Anterior Chamber Depth, Iridocorneal Angle Width, and Intraocular Pressure Changes After Phacoemulsification:  Narrow vs Open Iridocorneal Angles FREE

It is well known that narrowing of the anterior chamber angle (ACA) is the fundamental mechanism for the development of acute primary angle closure and chronic primary angle-closure glaucoma.^1- 2 Studies^3- 4 based on advanced imaging have demonstrated that the crystalline lens has an important role in the narrowing of the angle by pushing the peripheral iris anteriorly, in particular with the cataractous lens. Lens extraction for angle-closure and associated glaucoma has increasingly become a plausible alternative and is gaining popularity, especially in Asia.^5- 9

Numerous studies^10- 18 have demonstrated that intraocular pressure (IOP) is reduced after phacoemulsification with intraocular lens (IOL) implantation in eyes with narrow-angle (NA) glaucoma and open iridocorneal angles. The amount of IOP decrease seems to be inversely related to preoperative anterior chamber depth (ACD).^19- 20 The correlation between narrow iridocorneal angle widening and IOP reduction after cataract surgery has not been thoroughly investigated.

Anterior segment optical coherence tomography (AS-OCT) has been shown to be effective in assessing ACA structures. Examples of ACA variables that can be assessed by AS-OCT include angle opening distance (AOD), ACD, trabecular iris surface area, and angle recess area, often by using customized software.^21- 24

We conducted a prospective study to evaluate the effect of cataract extraction by phacoemulsification on the anatomy of the drainage angle in narrow iridocorneal angle eyes. The results were compared with those in eyes with open iridocorneal angles that had the same procedure. The association of IOP reduction after phacoemulsification and the change in the iridocorneal angle width as assessed by AS-OCT was also evaluated.

Institutional review board approval was obtained from the University of California, San Francisco, Committee on Human Research. Written informed consent was obtained from all the participants. Patients were recruited from the glaucoma service at the University of California, San Francisco, between March 1, 2009, and July 31, 2010. Eyes with open angles (OAs) were defined as those with Shaffer grades of 3 or 4 in 3 or all 4 quadrants. Eyes with NAs were defined as those with Shaffer grades of 2 or less in 3 or all 4 quadrants, as described previously.²⁵ Patients who had previously undergone laser peripheral iridotomy were recruited if they met the criterion of NA. All the patients possessed visually significant cataracts and had best-corrected visual acuity of less than 20/40 in the affected eye. The exclusion criteria included previous penetrating surgery, complications related to the cataract surgery (such as posterior capsular rupture and vitreous loss, primary or secondary glaucoma, and peripheral anterior synechiae), glaucomatous optic neuropathy, topical glaucoma therapy, and loss to follow-up. Optic nerves graded as having a cup-disc ratio greater than 0.6 (vertical meridian) were also excluded from this study. To maintain image quality, eyes with substantial corneal abnormality²⁶ (such as edema, dystrophy, abrasion, marginal degeneration, or pterygium) were also excluded. Glaucoma was diagnosed according to the International Society for Geographical and Epidemiological Ophthalmology.²⁷

Preoperative evaluation included slitlamp examination, visual acuity testing, routine fundus examination, gonioscopy, and IOP determination. The IOP was measured using Goldmann applanation tonometry by a single observer (S.C.L.). We checked patients in a narrow period of the day (1-3 PM) to avoid the effect of diurnal IOP fluctuation.^28- 29 Two values were assessed, and the average value was used for analysis. If the 2 values differed by more than 2 points, then a third value was obtained, and the middle figure was chosen as the measurement. Gonioscopy was performed using a Zeiss-style 4-mirror gonioscopic lens (model OPDSG; Ocular Instruments Inc, Bellevue, Washington) by a single glaucoma specialist physician (S.C.L.) in a dark room. Angles were graded based on the Shaffer method in all 4 quadrants (superior, nasal, temporal, and inferior). Funduscopy was performed using +90- and +20-diopter lenses. When funduscopy was impossible, the posterior segment was evaluated using B-scan ultrasonography. Follow-up time points included preoperative and 10 days and 1, 3, and 6 months after surgery.

