Abstract
Purpose.:
To study changes in lamina cribrosa position and prelaminar tissue thickness (PTT) after surgical IOP reduction in glaucoma patients.
Methods.:
Twenty-two patients (mean age, 71.4 years) were imaged with spectral domain optical coherence tomography (SD-OCT; 24 radial B-scans centered on the optic nerve head [ONH]) before trabeculectomy or tube shunt implantation. Follow up images were acquired 1 week, 1 month, 3 months, and 6 months postsurgery. Bruch's membrane opening (BMO), the internal limiting membrane (ILM) and the anterior laminar surface (ALS) were segmented in each radial scan with custom software. Surfaces were fitted to the ILM and ALS with the extracted three-dimesional coordinates. PTT was the distance between the ILM and ALS, perpendicular to a BMO reference plane. Serial postsurgical laminar displacement (LD), relative to the BMO reference plane, and changes in PTT were measured. Positive values indicated anterior LD.
Results.:
Mean (SD) presurgery IOP was 18.1 (6.5) mm Hg, and reduced by 4.7 (5.5), 2.4 (7.7), 7.0 (6.2), and 6.8 (7.5) mm Hg at 1 week, 1 month, 3 months, and 6 months postsurgery, respectively. At the four postsurgery time points, there was significant anterior LD (1.8 [9.5], −1.1 [8.9], 8.8 [20.2], and 17.9 [25.8] μm) and PTT increase (1.7 [13.3], 2.4 [11.9], 17.4 [13.7], and 13.9 [18.6] μm). LD was greater in ONHs with larger BMO area (P = 0.01) and deeper ALS (P = 0.04); however, PTT was not associated with any of the tested independent variables.
Conclusions.:
Both anterior LD and thickening of prelaminar tissue occur after surgical IOP reduction in patients with glaucoma.
Twenty-two patients with open-angle glaucoma from the Queen Elizabeth II Health Sciences Centre in Halifax, Nova Scotia were enrolled in this prospective observational study. Patients were scheduled for glaucoma surgery in the referral practices of two of the authors (LMS and MTN) and underwent SD-OCT imaging pre- and postsurgery. The research adhered to the tenets of the Declaration of Helsinki and subjects gave informed consent to participate in the study. The protocol was approved by the Capital Health Research Ethics Board.
Inclusion criteria were (1) a clinical diagnosis of open-angle glaucoma with documented progressive optic disc change and/or visual field damage compatible with glaucoma, (2) best corrected visual acuity equal to or better than 0.3 (20/40) logarithm minimum angle of resolution in the study eye, and (3) refraction within ±6.00 diopters (D) sphere and ±3.00 D astigmatism. Exclusion criteria were (1) concomitant ocular disease, and (2) systemic medication known to affect the visual field. Only one eye of each patient was included in this study.
The glaucoma surgical procedures were either trabeculectomy or tube shunt implantation. Surgery was performed when there was one or both of the following indications: (1) documented and confirmed visual field and/or optic disc progression, and (2) IOP considered clinically too high for the level of glaucomatous damage. Suture lysis, bleb needling revision, 5-fluorouracil subconjunctival injection, and topical ocular hypotensive medications were used at the discretion of the treating surgeon postoperatively, as clinically indicated.
The segmentation of the 24 radial B-scans yielded a 3D point cloud describing the ILM, the ALS, and BMO. Surfaces were fitted to the ILM and ALS with the extracted 3D coordinates. A spline was fitted to the BMO points around the ONH to generate a closed curve within which the BMO area was computed. In addition, a best-fit plane representing the BMO points was derived creating a BMO reference plane (
Fig. 2). Laminar position measurements in pre- and postsurgery images were made relative to the BMO reference plane.
For each time point, PTT was defined as the mean perpendicular distance (direction normal to the BMO reference plane) between the overlapping en face segmented ILM surface and ALS. The mean perpendicular distance from the BMO reference plane to the ALS, measured within BMO, was termed the ALS height. The LD resulting from IOP change was computed as the mean difference between post- and presurgery ALS heights, with positive values indicating anterior displacement and negative values indicating posterior displacement. The PTT change resulting from IOP change was computed as the difference between post- and presurgery PTT. In order to compute and compare serial LD and PTT changes, only data from the overlapping en face segmented ILM surface and ALS inside BMO in the pre- and postsurgery images (i.e., those locations with common valid values at the x and y coordinates throughout the follow up) were analyzed.
Repeated measures ANOVA was used to evaluate changes in IOP and ONH parameters following surgery. Spearman's rank correlation coefficient ρ was used to test the relationship between LD and PTT change, and their relationship with IOP. Multiple regression analysis was performed to establish whether the independent variables of age, presurgery IOP, ALS height, PTT and BMO area, and mean postsurgery IOP were associated with the mean postsurgery LD and PTT change. Analyses were carried out with Matlab (MathWorks, Natick, MA) and SPSS Statistics v. 19 (IBM SPSS Statistics, Armonk, NY).
