluded when a good-quality image (i.e., quality score >15) could not be obtained at more than five sections (when the quality score does not reach 15, the image acquisition process automatically stops, and images of the respective sections are not obtained). 

For this study, the follow-up period was divided into an initial 6-month follow-up (initial follow-up) and a subsequent follow-up of up to 2 or 3 years after surgery (subsequent follow-up). The patient's last follow-up appointment (at least 2 years after surgery) is hereafter referred to as the “final” follow-up. 

Optic discs were examined using SD-OCT (Spectralis; Heidelberg Engineering, Heidelberg, Germany) at 1 day before surgery and at 6 months and approximately 2 or 3 years postoperatively (final follow-up). The preoperative IOP was defined as the average of two measurements within 2 weeks before trabeculectomy. The IOP was recorded every 3 months during the subsequent follow-up period, and the average and standard deviation (SD) of the follow-up IOP measurements were defined as the mean follow-up IOP and the IOP fluctuation, respectively. When patients underwent an additional IOP-lowering procedure such as bleb needling, additional trabeculectomy, or glaucoma drainage device implantation, the IOPs measured within 3 months posttreatment were not included in the calculation. 

Enhanced Depth Imaging Optical Coherence Tomography of the Optic Disc

The optic nerve head (ONH) was imaged using the Spectralis (Heidelberg Engineering) OCT device with the EDI technique. The details and advantages of this technology for evaluating the LC have been described previously. ^10–13 Approximately 65 B-scan section images covering the ONH, 30 to 34 μm apart (the scan line distance being determined automatically by the machine), were obtained from each eye. Forty-two OCT frames were averaged for each section, which provided the best tradeoff between image quality and patient cooperation.¹⁰ Images are obtainable using the Spectralis OCT device only when the quality score is higher than 15. When this criterion is not fulfilled, the image acquisition process automatically stops and the image of the respective section is excluded. Only acceptable scans with a good-quality image (i.e., quality score >15) obtained for more than 60 sections and allowing clear delineation of the anterior and posterior borders of the LC were included. 

Quantification of the Lamina Cribrosa Depth

The depth of the LC was measured on the pre- and postoperative B-scan images. The details of this measurement method have been described previously. ^3,9 Excellent interobserver reproducibility in measuring the LC depth has been reported (intraclass correlation coefficient = 0.998). ³ The LC depth was determined by measuring the distance from the opening plane of Bruch's membrane to the level of the anterior LC surface. To this end, a reference line connecting the two termination points of Bruch's membrane (BMO) was drawn in each B-scan. The distance from the reference line to the level of anterior border of LC was then measured at three points: the maximally depressed point and two additional points (100 and 200 μm apart from the maximally depressed point in the temporal direction). Only the temporally adjacent points were selected because the maximally depressed point was often close to the central vessel trunk, whose shadow obscured the LC. In addition, the full-thickness LC was often not clearly discernible at the temporal periphery. Thus, data were collected from only two additional points. The distance was measured on the line perpendicular to the reference line using the Amira software (version 5.2.2; Visage Imaging, Berlin, Germany) manual caliper tool. The average of these three values was defined as the LC depth of the B-scan. 

To determine the mean LC depth, seven B-scan images that divided the optic disc diameter into eight equal parts vertically were selected in each eye from the three-dimensional image data set, as described previously. ⁹ The LC depth was measured in each B-scan as described above, and the average of the measurements was defined as the mean LC depth of the eye. 

For follow-up measurements, sets of B-scans were selected to correspond to those that had been selected for the baseline measurements. En face images as well as the low reflective shadow within the LC shown in B-scan images were used to confirm the correspondence of the selected B-scans between pre- and postoperative images. ^3,9  

The difference in the mean LC depth between each follow-up examination was defined as the mean LC depth decrease. A statistically significant change was accepted with an intersession SD of 1.96 times, since this corresponds to the 95% confidence interval for the true value of the measurement. The intersession reproducibility in the LC depth measurements (ICC = 0.991) and 1.96 times the intersession SD (23.3 μm) have been reported previously. ⁹  

The LC depth was measured by an observer (EJL) who was blind to the clinical information, including the IOP and the time point of the scanning for the follow-up OCT images. 