Images of the anterior segment were obtained using a commercially available AS-OCT device (Visante OCT; Carl Zeiss Meditec Inc, Dublin, California) by 2 experienced operators (G.H. and E.G.) who were masked to the results of the clinic ophthalmic examinations. A detailed description of the AS-OCT's functionality has been previously discussed.^30- 31 Standard-resolution scans captured the temporal and nasal quadrants (nasal-temporal 0°-180°) in 1 image with participants looking straight ahead and having a good central corneal reflex. All the images were taken in the same dark conditions (0-1 lux by digital light meter [EasyView model EA30; Extech Instruments Inc, Waltham, Massachusetts]) with patients in a sitting position. After several scans were acquired, the operator selected the best image with no motion artifacts or image artifacts from the eyelids. Assessment of the superior and inferior quadrants often requires manual manipulation of the eyelids, which may distort the angle. To prevent systematic bias in angle assessment of groups of patients who may require eyelid manipulation, only images of nasal and temporal quadrants were included in this study.³²

Images were analyzed using the Zhongshan Angle Assessment Program (ZAAP, Guangzhou, China), which has been shown to have good reproducibility for iris measurements.^22- 24 Initial attention was focused on quality control of the Fvisu, especially for NAs.³³ Images in which we could not clearly detect the scleral spur were removed from the analysis because quantitative evaluation of the anterior chamber variables by AS-OCT depends on correctly identifying the scleral spur as the landmark.^34- 35 To prevent mechanical influence on the angle due to the temporal clear corneal tunnel incision of phacoemulsification for cataract, only the nasal quadrant image of AS-OCT was chosen to be analyzed in this study.

All the operations were performed by the same surgeon (S.C.L.) using conventional surgical procedures. In brief, eyes were prepared for surgery by instilling tropicamide, 0.5%, and phenylephrine hydrochloride, 10%, for pupil dilation, and tetracaine hydrochloride, 0.5%, for topical anesthesia. All the surgical procedures were undertaken using a 3.2-mm temporal clear corneal tunnel incision. After the incision, the continuous curvilinear capsulorrhexis measuring approximately 5.5 mm in diameter was performed using a cystitome and Utrata forceps. Hydrodissection/hydrodilation, in-the-bag phacoemulsification using the divide-and-conquer technique, cortical aspiration, and insertion of a foldable acrylic IOL in the capsular bag were performed step by step. At the end of the operation, the surgeon always confirmed that the IOL was accurately implanted in the capsular bag. All the eyes received a single-piece acrylic IOL (model SA60AT; Alcon Laboratories, Fort Worth, Texas).

All the data are reported as mean (SD). Longitudinal changes in ACD, AOD at 500 μm anterior to the scleral spur (AOD500), IOP, and the other continuous variables between 2 groups were compared statistically using linear mixed models. We also used linear mixed models to compare preoperative and postoperative values of each variable. In each case, before testing for differences between preoperative and postoperative values, we first tested whether the postoperative values of the variable under consideration showed statistically significant differences. Specifically, we fit models including an effect for each time point (10 days and 1, 3, and 6 months) and compared these with a model that collapsed all postoperative time points and contrasted these using the likelihood ratio test. We also compared the NA and OA groups within each time stratum. Differences in the preoperative and postoperative groups were assessed by testing the significance of the interaction term between time and group. Correlations between measurements taken on each eye over time and between the 2 eyes of each patient were addressed by the inclusion of random effects. Hypothesis testing was conducted using the likelihood ratio test. Differences between preoperative and postoperative values were analyzed and compared using the paired t test. P < .05 was considered statistically significant. Data analysis was conducted using a commercially available statistical package (R version 2.12 for MacIntosh, R Foundation for Statistical Computing, Vienna, Austria).

The present study comprised 76 eyes of 49 consecutive patients who underwent cataract surgery; 6 of 32 eyes in the NA group and 5 of 42 eyes in the OA group were excluded because of the inability to detect the nasal scleral spur. Two eyes in the NA group had vitreous loss and were excluded from the study. Of the 63 eyes remaining for analysis, 26 were NA and 37 were OA. Sixteen eyes had undergone laser peripheral iridotomy and met the criterion of “narrow” in the NA group. Axial length in the NA group was significantly shorter than that in the OA group. No significant differences were related to age, sex, right and left eye ratio, and cumulative dissipated energy between groups. Table 1 provides the patient demographics and clinical data for each group.

