Background Central corneal thickness (CCT) and its association with intraocular pressure, which is a pivotal parameter in glaucoma management, has previously been reported. In this study, we intended to investigate the long-term change of CCT in terms of rate in eyes with primary angle closure (PAC). Additionally, we aimed to analyze events that could affect CCT.
Methods In this retrospective study, 27 patients with PAC who had a follow-up period of more than five years were evaluated. The rate of CCT changes from baseline was evaluated from the serial CCT measurements over the average follow-up period. The pattern of CCT change rate according to modes of treatment and history of angle closure attack was analyzed using the repeated linear mixed model analysis.
Results A total of 52 eyes were enrolled. The CCT reduction rate of the entire study population was − 0.72 ± 0.22 µm/yr (P = 0.001) with statistical significance. The CCT thinning rate of the laser peripheral iridotomy (PI) group was − 0.53 ± 0.25 µm/yr (P = 0.034), that of the surgical trabeculectomy group was − 1.32 ± 0.43 µm/yr (P = 0.002), and it was not statistically significant (P = 0.112). The rate of CCT thinning in patients with a history of acute angle closure attack was − 0.81 ± 0.31 µm/yr (P = 0.009) and that in patients without an attack was − 0.63 ± 0.30 µm/yr (P = 0.001), and it was not statistically significant (P = 0.680). Baseline CCT appeared to be the only significant factor affecting the rate of CCT changes (P < 0.001).
Conclusions We found a significant reduction in CCT over a long-time observation period in PAC eyes. We also found that the rates of CCT reduction were not affected by different treatment modalities or acute angle closure attack. The analysis of long-term CCT changes in conjunction with baseline CCT would also be helpful in the clinical evaluation of the PAC patients.
Significant reduction in central corneal thickness (CCT) was observed over a long-time period in primary angle closure eyes. The rates of CCT reduction were not affected by treatment modalities or angle closure attack but affected by baseline CCT.
As far as is known, intraocular pressure (IOP) is the only modifiable risk factor for glaucoma, which is a multivariate figure and small error of any variable can result in under- or over-estimation, which may significantly affect treatment and prognosis.[1, 2]
Primary angle closure (PAC) is a more common form of glaucoma in Asians compared to other ethnicities. It is part of a broader disease spectrum that includes primary angle closure suspect, PACS, primary angle closure glaucoma, PACG, and PAC. PAC is a condition with physical narrowing of the anterior chamber owing to synechial or appositional forces, which then leads to progressive damage of the optic nerve. In the case of acute angle closure, a rapid increase of IOP causes swelling and endothelial damage of cornea, resulting in various biomechanical changes. Furthermore, many recent publications have suggested that long-term use of anti-glaucoma medications, laser procedures, and surgeries affect corneal biomechanics, such as central corneal thickness (CCT).[3, 4]
The importance of CCT in glaucoma is well-established. It is widely accepted that true IOP with thinner cornea is higher than its measurement, and vice versa. Famously, the ocular hypertension treatment study, European glaucoma prevention study, and Barbados eye study showed that CCT could be a risk factor for development and progression of glaucoma disease.[5-8] Currently, multiple studies focus on its relationship to multiple types of glaucoma, age, ethnicity, and so on. Unfortunately, only a few of them showed that CCT is a substantial risk factor for glaucoma, and most of the studies were short-term.
CCT is important in the evaluation of glaucoma. We believe, understanding the corneal changes in PAC eyes according to time is crucial since PAC itself and even its treatments can induce changes in various corneal parameters.
Based on our literature review, no publication has reported a long-term change of CCT rate, in particular. This study aimed to investigate the long-term rate of CCT change in eyes with PAC.
The Institutional Review Board of Hanyang University Hospital approved this study (IRB No. 2020-04-057-001). The study design followed the tenets of the Declaration of Helsinki for biomedical research.