There were 13 (60%) men and 9 (40%) women of European ancestry in the study. The mean (range) age was 71.4 (47.0–88.2) years and the visual field mean deviation, −9.5 (−2.4 to −18.8) dB. Eighteen (81%) patients underwent trabeculectomy and four (19%) tube shunt implantation. None of the 22 patients had IOP less than 5 mm Hg at any time during the follow up, gross macular or optic disc edema after surgery, major surgical complications, or required additional surgery during the follow up period. There was a statistically significant IOP reduction over 6 months of follow up (
Table 1,
P < 0.01). The mean (SD) value of the mean absolute deviation of the BMO points from the fitted BMO reference plane was 15.4 (6.0) μm.
Table 1. Intraocular Pressure Characteristics Pre- and Postsurgery*
Table 1. Intraocular Pressure Characteristics Pre- and Postsurgery*
| | Presurgery | Postsurgery |
| 1 Wk | 1 Mo | 3 Mos | 6 Mos |
| IOP† | 18.1 (6.5) | 13.6 (6.1) | 15.6 (10.3) | 11.3 (4.3) | 11.4 (5.3) |
| Absolute IOP reduction | | 4.7 (5.5) | 2.4 (7.7) | 7.0 (6.2) | 6.8 (7.5) |
| % IOP reduction | | 22.8 (31.3) | 14.9 (44.1) | 35.8 (21.6) | 33.9 (29.3) |
The mean (SD) BMO area was 1.81 (0.42) mm2 presurgery, and was not statistically different postsurgery: 1.79 (0.39) mm2, 1.83 (0.38) mm2, 1.84 (0.43) mm2, and 1.81 (0.41) mm2, at 1 week, 1 month, 3 months, and 6 months, respectively (P = 0.30). The ALS area segmented in the study was on average 57 (18%) of the BMO area, and the overlap in the segmented en face ALS in the pre- and postsurgery images was 47 (16%).
The mean presurgery PTT was 232.0 μm, and increased significantly by 1.7 to 17.4 μm in the postsurgery follow up (
P = 0.02;
Table 2 and
Fig. 3). Additionally, postsurgery, the ALS displaced anteriorly with mean LD of 1.8 to 17.9 μm (
P < 0.01;
Table 2 and
Fig. 3). There was no difference between LD and PTT change at any of the postsurgery time points (
P > 0.12) and generally a weak negative correlation between these two variables (ρ from −0.19 to −0.56,
Fig. 4).
Table 2. Laminar Displacement and Prelaminar Tissue Thickness Characteristics*
Table 2. Laminar Displacement and Prelaminar Tissue Thickness Characteristics*
| | Presurgery | Postsurgery |
| 1 Wk | 1 Mo | 3 Mos | 6 Mos |
| LD | | 1.8 (9.5) | −1.1 (8.9) | 8.8 (20.2) | 17.9 (25.8) |
| PTT | 232.0 (117.4) | 227.9 (124.8) | 203.8 (99.3) | 235.9 (109.7) | 245.9 (117.7) |
| PTT change | | 1.7 (13.3) | 2.4 (11.9) | 17.4 (13.7) | 13.9 (18.6) |
In multivariate regression analyses, LD was positively associated with baseline BMO area (
P = 0.01;
Table 3) and negatively associated with baseline ALS height. PTT change was not associated with any of the studied variables (
Table 3). The models for LD and PTT yielded adjusted
R 2 values of 0.46 (
P = 0.01), and −0.13 (
P = 0.75), respectively.
Table 3. Results of Multivariate Regression Analyses of Mean Postsurgery Laminar Displacement and Prelaminar Tissue Thickness Change*
Table 3. Results of Multivariate Regression Analyses of Mean Postsurgery Laminar Displacement and Prelaminar Tissue Thickness Change*
| | LD | PTT Change |
| Coefficient | P | Coefficient | P |
| Age | −0.36 | 0.12 | −0.32 | 0.24 |
| Presurgery IOP | −0.80 | 0.12 | 0.81 | 0.18 |
| Presurgery ALS height | −0.04† | 0.04 | 0.04 | 0.10 |
| Presurgery PTT | 0.01 | 0.68 | 0.00 | 0.93 |
| Presurgery BMO area | 15.62‡ | 0.01 | 1.80 | 0.78 |
| Absolute IOP reduction | 0.26 | 0.70 | −0.74 | 0.37 |
A representative case example of a 62-year-old patient with anterior LD and PTT change is shown in
Figure 5. The presurgery IOP was 14 mm Hg and the postsurgery IOP was 9, 6, 5, and 8 mm Hg at 1 week, 1 month, 3 months, and 6 months, respectively. The LD and PTT change for this patient are indicated in
Figure 4.