Data Analysis

The LC depths measured preoperatively and at 6 and 24 months postoperatively were compared using repeated-measures analysis of variance. Linear regression analysis was used to determine the factors associated with the change in the LC depth during the subsequent follow-up. Statistical analyses were performed using PASW Statistics software (version 18.0.0; SPSS, Chicago, IL). The level of statistical significance was set at P < 0.05. 

Results

Thirty-five POAG patients who underwent trabeculectomy were included. Of these, 7 patients were lost to follow-up; the remaining 28 patients were followed up for 27.1 ± 3.3 months (mean ± SD; range, 23–34 months). 

The age was 48.6 ± 17.2 years (range, 15–80 years); 10 subjects were women and 18 were men. The visual acuity ranged from 20/40 to 20/20, and the refractive error (spherical equivalent) was −2.9 ± 2.8 diopters (range, −7.00 to +2.25 diopters). The visual field mean deviation was −14.5 ± 9.9 dB (range, −31.79 to −1.41 dB) (Table 1). 

Table 1

 

Patient Clinical Demographics (n = 28)

Variable
Age, y	48.6 ± 17.2
Gender, male/female	18/10
Spherical equivalent, diopters	−2.9 ± 2.8
Central corneal thickness, μm	546.0 ± 55.2
Axial length, mm	25.0 ± 1.7
Visual field MD, dB	−14.5 ± 9.9
Visual field PSD, dB	8.3 ± 4.1

 PSD, pattern standard deviation. Values are shown as mean ± standard deviation.

Eleven of the 28 eyes required a total of 15 additional surgical interventions during the subsequent follow-up period. Three eyes required needling of an encapsulated bleb with adjunctive 5-fluorouracil delivered by subconjunctival injection (two, two, and five times for each eye, respectively). Five eyes underwent an additional trabeculectomy at 8, 11, 16, 26, and 29 months after initial surgery. Of these five eyes, one also had glaucoma drainage implantation surgery. 

Table 2 lists the IOP and LC depth at each follow-up visit. The IOP decreased from 27.4 ± 9.0 (range, 14–47) to 9.7 ± 3.1 mm Hg (range, 6–16 mm Hg) at postoperative 6 months (P < 0.001) and to 12.7 ± 5.6 mm Hg (range, 6–30 mm Hg) at the final follow-up (P < 0.001). None of the patients had signs of ocular hypotony in terms of reduced visual acuity or retinal folds at the time of SD-OCT. The mean LC depth decreased significantly from a preoperative level of 625.6 ± 186.3 to 499.6 ± 140.6 μm at 6 months postoperatively (P < 0.001), and to 519.0 ± 133.4 μm at final follow-up (P < 0.001). 

Table 2

 

Intraocular Pressure and the Mean Lamina Cribrosa Depth at Each Follow-up Visit

	Preop	Postop 6 Mo	P Value*	Final	P Value*	P Value†
IOP, mm Hg	27.4 ± 9.0	9.7 ± 3.1	<0.001	12.7 ± 5.1	<0.001	0.001
Mean LC depth, μm	625.6 ± 186.3	499.6 ± 140.6	<0.001	519.0 ± 133.4	<0.001	0.116

P  values are based on repeated-measures analysis of variance.

 Values are shown as mean ± standard deviation (μm).

 Values with statistical significance are shown in bold.

*   P values comparing the postoperative and preoperative measurements.

†   P values comparing the final and 6 months postoperative measurements.