A total of 27 patients with PAC who were followed-up for more than five years at the Department of Ophthalmology of Hanyang University Hospital from January 2002 to December 2010 were enrolled in this study and their medical charts were retrospectively reviewed. Only the patients with PAC were included in the study. Clinical definitions of PAC and its disease spectrums are further described in oncoming section. Exclusion criteria included secondary angle closure due to any possible causes, such as neovascular, uveitic glaucoma or trauma, corneal disorders that prevent accurate measurements, previous history of ocular surgery, trauma and other ocular disorders after intraocular surgery. Patients with less than 5-year follow-up period were also excluded from the study.
All subjects underwent a complete ophthalmologic examination, including visual acuity testing, manifest refraction assessment, slit-lamp examination, IOP measurements using Goldmann applanation tonometry, gonioscopy, dilated fundus examination, axial length measurement (IOLMaster; Carl Zeiss Meditec, Dublin, Ca, USA, Aviso; Quantel medical, Quebec, Canada), stereo-disc photography and red-free RNFL photography (TRC-50X; Topcon Corporation, Tokyo, Japan, F-10; Nidek, Gamagori, Japan), and Swedish interactive thresholding algorithm (SITA) 30-2 perimetry (Humphrey Field Analyzer II; Carl Zeiss Meditec, Jena, Germany). The CCT measurements were performed using an ultrasound pachymeter (Tomey SP-3000; Tomey Corporation, Nagoya, Japan) by the same technician, recording a mean of ten consecutive readings. Above explained ophthalmologic examinations were performed every year for all study patients.
PAC and its disease spectrum
The latest classification scheme by the International Society of Geographical and Epidemiological Ophthalmology (ISGEO) describes features of the PAC spectrum (PACS, PAC, PACG). Accordingly, PAC was defined as present in an eye with an “occludable angle” with normal IOP, less than 21 mmHg. The occludable angle was defined as when the posterior trabecular meshwork was not visible on the nonindentation gonioscopy for at least two quadrants at the primary position. PAC was defined as PACS eye with features of increased IOP, trabecular obstruction, such as peripheral anterior synechiae (PAS), increased IOP, iris whirling, and glaucomflecken. Both PAC and PACS should not have had glaucomatous optic damage. PACG was defined as the presence of glaucomatous optic neuropathy with compatible visual field loss in an association with occludable angle.
In this study, a single glaucoma specialist (KBU) made clinical diagnoses using non-indentation gonioscopy. All surgical and laser treatment decisions were made and undertaken by a glaucoma specialist (KBU). The laser peripheral iridotomy (PI) was done for patients with acute angle closure by the trained doctor mentioned above. Furthermore, prophylactic PI was performed for patients without an attack, who were considered to have PAC in a broad sense. Trabeculectomy was performed in cases with severe corneal edema or failed PI, such as persistent IIOP or progression of glaucomatous damage after PI.
PI was performed in a standard manner. After administration of 2% pilocarpine eye drops, PI was done using argon and Nd:YAG lasers, sequentially. The argon laser was used to irradiate the iris, using an Abraham lens. First, 3-6 pulses at a power of 200 mW and a spot diameter of 200 micron with a duration of 0.2 seconds for iris extension, was performed. Then, 10-40 pulses at 800 mW power, the spot diameter of 50 micron, and a duration of 0.05 to create perforation in the iris were done. Irradiations were applied to the superior iris in order to avoid corneal complications. Finally, pulses of 3.5 to 5.0 mJ of the Nd:YAG laser was used for complete perforation of the wound. Preoperatively, the administration of anti-glaucoma medications and steroid medications were used for one week. When IOP was not adequately controlled, even after surgical or laser procedures, additional anti-glaucoma medications were prescribed by the single glaucoma specialist (KBU).
