Reviewed in this study were 50 PAC eyes of 50 patients who had undergone uneventful cataract surgery. The medical records of all subjects evaluated by one glaucoma specialist (KRS) from January 2011 to April 2017 in the glaucoma clinic of the Asan Medical Center in Seoul, Korea, were retrospectively examined. Previously, we reported that PAC and PACG do not differ in terms of anterior segment parameters [16]. Therefore, eyes with PAC or PACG were pooled, and both were included in the current analysis. The presence of PAC or PACG was determined on the basis of gonioscopic examination. Specifically, PAC was diagnosed when an eye had an occludable angle and exhibited features indicative of trabecular obstruction by the peripheral iris (elevated intraocular pressure [IOP], peripheral anterior synechiae, iris whorling [distortion of the radially orientated iris fibers], “glaukomflecken” lens opacity, or excessive pigment deposition on the trabecular surface) without a glaucomatous optic disc or any visual field change [17]. PAC eyes showing glaucomatous optic disc changes (e.g., neuroretinal rim thinning, disc excavation, and/or optic disc hemorrhage due to glaucoma) or a glaucomatous visual field change (pattern standard deviation <5% and values outside normal limits in the glaucoma hemifield test) were considered to have PACG [17]. Only reliable visual field test results (false-positives <15%, false-negatives <15%, and fixation loss <20%) were included in the analysis. Subjects with secondary angle closure due to an intumescent lens, uveitis, ocular trauma, choroidal effusion, suprachoroidal hemorrhage, medication (e.g., topiramate use), or previous intraocular surgical procedure were excluded from this study. In addition, individuals with past or present ocular disease other than PAC or mild cataract were excluded.

The study protocol was approved by the institutional review board of the Asan Medical Center (2017-0424), and the study design followed the principles of the Declaration of Helsinki. Informed consent was waived due to the retrospective nature of the study.

Initial testing comprised a comprehensive ophthalmologic examination, including a review of the medical history, measurement of best-corrected visual acuity, performance of slit-lamp biomicroscopy, and repeated IOP measurement by Goldmann applanation tonometry (Haag-Streit AG, Koeniz, Switzerland). Preoperative and postoperative refractive errors were measured with manifest refraction. Anterior segment optical coherence tomography (AS-OCT) imaging (Visante ver. 2.0; Carl Zeiss Meditec, Jena, Germany) was performed to measure anterior chamber depth (ACD), LV, central corneal thickness, and angle-opening distance at 500 µm (AOD 500).

Preoperative IOL power calculations were performed with the IOLMaster Optical Biometer (Carl Zeiss Meditec). Predicted refractive outcomes were calculated using the Haigis, SRK/II, and SRK/T formulas [18]. The postoperative refractive errors measured at one month after surgery were analyzed. Mean refractive error (MRE) was calculated as the difference of spherical equivalent between predicted and actual postoperative refraction values assessed at 1-month post-surgery for each of the three IOL calculation formulas. Mean absolute error (MAE) indicated the absolute number of calculated MRE values.

The affected eyes of all participants were imaged at the nasal and temporal angles using AS-OCT (Visante ver. 2.0) operating in the enhanced anterior segment single mode (scan length 16 mm, 256 A-scans). The anterior segment parameters in each image were evaluated by an independent examiner (WKS), who was blinded to all other test results and the clinical information of the participants. All AS-OCT parameters were determined using the Image J software ver. 1.46 (National Institutes of Health, Bethesda, MD, USA). ACD was defined as the maximum distance, measured perpendicularly to the horizontal line connecting the two scleral spurs, between the posterior corneal surface and the anterior lens surface. LV was defined as the perpendicular distance between the anterior pole of the crystalline lens and the horizontal line joining the two scleral spurs. AOD 500 was defined as the linear distance between the point of the inner corneoscleral wall (500 µm anterior to the scleral spur) and the iris [19,20].

All patients underwent spectral-domain optical coherence tomography (SD-OCT; Spectralis, Heidelberg Engineering, Heidelberg, Germany) imaging following pupil dilatation. Choroid imaging was obtained with enhanced depth imaging technique. A single independent ophthalmologist (JWS), blinded to the clinical diagnosis, performed measurements of the macular regional CT, whereby horizontal and vertical scans passing directly through the center of the fovea were selected. The resulting images were viewed and measured with the Heidelberg Eye Explorer ver. 1.5.12.0 (Heidelberg Engineering). The subfoveal CT was defined as the distance from the outer portion of the hyperreflective line corresponding to the RPE to the inner surface of the sclera. Measurements were performed at the subfovea and at 1 and 3 mm from the fovea superiorly, inferiorly, temporally, and nasally (Fig. 1A, 1B) [21,22,23]. Postoperative subfoveal CT was acquired at one month postoperatively using the same method.

All of the procedures were performed by one surgeon (KRS). A temporal corneal incision, continuous curvilinear capsulorrhexis, hydrodissection, and phacoemulsification with the OZil torsional system (Infiniti; Alcon, Fort Worth, TX, USA) were performed. A foldable posterior chamber IOL was implanted in the capsular bag. The IOLs used in the study were one-piece acrylic versions (Hoya iSert 250; Hoya Surgical Optics, Tokyo, Japan). No intraoperative complications occurred. The power of the implanted IOL was calculated with the preoperative biometric data and the IOL power formula for each patient. The predicted refraction values were estimated using the calculated power of the implanted IOL.