At the end of the initial follow-up (i.e., 6 months after surgery), significant LC displacement reversal was observed in 23 of the 28 eyes. In 8 of the 23 eyes (34.8%), LC depth increased again and exceeded the intersession variability (491.9 ± 171.7–585.6 ± 159.2 μm, P = 0.012, Wilcoxon signed-rank test), while 4 eyes (17.4%) exhibited a further LC displacement reversal to a significant level during the subsequent follow-up (632.1 ± 138.5–569.0 ± 109.0 μm, P = 0.068, Wilcoxon signed-rank test). The change in the mean LC depth during the subsequent follow-up period in the remaining 11 eyes (47.8%) was not statistically significant (494.9 ± 122.7–495.8 ± 125.6 μm, P = 0.594, Wilcoxon signed-rank test) (Fig. 1A). 

Figure 1

 

Mean lamina cribrosa (LC) depth at each follow-up in eyes in which a significant LC displacement reversal was observed ([A], n = 23) or not observed ([B], n = 5) during the initial follow-up. Of the 23 eyes in which a significant reversal was observed at postoperative 6 months, 8 showed redisplacement of the LC at the end of the subsequent follow-up (red lines with red circles). Of the five eyes in which significant LC displacement reversal was not found during the initial follow-up, an increase in the mean LC depth was observed in one eye during the subsequent follow-up.

Of the five eyes in which significant LC displacement reversal was not observed during the initial follow-up, an increase in the mean LC depth was observed in one eye during the subsequent follow-up. In another eye, a significant LC displacement reversal was observed at the subsequent follow-up. In the remaining three eyes, there was no change in the mean LC depth during the subsequent follow-up (Fig. 1B). 

Eyes in which the LC depth increased again after the 6-month follow-up were observed in patients who were younger (P = 0.001) and had a higher IOP at final follow-up (P = 0.035), a greater IOP fluctuation (P = 0.007), and a higher mean follow-up IOP from 6 months after surgery (P = 0.022). There was no significant difference in the reduction of the visual field (VF) mean deviation (MD) between eyes with LC redisplacement and eyes without LC redisplacement (P = 0.951) (Table 3). 

Table 3

 

Comparison Between Eyes With and Without a Significant Redisplacement of the Lamina Cribrosa During the Subsequent Follow-up

	Eyes With Significant LC Redisplacement, n = 9	Eyes Without Significant LC Redisplacement, n = 19	P Value*
Age, y	34.3 ± 14.6	55.4 ± 14.1	0.001
Gender, male/female	4/5	14/5	0.132†
IOP at baseline, mm Hg	31.0 ± 10.1	25.7 ± 8.1	0.147
IOP at postoperative 6 mo, mm Hg	10. 3 ± 4.0	9.4 ± 2.6	0.461
IOP at final follow-up, mm Hg	16.6 ± 6.6	10.9 ± 2.9	0.035
IOP reduction from 6 mo, %	−84.7 ± 122.3	−19.6 ± 26.1	0.151
IOP fluctuation, SD	4.9 ± 2.4	2.0 ± 1.1	0.007
Mean follow-up IOP, mm Hg	14.3 ± 4.5	10.9 ± 2.8	0.022
Mean LC depth at baseline, μm	710.0 ± 175.0	585.6 ± 182.3	0.1
Mean LC depth at 6 mo, μm	486.0 ± 161.6	506.0 ± 133.8	0.732
Global RNFL thickness, μm	61.6 ± 15.7	53.3 ± 14.4	0.178
Visual field MD at baseline, dB	−13.7 ± 11.6	−14.9 ± 9.3	0.769
Visual field MD at final follow-up, dB	−15.3 ± 12.2	−16.6 ± 9.8	0.768
Amount of visual field MD deterioration, dB	−1.6 ± 2.0	−1.7 ± 3.0	0.951
Central corneal thickness, μm	548.0 ± 67.9	545.4 ± 53.4	0.928
Axial length, mm	25.0 ± 1.4	25.1 ± 1.9	0.908
Spherical equivalent, diopters	−4.6 ± 3.1	−3.1 ± 4.7	0.390
Follow-up duration, mo	25.0 ± 4.2	26.9 ± 4.8	0.321

 RNFL, retinal nerve fiber layer.

 Values with statistical significance are shown in bold.