Calculation of Central Corneal Thickness Changing Rates & Statistical Analyses
All statistical tests were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Patient characteristics with continuous variables were expressed as the mean ± SD, and nominal variables were expressed as frequencies and percentages. The normality of the distribution of the CCT scores was verified using the Shapiro-Wilk’s test. The rate of CCT changes from baseline was determined from the serial measurements using repeated linear mixed model analysis (expressed in µm per year), with a restricted maximum likelihood estimation. Fixed effects were treatment group (trabeculectomy vs. PI and angle closure attack vs. no attack), time of measurement, and the treatment group by time interaction. The rates of change were compared among groups through testing of the interaction term in the linear mixed models. In this model, the treatment by time interaction was not statistically significant. It indicates that there were no differential changes in CCT over time, depending on treatment groups. The covariance pattern between the repeated measurements was assumed to be compound symmetry. We considered different forms of the random effects terms ranging from the simplest model with no random effects to the largest model with random intercepts and random slopes. We computed the Akaike information criterion (AIC) for a set of candidate models with different forms of random effects and selected the one with the smallest AIC value indicating better fitting model. Finally, we applied eye-specific random effects model. Additionally, we analyzed the associated CCT change rate in PAC patients with clinicopathological factors of interest using a linear mixed model, and estimate, SE, and its P-values were calculated. The level of significance was set at P-value < 0.05.
The study evaluated 54 eyes of 27 patients with PAC, and two eyes of one patient, which developed bullous keratopathy due to failure of IOP control, were excluded. Finally, 52 eyes of 26 patients were analyzed.
Table 1 shows the clinical demographics of all patients at the time of enrollment. A total of 40 eyes underwent PI only. A total of 12 patients had trabeculectomy. The latter group included patients who underwent a PI before trabeculectomy. Twenty-three eyes had a history of angle closure attack.
The average number of the examinations was 7.0 ± 2.1 (range, 4–10) over a mean follow up period of 94.5 ± 28.7 months (range, 66–134 months). Most of the study population consisted of females (92.3% of all patients) with a mean age of 63.4 ± 8.7 years (range, 48 –78 years). The baseline CCT was 549.1 ± 29.2 µm (range, 484–619 µm), and mean presenting IOP was 29.6 ± 18.2 mmHg (range, 6–64 mmHg).
Central corneal thickness change rate
In each group there was a statistically significant reduction in CCT, as indicated in Supplementary Table 1. The CCT thinning rate of all enrolled patients was -0.72 ± 0.22 µm/yr, and it was a statistically significant reduction (P = 0.001).
Overall, the CCT thinning rate in each of the two groups was statistically significant. The CCT thinning rate of the PI group was -0.53 ± 0.25 µm/yr, that of the trabeculectomy group was -1.32 ± 0.43 µm/yr (P = 0.034 and P = 0.002, respectively). However, the statistical analysis showed that a higher CCT thinning rate of the trabeculectomy group was not statistically significant (P = 0.112) (Table 2 and Figure 1). Similar results were observed in CCT comparison with regard to the presence of angle closure attack. The rate of CCT thinning in patients with a history of acute angle closure attack was -0.81 ± 0.31 µm/yr and that in patients without an attack was -0.63 ± 0.30 µm/yr (P = 0.009 and P = 0.001, respectively). There was no statistically significant difference in the CCT thinning rate between the two groups (P = 0.680). (Table 2 and Figure 2)
Figures 1 and 2 show CCT changes during the follow-up period, with the maximum being close to 11 years. It demonstrates the changes of CCT over time in all study patients, and that divided by subgroups (trabeculectomy vs. PI , and angle closure attack vs. no attack). The mean CCT changes were schematically drawn to compare the total study groups and the subgroups easily.
The results of the analysis for identifying factors associated with the CCT change rates are summarized in Table 3. Only baseline CCT was associated with the rates of its thinning (P < 0.0001). In other words, higher baseline value was associated with a higher rate of CCT reduction. However, in this statistical analysis, CCT change rate was not significantly associated with age (P = 0.297), sex (P = 0.231), IOP (P = 0.295), type of intervention (P = 0.913), and presence of angle closure attack (P = 0.238).