Table 2 summarizes the mean ACD deepening induced by cataract surgery and posterior chamber IOL implantation. Before surgery, the mean (SD) ACD was 2.23 (0.07) mm in the NA group and 2.76 (0.08) mm in the OA group. The preoperative ACD in the NA group was significantly shallower than that in the OA group (P < .001). After cataract surgery, the ACD deepened significantly in both groups (P < .001 in both groups), and thereafter no significant changes were observed throughout the postoperative period in any group (P >> .05). The NA group had smaller values but deepened 0.313 (0.05) mm more than did the OA group (P < .001). Six months after surgery, the mean (SD) deepening of the ACD was 1.52 (0.04) mm in the NA group and 1.20 (0.06) mm in the OA group. Six months after surgery, the mean (SD) postoperative ACD was 3.75 (0.05) mm in the NA group and 3.95 (0.05) mm in the OA group, showing a significant difference between the 2 groups. Changes in ACD were significantly related to their respective preoperative variables in both groups (slope = −0.658 mm/mm in the NA group, P < .001; slope = −0.690 mm/mm in the OA group, P < .001). We found no evidence that the 6-month change per millimeter of preoperative ACD differed between the groups (P = .87). Figure 1 shows the negative correlation between ACD deepening postoperatively and preoperative ACD; that is, shallower depth before surgery was associated with greater ACD increase after surgery. Although the relationship between ACD deepening and IOP change was not significant (P = .08), the trend was toward an inverse correlation (decrease of 0.083 mm Hg per 0.1-mm increase in ACD) of those 2 factors.

Table 3 gives the mean ACA widening after cataract surgery. The mean preoperative width of AOD500 in the NA group was found to be significantly shallower than that in the OA group (P < .001). Immediately after cataract surgery, the ACA widened significantly (P < .001), and no further significant change occurred throughout the postoperative period in each group (P >> .05). The NA group had lower values, but they increased more. The mean (SD) change in AOD500 in eyes with NAs was 0.062 (0.021) mm larger than that in eyes with OAs (P = .003), and the percentage of widening in the NA group was almost twice that in the OA group. Six months after surgery, changes in AOD500 were significantly related to their respective preoperative variables in both groups (slope = −0.766 mm/mm, P < .001; slope = −0.785 mm/mm, P < .001; no evidence of a difference was seen in the slope in the NA group vs the OA group, P = .95). Figure 2 shows the correlation between postoperative ACA widening and preoperative ACA width in the 2 groups.

The mean IOPs and the changes in IOP are given in Table 4. Before surgery, no significant difference in IOP was noted between the 2 groups. After surgery, the mean IOP decreased significantly in both groups, and no further significant changes were found throughout the postoperative period in each group compared with 10 days after surgery. The NA group had higher IOP but dropped more so that the IOPs after surgery were approximately the same. The mean (SD) IOP reduction in the NA group was 1.07 (0.369) mm Hg more than that in the OA group after surgery (P = .004). Six months after surgery, the mean (SD) amount of IOP reduction was 2.75 (0.60) mm Hg (17.82%) in the NA group and 1.55 (0.47) mm Hg (9.60%) in the OA group (P = .004). The percentage of IOP reduction in the NA group was almost twice that in the OA group 6 months postoperatively.

Finally, we assessed the relationship between AOD and IOP using a linear mixed-effects model with time and the AOD500 at each time point, together with interaction terms (and random effects to include the correlation between measurements longitudinally on the same eye and between the eyes of a given person). We found no evidence of interaction or of a difference between the postoperative times. In the NA group, we found that each 0.1-mm increase in AOD corresponded to a mean (SD) 0.47 (0.12)−mm Hg decrease in IOP (averaging across all 4 postoperative measurements) (P < .001) but that there was a significant effect of time as well (not all the change in IOP is attributable to changes in AOD). Restricting the analysis to the 6-month postoperative visit only (compared with preoperative value), the same results were found; in the NA group each 0.1-mm increase in AOD resulted in a mean (SD) decrease of 0.42 (0.18) mm Hg in IOP (P = .001). In the OA group, each 0.1-mm increase in AOD corresponded to a mean (SD) decrease of 0.32 (0.16) mm Hg (P = .046).

Axial length (from IOLMaster [Carl Zeiss Meditec Inc]) was not associated with IOP change (P = .33) or angle widening (P = .84). Figure 3 shows the anterior chamber configuration before and after surgery in NA and OA eyes.

The results of this study suggest that phacoemulsification with IOL implantation results in deepening of the central anterior chamber and widening of the ACA in NA and OA eyes based on quantitative assessment of AS-OCT imaging. Furthermore, these results suggest a correlation between IOP reduction and angle widening in the 2 groups. To our knowledge, this is the first study to demonstrate that the postoperative reduction in IOP is related to the increase in angle width after phacoemulsification for NA and OA eyes.