*  Except where indicated otherwise, the comparison was performed using independent-samples t-test.

†  Comparison was performed using χ² test.

The factors associated with the increase of LC depth during the subsequent follow-up period were assessed using linear regression analysis. In the univariate analysis, younger age (P = 0.012), female gender (P = 0.016), higher IOP at the final follow-up (P = 0.004), smaller percentage IOP reduction from 6 months to the final follow-up (P = 0.049), greater IOP fluctuation (P < 0.001), and higher mean follow-up IOP (P = 0.005) after postoperative 6 months were significantly associated with an increase of LC depth in the subsequent follow-up period. Since the IOP-related factors had high variance inflation factors, multivariate analysis was performed in four ways to avoid multicollinearity; this revealed significant associations with younger age (P < 0.05), IOP at final follow-up (P = 0.029), greater IOP fluctuation (P < 0.001), and higher mean IOP during the subsequent follow-up period (P = 0.034). (Tables 4, 5, Fig. 2). Figure 3 shows two representative cases with POAG in which there were further LC displacement reversal and increase in the LC depth during the subsequent follow-up period. In case A, the LC depth continued to decrease after 6 months (Fig. 3A-3). However, the LC depth in case B increased at the final follow-up compared with the 6-month LC depth (Fig. 3B-3). 

Figure 2

 

Relationship between the change in the mean lamina cribrosa (LC) depth from postoperative 6 months to the end of the subsequent follow-up period and age (A), intraocular pressure (IOP) at the final follow-up (B), IOP fluctuation (C), and mean follow-up IOP from 6 months after surgery (D). The dotted line indicates 1.96 times the intersession standard deviation (SD; 23.3 μm). Positive and negative values indicate LC redisplacement and further LC depth reduction, respectively.

Figure 3

 

En face (A, B) and B-scan images obtained at baseline (A-1, B-1), postoperative 6 months (A-2, B-2), and final follow-up (A-3, B-3) in two eyes with primary open-angle glaucoma that underwent trabeculectomy. (A) The left eye of an 81-year-old female patient. The IOP had decreased from 17 to 6 mm Hg at 6 months and was 7 mm Hg at 25 months after surgery. Note that the magnitude of the LC depth continuously decreased throughout the entire follow-up period. (B) The right eye of a 51-year-old female patient. The IOP had decreased from 41 to 16 mm Hg at postoperative 6 months, but had increased to 30 mm Hg at 23 months. Note that there was a clear LC depth reduction and then an increase in the LC depth at 6 months and final follow-up, respectively.

Table 4

 

Univariate Analysis of the Factors Associated With Increasing Depth of the Lamina Cribrosa More Than 6 Months Postoperatively

	Univariate Analysis
	Beta (95% CI)	P Value	VIF
Age, per 1 y older	−1.73 (−3.03 to −0.42)	0.012	1.33
Female gender	58.62 (12.05–105.19)	0.016	1.85
IOP at baseline, per 1 mm Hg increase	1.75 (−0.99–4.50)	0.201
IOP at postoperative 6 mo, per 1 mm Hg increase	2.68 (−5.55–10.90)	0.510
IOP at final follow-up, per 1 mm Hg increase	6.47 (2.21–10.73)	0.004	9.07
% IOP reduction from 6 mo, per 10% reduction	−3.10 (−6.19 to −0.01)	0.049	8.00
IOP fluctuation, per 1 SD increase	22.80 (15.08–30.53)	<0.001	4.44
Mean follow-up IOP, per 1 mm Hg increase	8.80 (2.92–14.69)	0.005	12.37
Mean LC depth at baseline, per 100 μm deeper	0.64 (−13.03–14.32)	0.924
Mean LC depth at 6 mo, per 100 μm deeper	−15.09 (−32.17–1.99)	0.081	1.84
Global RNFL thickness, per 1 μm thicker	1.06 (−0.58–2.70)	0.195
Visual field MD, per 1 dB worse	−0.16 (−2.95–2.63)	0.907
Central corneal thickness, per 1 μm thicker	0.31 (−0.15–0.76)	0.176
Axial length, per 1 mm longer	3.25 (−14.77–21.27)	0.712
Follow-up duration, per 1 mo longer	−0.85 (−104.61–188.21)	0.752

 95% CI, 95% confidence interval; VIF, variance inflation factor.