Biometrics of PAC eyes have been subjected to study for many years. Recent interest in CCT is due to its influence on the accuracy of IOP measurements. In this retrospective study, we found a marked reduction of CCT in all PAC patients enrolled in this study. However, no significant difference was observed regardless of acute angle closure attack and modes of glaucoma treatment. Only a few studies observed long-term reduction of CCT, especially in the study population with PAC. We believe this study has its originality in that it conducted a long-term evaluation of CCT changes in PAC patients and analyzed it with the concept of rate.
Several studies, mostly cross-sectional, have compared CCT in normal controls and other glaucoma subtypes, including PAC. The results of the studies mostly suggested that there was no statistically significant difference in CCT among glaucoma subtypes and normal controls.[11-13] A large population study in Beijing conducted by Xu et al. also suggested no significant difference between glaucomatous and normal eyes.
Comparable to our study designs, some studies have studied CCT changes in PAC eyes only. Chen et al. studied corneal status in the PAC disease spectrum. In this study, PAC eyes with a history of ACG attack were compared with fellow PAC eyes without attack and found no significant difference in CCT between the two groups. Although previous studies published significant results with large study groups, the weakness was that they were conducted in the short-term. We believe long-term change of corneal characteristics would be more applicable in assessing glaucoma since corneal characteristics are changeable with aging and corneal insults, such as surgeries, angle closure attack, and topical medications.
In this study, final mean CCT and its change rate were not significantly different between the PI and trabeculectomy groups. The same result was observed for eyes with and without acute angle closure attack. This is consistent with a few previous studies, Pillunat et al. investigated corneal biomechanical changes after trabeculectomy, and showed that despite a decline in IOP, CCT was not altered. A few other studies suggested similar findings with PI.[17, 18]
Instead of CCT, multiple studies have shown irreversible corneal endothelial damage caused by surgeries, laser procedures, and even topical medications.[16-22] In relevance to the field of glaucoma, many researchers have studied associations between the corneal endothelial loss with glaucoma subtypes. Shiota et al. investigated corneal endothelial status across subtypes of angle closure glaucoma. They found that the previous acute angle closure and chronic PACG had significantly lower endothelial cell density (ECD). At the same time, there are conflicting studies. Varadaraj et al. and Verma et al. studied endothelial changes in the PAC disease spectrum, including eyes with a history of acute angle closure attack and found no significant difference in all groups.[24, 25] Interestingly, a recent study of a rat model showed eyes with acute angle closure had lower ECD at first, but showed a gradual resolution once IOP stabilized to normal levels. It may be more useful to assess endothelial changes since CCT does not reflect the overall corneal status. Unfortunately, we were not able to perform endothelial studies on our study population owing to the retrospective nature of this study. A long-term evaluation of CCT in conjunction with ECD changes is required for future studies, and assessment of corneal status with various parameters will enrich future study results.
The repeated mixed model in this study showed that baseline CCT was the only factor associated with the rate of its reduction. Two possible mechanisms can explain such a finding. First, eyes with thick cornea can be intrinsically prone to CCT thinning. This mechanism requires further study or statistical analysis that directly compares the CCT thinning rate between eyes with a thin cornea and thick cornea. Second, corneal edema caused by an acute angle closure attack or anti-glaucoma procedures can be another possible explanation. Corneal edema can quickly resolve as IOP normalizes, which in turn, can overestimate CCT thinning rate. Correlation with the interval between the acute insult and CCT measurement will be helpful to clarify this finding further.
The primary strength of this study is that the longitudinal change of CCT in the extended time was evaluated, with the maximum follow-up duration being close to 11 years. Another strength is that we evaluated CCT with account for events that can affect the corneal status, such as acute angle closure attack and surgeries. We are aware of the numerous publications with notable results that studied CCT changes in normal and different glaucoma types.[20-22] However, these studies had a small sample size and, more importantly, had short observation periods, ranging from months to 4–5 years. The cornea is a structure that is prone to change over time. It can change with aging, long-term topical medication use, surgeries, and even diurnally.[27-30] Considering this, we believe that a long-term investigation of CCT concerning acute events is more applicable in glaucoma evaluation than cross-sectional studies of CCT.