Cataract extraction and IOL implantation generally result in significant lowering of IOP in glaucomatous and nonglaucomatous eyes.^10- 18 Wide variation in mean IOP reductions (1.1-13.5 mm Hg) have been reported in such studies.^10- 18 The present study also found wide variation in the IOP response in the study cohort. We found that the postoperative IOP was reduced significantly in both groups. Six months after surgery, the mean (SD) IOP reduction was 2.75 (0.60) mm Hg (17.82%) in the NA group and 1.55 (0.47) mm Hg (9.60%) in the OA group, which were significantly different between the 2 groups (P = .004). Phacoemulsification with IOL implantation seems to reduce IOP more in NA eyes than in OA eyes. There is potential clinical relevance of these findings to the treatment of patients with glaucoma. Patients with glaucoma who have more NAs before phacoemulsification may have greater IOP lowering afterward and a greater possibility of discontinuing 1 or more medications postoperatively. Perhaps a further clinical implication is that preoperative angle assessment is helpful in predicting the IOP benefit of cataract extraction.

The exact mechanism of this IOP reduction after cataract surgery is still not fully understood. Hypothetically, cataract surgery removes the anatomical cause of NAs, resulting in deepening of the ACD and widening of the ACA. It would be expected that access of aqueous to the filtering portion of the trabecular meshwork would be improved due to widening of the drainage angle. The results of this study support this scenario as a likely contributory mechanism for IOP lowering. In both groups, the IOP reduction was statistically significantly associated with angle widening as assessed by AOD500, with the IOP reduction per 0.1-mm increase in AOD500 greater in NA eyes than in OA eyes. Nolan et al³⁶ investigated the changes in angle configuration after phacoemulsification in 21 patients using AS-OCT and found that the mean AOD500 for the nasal quadrant (in dark conditions) increased from 243 to 457 μm, similar to the findings in the present OA group. Their study cohort included 7 eyes with iridotrabecular contact or peripheral anterior synechiae in 1 or more quadrants and 14 OA eyes. However, IOP information was unavailable to correlate with the degree of angle opening. In the present study, there was a trend toward greater IOP lowering with more ACD deepening, although it did not reach statistical significance (P = .08). Previous studies^19- 20 have supported a significant relationship. This slight discrepancy may be related to the mixed population of open and closed angles and the fact that the angle opening is a more significant factor than is central ACD increase.

Studies have also focused on changes in angle configuration after cataract surgery using either Scheimpflug or ultrasound biomicroscopic images and support the finding that IOP reduction is greater in angle-closure eyes. Hayashi et al¹⁰ demonstrated by Scheimpflug imaging that the width and depth of the ACA of angle-closure glaucoma and OA glaucoma increase significantly after cataract extraction and IOL implantation. They also found that the IOP decreased significantly after surgery and that the amount of reduction in the angle-closure glaucoma group was higher than that in the OA glaucoma and OA groups. However, no association was noted between ACA widening or deepening and IOP decrease in this study. Clinically, though, it is difficult to avoid image artifacts because the light of the rotating Scheimpflug camera often cannot penetrate the corneosclera limbus.^37- 38 A study based on ultrasound biomicroscopy by Tham et al³⁹ demonstrated that the mean AOD500 in pseudophakic primary angle-closure glaucoma eyes approached or exceeded the reported value in normal phakic eyes. The present findings agree with these studies, although they are based on different technologies for measurement of the anterior chamber configuration.

Although angle opening is an apparent mechanism that can partially account for IOP lowering, there are other potential contributory factors, such as ultrasound activation of cytokines, endogenous prostaglandin F₂ release, and increase in aqueous outflow by expansion of the trabecular meshwork and lumen of the Schlemm canal. Wang et al⁴⁰ demonstrated that phacoemulsification ultrasound activates the interleukin 1α/nuclear factor kappa B/endothelial leukocyte adhesion molecule 1 pathway, facilitating aqueous outflow and reduction in IOP. Mathalone et al⁴¹ suggested that the endogenous prostaglandin F₂ released postoperatively may enhance uveoscleral outflow. Theoretically, postoperative shrinkage of the lens capsule can result in increasing posterior traction on the scleral spur, expanding the trabecular meshwork and lumen of the Schlemm canal.^12,42- 43

This study has important limitations. The cohorts were relatively small, with only 26 eyes that met the criterion for the NA group and 37 eyes in the OA group. In addition, a longer follow-up duration would provide more valuable evaluation of the long-term effects of phacoemulsification with IOL implantation in eyes with NAs.

In summary, phacoemulsification with foldable IOL implantation can significantly deepen the ACD, widen the anterior chamber drainage angle, and lower IOP in NA and OA eyes. The amount of IOP reduction was approximately 18% in NA eyes and 10% in OA eyes. The postoperative reduction in IOP was proportional to the increase in angle width in NA and OA eyes. Future studies will further help to elucidate the long-term relationship of the anterior chamber configuration and IOP after phacoemulsification with foldable IOL implantation.