All statistical analyses were performed using commercial software (IBM SPSS Statistics ver. 21.0; IBM Corp., Armonk, NY, USA). The one-way analysis of variance and Bonferroni post-hoc tests were used to compare the predictive accuracy of the three IOL calculation formulas. Univariate and multivariate regression analyses were performed to evaluate the influence of preoperative biometric factors on refractive outcome and correlation between preoperative and postoperative subfoveal CT values and refractive outcome. In the multivariate analysis, variables that retained significance at p < 0.1 in the univariate analysis were included. Probability values with p < 0.05 were considered to be statistically significant.

Fifty eyes of 50 angle closure patients who had undergone uneventful cataract surgery were included in this study. The preoperative and postoperative clinical data are summarized in Table 1. The mean age was 69.4 ± 8.3 years, and 42 of the 50 subjects were female. Of the 50 subjects, 17 (34.0%) had undergone LPI before cataract surgery.

The comparison of predicted and actual postoperative 1-month refractive errors showed a hyperopic shift for the SRK/T (MRE, 0.199 ± 0.567 diopters [D]) and Haigis (MRE, 0.190 ± 0.727 D) formulas and a myopic shift for the SRK/II formula (MRE, −0.077 ± 0.623 D) (Table 2). The MREs of the SRK/II formula from the IOLMaster Optical Biometer (Carl Zeiss Meditec) were significantly lower than those of the other two formulas, indicating more myopic prediction with this formula (p < 0.001). The MREs of the SRK/T and Haigis formulas were not statistically different from each other (p = 1.00).

In contrast with the MRE results, the MAE results indicated that there were no significant differences among all three formulas (MAE, 0.512 ± 0.364 D for SRK/II; MAE, 0.512 ± 0.328 D for SRK/T; MAE, 0.593 ± 0.461 D for Haigis) (Table 3). The MAEs of the SRK/T formula from the IOLMaster Optical Biometer tended to be higher than those of the Haigis formula, but not in significantly significant manner (p = 0.083).

The associations between the considered preoperative biometric factors and MRE values were investigated using univariate and multivariate regression analysis. The univariate and multivariate analysis results according to the different IOL calculation formulas (SRK/II, SRK/T, and Haigis) are provided in Tables 4, 5, and 6.

Using the MREs calculated by the three formulas, univariate regression analysis showed that higher LV and shallower ACD were associated with a more significant hyperopic shift of the refractive outcome, but this association was not observed in multivariate regression analysis (backward variable elimination) (LV SRK/II: β = −0.152, p = 0.323; SRK/T: β = −0.135, p = 0.368; Haigis: β = −0.140, p = 0.373 and ACD SRK/II: β = −0.215, p = 0.150; SRK/T: β = −0.113, p = 0.432; Haigis: β = −0.128, p = 0.398) (Table 3, 4, 5). In addition, a thicker preoperative subfoveal choroid was associated with a more significant myopic refractive shift after cataract surgery. When MREs calculated by the three IOL calculation formulas (SRK/II, SRK/T, and Haigis) were incorporated into univariate and multivariate analyses in the same manner, thicker preoperative CT showed strong negative correlation with MRE (SRK/II: β = −0.511, p < 0.001; SRK/T: β = −0.652, p < 0.001; Haigis: β = −0.671, p < 0.001) (Table 3, 4, 5).

We observed significant linear correlations between MRE and subfoveal CT with the SRK/II, SRK/T, and Haigis IOL calculation formulas. Using linear regression analysis, the association between subfoveal CT and postoperative MRE showed a more myopic refractive shift after cataract surgery (SRK/II: r² = 0.350, p < 0.001; SRK/T: r² = 0.502, p < 0.001; Haigis: r² = 0.450, p < 0.001) (Fig. 2A-2C).

To clarify the relationship between preoperative subfoveal CT and the amount of refractive error after cataract surgery, we counted the number of patients with the cutoff criteria of a subfoveal CT of 300 µm and a MAE value of 0.5 D. Using Pearson's chi-square method, patients with greater than or equal to 300 µm of CT were highly correlated with a higher MAE value of 0.5 D in the SRK/T and Haigis formula groups, which indicates the higher absolute value of the refractive error (odds ratio, 1.970; Fisher exact test, p = 0.014 for SRK/T; odds ratio, 2.364; Fisher exact test, p = 0.005 for Haigis) (Table 7).

We conducted a subgroup analysis of 16 patients who underwent postoperative subfoveal CT evaluation. In subgroup analysis, postoperative subfoveal CT was significantly thinner than preoperative subfoveal CT (preoperative group, 258.88 ± 66.03 µm; postoperative group, 254.81 ± 61.88 µm; p = 0.024).

Using the MREs calculated by the three formulas, univariate regression analysis showed that a larger degree of postoperative reduction of subfoveal CT was significantly associated with a more significant amount of myopic shift after cataract surgery in all formulas (SRK/II: β = 0.579, p = 0.0.019; SRK/T: β = 0.818, p < 0.001; Haigis: β = 0.619, p = 0.011). In addition, the association between postoperative subfoveal choroidal thinning and postoperative MRE showed a more significant myopic refractive shift after cataract surgery (SRK/II: r² = 0.335, p = 0.019; SRK/T: r² = 0.669, p < 0.001; Haigis: r² = 0.384, p = 0.011) (Fig. 3A-3C).