 Values with statistical significance are shown in bold.

Table 5

 

Multivariate Analysis of the Factors Associated With Increasing Depth of the Lamina Cribrosa More Than 6 Months Postoperatively

	Multivariate 1		Multivariate 2		Multivariate 3		Multivariate 4
	Beta (95% CI)	P Value	Beta (95% CI)	P Value	Beta (95% CI)	P Value	Beta (95% CI)	P Value
Age, per 1 y older	–1.51 (−2.65 to −0.38)	0.011	−1.48 (−2.71 to −0.25)	0.021	−1.34 (−2.18 to −0.50)	0.003	−1.64 (−2.76 to −0.52)	0.006
Female gender	24.95 (−18.98–68.88)	0.252	43.79 (−3.85–91.42)	0.070	11.67 (−21.57–44.92)	0.475	13.51 (−33.58–60.60)	0.559
IOP at final follow-up, per 1 mm Hg increase	4.26 (0.48–8.05)	0.029
% IOP reduction from 6 mo, per 10% reduction			−2.24 (−5.03–0.55)	0.111
IOP fluctuation, per 1 SD increase					18.03 (11.04–25.01)	<0.001
Mean follow-up IOP, per 1 mm Hg increase							5.99 (0.49–11.49)	0.034
Mean LC depth at 6 mo, per 100 μm deeper	−12.42 (−27.64–2.81)	0.105	−8.38 (−25.63–8.86)	0.325	−10.42 (−21.77–0.93)	0.070	−14.59 (−29.90–0.71)	0.061

Discussion

In the current study we investigated the long-term changes in LC depth after trabeculectomy in eyes with POAG. Six months after surgery (i.e., the initial follow-up), significant LC depth reduction was observed in most of the eyes. This reversal of LC displacement exhibited a variable course thereafter, either remaining stable, being further reversed, or increasing in depth again. This is the first study to investigate the long-term changes of the LC following IOP-lowering surgery. 

The present study showed a significant increase of the LC depth during the subsequent (i.e., post-6-month) follow-up after IOP-lowering surgery in 9 of 28 eyes (32.1%). These eyes exhibited a higher IOP at final follow-up, a greater IOP fluctuation, and a higher mean IOP over the subsequent follow-up period. This finding suggests that sustained reduction of the IOP is important for maintenance of the reversed LC displacement that occurs after trabeculectomy. 

The deepening of the LC over the subsequent follow-up period in eyes with unstable IOP is in line with findings from previous experimental studies. Burgoyne et al. ¹⁴ and Yang et al. ^15,16 showed that the LC was posteriorly displaced according to IOP elevation in monkey eyes with early experimental glaucoma. A similar finding was demonstrated in the study by Bellezza et al., ¹⁷ in which the LC displacement was observed at the onset of ONH surface change detected by confocal scanning laser tomography. Recently, Strouthidis et al. ¹⁸ showed that the IOP-induced LC displacement was also detectable using longitudinal SD-OCT images in early glaucomatous monkey eyes. Although the current study involved eyes that received trabeculectomy, our observation confirms that displacement of LC according to IOP elevation, which has been demonstrated in previous experimental studies, is a realistic finding in the clinical situation. 

