Changes of Anterior Segment Parameters in Primary Angle Closure with Axial Myopia: a Retrospective Study of 369 Patients Grouped by Axial Length
Mengwei Li,Xiangmei Kong,Wenqing Zhu,Yuhong Chen,Junyi Chen,Xinghuai Sun
DOI: https://doi.org/10.1111/aos.13048
2016-01-01
Acta Ophthalmologica
Abstract:Myopia is uncommon in primary angle closure. A recent research showed that myopic primary angle-closure patients had no significant differences in anterior segment parameters compared with emmetropic and hyperopic counterparts (Yong et al. 2014). However, the refractive status of these patients is often misleading due to lenticular myopia, including cataract formation after the acute attack and age-related nuclear cataract. Therefore, refractive myopia might occur in short-axial eyes, which still reflects the biometric risk factor for these diseases. The real atypical primary angle closure is related to axial myopia rather than refractive myopia. We sought to establish whether anterior segment parameters would show significant differences when grouped by axial length (AL). This retrospective study was conducted of our primary angle-closure clinical database. Patients who received intra-ocular surgery or laser peripheral iridotomy or who had plateau iris or other ocular diseases were excluded. Primary angle-closure diseases were classified as primary angle-closure suspect (PACS), primary angle closure (PAC) and primary angle-closure glaucoma (PACG) based on the criteria proposed by Foster et al. (2002). We studied 369 patients (82 PACS, 150 PAC and 137 PACG). Anterior chamber depth (ACD), corneal curvature (CC) and AL were obtained by A-scan biometry (AL-3000 Biometer/Pachymeter; Tomey, Nagoya, Japan). Axial length (AL) was categorized as short (<22.5 mm), medium (≥22.5 to <23.5 mm) or long (≥23.5 mm). Data of left eyes were used for analyses. spss 19.0 was performed for all statistical analyses using Chi-square, anova and pairwise t-tests with Bonferroni adjustment. p values < 0.05 were considered statistically significant. Mean age of the patients was 60.8 ± 10.4 years, with most being female (n = 255, 69.1%). Short AL was present in 220 patients (59.6%), medium AL in 108 patients (29.3%) and long AL in 41 patients (11.1%). Across the angle-closure subgroups, there were no significant differences in mean age (p = 0.480) or gender ratio (p = 0.904). The proportions of angle-closure subgroups among different AL groups were similar (p = 0.821). Among the primary angle-closure patients, those with long AL had deeper ACD and flatter CC than those with medium (p = 0.029 for ACD; p = 0.001 for CC) or short AL (p < 0.001 for ACD; p < 0.001 for CC) (Fig. 1A). Significant correlations were found between AL and ACD (r = 0.273, p < 0.001) and between AL and CC (r = 0.576, p < 0.001) (Fig. 1B). Our findings (Fig. 1Cb) are contrary to those of Yong et al. (2014) who reported that there were no significant differences in ACD and CC in eyes with axial myopic angle closure by comparing different refractive status (Fig. 1Ca). Our results might explain the low incidence of myopia in primary angle-closure patients. Lowe (1970) found that 5.5% of primary angle-closure patients were myopic, whereas Chakravarti & Spaeth (2007) reported that 1.9% of 322 patients had high myopia. Additionally, our study might suggest different mechanisms of these atypical primary angle closure. Future research might assess other anterior segment parameters such as anterior chamber width, peripheral iris thickness and lens vault to further verify this. Our study had several limitations. Firstly, the retrospective nature may have led to bias. Secondly, lens thickness was not routinely measured. Thus, we could not directly evaluate lens element for angle closure. However, we did access ACD, which is regarded as the most cardinal factor for angle closure (Foster et al. 2000), and can indirectly represent lens condition. Moreover, AL and ACD were only weakly correlated, so larger cohort studies are needed to establish the clinical significance of such association.