Anterior displacement of the ONH surface resulting from acute reduction in IOP has been observed in adult glaucoma patients undergoing trabeculectomy
12–17 and in a subset of patients with a large IOP reduction after medical treatment.
18,19 Because these studies used CSLT in which the ALS cannot be visualized, the source of ONH surface changes cannot be precisely elucidated. Backward bowing of the lamina has long been recognized as a characteristic feature of glaucomatous optic neuropathy.
30–32 It can be speculated that with postsurgical IOP reduction, anterior repositioning of the posteriorly displaced lamina and partial filling in of the optic disc cup occurs.
11 The advent of SD-OCT has made it possible to image the ALS in the presence of overlying prelaminar tissue and to quantify its displacement with IOP change.
27
The main finding of this study was that with mean IOP reduction of around 7 mm Hg at 3 and 6 months postsurgery, there was both prelaminar tissue thickening and anterior laminar displacement. On average PTT increased by 17 and 14 μm at 3 and 6 months postsurgery, respectively, while the corresponding average anterior LD was 9 and 18 μm.
Previous investigators hypothesized that anterior ONH surface changes after IOP reduction could be due to anterior movement of the lamina cribrosa or thickening of prelaminar tissue.
10,11,33 Our results support both these hypotheses, and indicate that anterior displacement of the lamina cribrosa alone cannot fully explain ONH surface changes. Prelaminar tissue thickening can occur due to an increase in blood volume with associated changes in blood flow,
34 or a shift of axoplasmic fluid from both the peripapillary retinal nerve fiber layer into prelaminar tissue and fluid that had been pushed downstream through the lamina at higher IOP levels.
3,5,35 Subclinical ONH edema is also a possible explanation for ONH surface changes following IOP reduction
36 ; however, since this most likely occurs in the early postoperative period, it probably does not explain the PTT changes persisting 6 months after surgery.
Previous research has demonstrated that the degree of ONH surface displacement is generally related to the degree of IOP change,
13–17 although the effects are variable
37 and may not be linear.
33,38 However, the various effects of IOP change on ONH structures described in experimental
27,33,38–40 and computer simulation
41–43 studies demonstrate that complex interactions between multiple variables are likely responsible for the observed results. Factors such as scleral
2,44 and laminar thickness
31,45 may be important to structural compliance. Recent work from Sigal and colleagues
46 suggests that the lamina does not respond to changes in IOP in isolation, but rather that the ONH and peripapillary sclera behave as a mechanical system, and that the magnitude of the final response of ONH structures to IOP may be blunted due to interactions of other factors to maintain the ONH as a robust structure.
47
In the present study, both LD and PTT change were more pronounced at 3 and 6 months postsurgery when there was maximal decrease in IOP. However, in multivariate regression analyses there appeared to be no statistically significant association between the degree of mean postsurgery IOP and either mean postsurgery LD or PTT change. LD was positively related to BMO area, indicating that the effects of IOP reduction on the lamina were more pronounced in ONHs with a larger neural canal opening. Additionally, there was greater anterior LD in ONHs with a deeper anterior lamina position relative to the BMO reference plane.
This study has some limitations. Because the entire ALS could not be visualized, largely because of the shadowing from the large vessels, it was not fully segmented. The LD values are reported from an average of 47% of BMO area, hence, it is possible that the lamina and prelaminar tissue in nonsegmented portions could behave differently under IOP modulation compared with those that were segmented. We could not consistently visualize the posterior laminar surface, making it impossible to assess information regarding deformation of the entire lamina. Although we had highly reliable lateral anatomical registration over time, the nature of serial SD-OCT examinations necessitates a relative axial reference plane. We chose the BMO as a reference plane to derive ALS height. Although the BMO area was not altered as a result of surgical IOP reduction, we could not determine whether the BMO plane was axially stable because all measurements were made relative to it. For this reason, we also calculated the LD relative to a peripheral ILM reference plane, akin to the reference ring used in the CSLT. These analyses yielded almost identical results to those obtained with a BMO reference plane, suggesting that there were no nonuniform changes in these two reference planes. Unlike LD, changes in PTT did not depend on a reference plane as PTT was computed as the mean perpendicular distance from the ILM to the ALS. While we performed analysis at the population level, our study lacked a control group of glaucoma patients who were matched for the examined baseline characteristics followed in the same manner to gauge the reproducibility of measurements within and between sessions at the individual patient level. The relatively small sample size could have resulted in a failure to statistically confirm the significance of other factors determining LD and PTT change.
In summary, both forward anterior laminar surface displacement and prelaminar tissue thickening occur in response to IOP decrease after glaucoma surgery. The changes are persistent to at least 6 months postsurgery.