Several clinical trials have emphasized the importance of IOP control for disease prognosis within the glaucoma continuum. The European Glaucoma Prevention Study showed that the mean IOP over time was an important factor for the development of open-angle glaucoma. ¹⁹ In the Advanced Glaucoma Intervention Study, both higher mean follow-up IOP and greater IOP fluctuation were associated with an increased risk of visual field progression. ²⁰ Others ^21,22 have also demonstrated a significant influence of mean IOP and fluctuation of IOP over the follow-up period on visual field progression. We propose that increased LC depth indicates an increased IOP-related stress on the ONH, which is likely to be related to a worse disease prognosis. In the present study, however, there was no difference in the reduction of the VF MD between the groups with LC redisplacement and without LC redisplacement during the study period. However, it is well known that VF data fluctuate considerably and that VF changes should therefore be compared only with multiple examinations. Further, the LC displacement group in the current study included only nine patients; thus the comparison between the two groups had only limited power to detect group differences. The current study was not designed to investigate the influence of LC change on future visual field changes, and this relationship remains to be elucidated. A longitudinal study is currently under way to investigate the relationship between the reversal of the LC displacement and the rate of disease progression. 

In the present study, BMO level was used as the reference plane. Although the BMO opening is a solid structure that can be identified consistently among patients, there is a possibility that the BMO level moves due to choroidal thickness change or retinal edema. In eyes with long-lasting hypotony, potential thickening of the choroid may move the BMO anteriorly, resulting in an artifactual increase of LC depth. Such an effect can be more significant in patients of younger age who are more susceptible to hypotony maculopathy or edema at low IOP. Thus, the association of LC redisplacement with younger age may be partly attributable to this artifact. However, we believe that such artifact was not a major component of LC depth increase seen in our patients, for the following reasons. First, none of the eyes included had signs of ocular hypotony at the time of SD-OCT scanning. Second, the extent of LC redisplacement was generally much larger than the potentially overlooked choroidal thickness change as shown in sample figures (Fig. 3). Third, the IOP at postoperative 24 months of the nine patients who displayed LC redisplacement was equal to or higher than the measurement at postoperative 6 months, ranging from 10 to 30 mm Hg (data not presented). 

In the present study, younger patients were more susceptible to LC redisplacement. This finding can be attributed to the following factors. First, the LC is probably less stiff in younger patients, and thus the extent of LC responses to varying IOP may be larger. Second, as discussed above, there is a higher likelihood that the choroid thickens in eyes with long-lasting low IOP, leading to anterior displacement of the Bruch's membrane. This may increase the distance between the BMO and the anterior LC surface. 

This study was subject to some limitations. First, the LC depth was measured in the central area because the LC was often not clearly visible in the peripheral area. However, it may be assumed that given the posteriorly bowed nature of the LC, the LC displacement reversal may actually occur largely in the central region. Thus, we believe that evaluation of the central LC area may be an effective way of demonstrating the changes occurring with IOP variation. 

Second, parameters potentially relevant to the resilient response of the LC, such as the material properties of the laminar and peripapillary scleral connective tissue, ^1,15,17,23 were not considered in this study. Unfortunately, those parameters are not currently measurable. It is possible that such factors play a significant role in the reduction of LC displacement and the increase in the thickness of the LC and prelaminar tissue. 

Lastly, our study included many young, myopic patients with high preoperative IOP. It is possible that the LC response in these patients is different from those in elderly patients with modestly high IOP, who comprised typical OAG patients who underwent trabeculectomy. 

In conclusion, we have described herein the long-term follow-up outcome of trabeculectomy on the LC depth in 28 patients with POAG. The LC depth change following the initial 6 months after surgery varied among the patients; younger age, higher IOP at final follow-up, greater IOP fluctuation, and a higher mean follow-up IOP were associated with an increase in LC depth during the longer-term postoperative follow-up period. Further studies should determine the influence on disease prognosis of the reversal of LC displacement that occurs following IOP reduction. 

Acknowledgments

Supported by a National Research Foundation of Korea grant funded by the Korean Government (2010-0004210) and in part by an unrestricted grant from Research to Prevent Blindness (New York, NY). RNW has received instruments from Heidelberg Engineering for use in research. The authors alone are responsible for the content and writing of this paper. 

Disclosure: E.J. Lee, None; T.-W. Kim, None; R.N. Weinreb, Carl Zeiss Meditec (F, C), Heidelberg Engineering (F), Optovue (F, C), Topcon (F), Nidek (F)