Wang and Zhang
1 have raised a number of questions regarding our published article “Comparison of anterior segment parameters between the acute primary angle closure eye and the fellow eye.”
2 The first comment was a recommendation that we control for pupil diameter for the anterior segment measurements in acute primary angle closure (APAC) eyes and fellow eyes in statistical analysis. In our analysis, we included pupil diameter in the univariate regression model and found that it was not significant (
P = 0.472).
2 In the Methods section, we stated that only a
P value less than 0.2 in univariate analysis would be included in the multivariate analysis and, therefore, did not include pupil diameter in our multivariate analysis. We also compared pupil diameter between APAC and fellow eyes (Table 2 of our published article) and found that the difference was not significant (
P = 0.525).
2
The second comment was a recommendation that we cite the results of Sng et al.
3 and discuss the differences between their results and our findings. Unfortunately, we could not cite that report because we submitted our study to
Investigative Ophthalmology and Visual Science (
IOVS) before its publication. We submitted our original manuscript on August 7, 2013 and our revised manuscript on November 11, 2013, receiving the decision letter on April 21, 2014. The study by Sng et al.
3 was e-published on October 31, 2013 and, thus, was not available until we had almost completed our revised submission. Thus, we are grateful for the opportunity to discuss the excellent results of Sng et al.
3 in this response.
In the first instance, however, we do not accept that the results of Sng et al.
3 contradict our data. The main finding of Sng et al.
3 was that a shallower anterior chamber depth (ACD) and smaller iris curvature (I-Curv) were the two main anterior segment biometric parameters associated with APAC during an attack, as mentioned by Wang and Zhang.
1 Our results also showed that a shallow ACD and smaller I-Curv were more associated with APAC eyes than fellow eyes.
2 In our analysis, however, lens vault (LV) was found to be the most prominent feature, showing an extremely high odds ratio, even though a shallow ACD and small I-Curv still were significantly associated with APAC eyes. We now have carefully compared our results (Table 2)
2 with those of Sng et al. (Table 1)
3 in terms of anterior segment parameters and found that the numerical values in both studies are very similar. However, the LV in our APAC eyes (1.26 ± 0.36 mm) was higher than that reported by Sng et al.
3 (1.18 mm, 95% confidence interval [CI], 1.09–1.26), whereas the LVs of the fellow eyes were similar between the two studies (1.06 ± 0.31 mm in our study and 1.06 [0.96–1.16] mm in their study).
Although it is difficult to compare values between different studies directly, it seems that our participants had a greater difference in the LV between APAC eyes and fellow eyes (1.26 ± 0.36 vs. 1.06 ± 0.31 mm;
P = 0.017) than of the subjects analyzed by Sng et al.
3 (1.18 vs. 1.06 mm;
P = 0.041). In other words, our participants showed a greater difference between the APAC and fellow eye in terms of LV during an acute attack. This difference may explain why LV was the most prominent feature in our results. This result could be due to ethnic differences (our participants were all Koreans; those assessed by Sng et al.
3 were predominantly Chinese), but we have to interpret the results carefully, because our study and that of Sng et al.
3 included only a small sample size (our study, 36 participants; report of Sng et al.,
3 31 participants). In addition, since our participants were older on average (our study, 67.8 ± 8.2 years; report of Sng et al.,
3 60.9 ± 7.5 years), the role of the lens would likely be more significant in our cohort because LV is significantly affected by aging.
4
One interesting point to note is the possible effect of central corneal thickness (CCT) on ACD measurement. Acute IOP elevation causes corneal edema, which may decrease the ACD. Indeed, we found in our analysis that the CCT was significantly greater in APAC eyes than in fellow eyes (Table 2)
2 and showed a significant association with APAC eyes in univariate and multivariate analyses. Increased CCT and a shallower ACD are evidenced by our data in Figure 3.
2 The ACD is influenced by LV, but also may be affected by the CCT. Corneal edema is a result of acute IOP elevation and, thus, may not be related to the pathogenesis of acute angle closure, although an increased CCT may decrease the ACD further and subsequently strengthen the relationship between ACD and APAC. Since our participants and those of Sng et al.
3 were imaged in the acute stage before any intervention or treatment, some eyes may have shown an increased CCT and subsequently a more decreased ACD. However, this is just speculation as Sng et al.
3 did not include CCT in their analysis, and, thus, we could not evaluate the effects of CCT on ACD measurement or APAC development in their results. A recent study from Razeghinejad et al.
5 suggested that an increased CCT is a risk factor for APAC development. Therefore, this is an interesting issue that needs further investigation. As shown in Table 4 of our study,
2 the ACD, CCT, and LV all were related to APAC occurrence in our analysis, but the LV showed an extremely high odds ratio, so we concluded that this was the most prominent feature.
We again thank Wang and Zhang
1 for their interest in our study and hope to continue our constructive discussions in the forthcoming longitudinal study that they suggest.