Despite the distinct merits of this study entails, there are some limitations. One is that the study has no data regarding CCT changes with age, and that in the normal controls or in other forms of glaucoma. Many published reports are showing a decrementing trend of CCT with age, but some show contradictory results. Pang et al. and Day et al. studied CCT and its relationship with glaucoma in Asian and East Asian patients; their results showed statistically significant CCT reduction with increasing age. In this study, the CCT change rate and age were found to have no association. However, since the study only included PAC eyes, comparison with normal controls or other forms of glaucoma is required for a more fair comparison. Another limitation is that other possible confounding factors were not considered in this study. It has been previously reported that systemic diseases, such as diabetes and some anti-glaucoma medications, can affect corneal biometrics. Particularly, prostaglandin analogues, such as latanoprost and travoprost,[20-22, 31] were associated with thinning of CCT in both short-and long-term studies. However, owing to the extensive study period, information on the duration of usage, changes in regimen was difficult to follow-up, as well as to statistically analyze. For this reason, these possible contributing factors were ignored in this study and further study will be required to assess its conduciveness. Additionally, a larger sample size will be required for further studies. This may have been due to the fact that the study was restrospectively designed and included patients with very long-term follow up period, which makes it difficult to collect sufficient sample size. Lastly, as aforementioned, other corneal parameters, such as ECD, were not taken into consideration owing to retrospective nature of this study. Evaluation of PAC eyes with various parameters would be more interesting for future study and more valuable in understanding the true changes in corneal biomechanics of PAC eyes.
To conclude, we found a significant decline in CCT over a long time-period in PAC eyes. The rates of this reduction were not influenced by treatment modalities or the presence of an ACG attack. Instead, baseline CCT appeared to be the only significant factor affecting the rate of its changes. It holds for more future investigations. The analysis of other corneal parameters, such as ECD, in conjunction with CCT would be more interesting and beneficial to understand true corneal biomechanical changes in PAC eyes. Observation of long-term CCT changes by its baseline would also further validate this study’s results.
CCT = central corneal thickness, PAC = primary angle closure, PI = laser peripheral iridotomy, IOP = intraocular pressure, PACS = primary angle closure suspect, PACG = primary angle closure glaucoma, ISGEO = International Society of Geographical and Epidemiological Ophthalmology
Meeting Presentation: Presented at the 120th Annual Meeting of the Korean Ophthalmological Society, Seoul, Korea
Financial Support: This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government [(Ministry of Education) (No. NRF-2018R1D1A1B07044619).
Conflicts of Interest: No conflicting relationship exists for any author.
Ethics approval and consent to participate
This non-interventional retrospective chart-review study was approved by the Institutional Review Board of the Hanyang University Hospital (IRB No. 2020-04-057-001), and the committee (Institutional Review Board of the Hanyang University Hospital) waived the need for informed consent from the patients because the data were anonymized.
Consent for publication
Availability of data and materials
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.
No conflicting relationship exists for any author.
This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government [(Ministry of Education) (No. NRF-2018R1D1A1B07044619).
Conceptualization: Hae Min Park, Jiin Choi, Won June Lee, Ki Bang Uhm
Data curation: Hae Min Park, Jiin Choi, Won June Lee, Ki Bang Uhm
Formal analysis: Hae Min Park, Jiin Choi, Won June Lee, Ki Bang Uhm
Funding acquisition: Won June Lee
Investigation: Hae Min Park, Jiin Choi, Won June Lee, Ki Bang Uhm
Methodology: Jiin Choi and Won June Lee
Supervision: Won June Lee
Validation: Hae Min Park and Won June Lee
Visualization: Hae Min Park, Jiin Choi, Won June Lee
Writing – original draft: Hae Min Park, Jiin Choi, Won June Lee, Ki Bang Uhm
Writing – review & editing: Hae Min Park, Jiin Choi, Won June Lee, Ki Bang Uhm
WJL had full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The authors alone are responsible for the content and writing of the paper.
We thank Biostatistical Consulting and Research Lab, Hanyang University, Seoul, Korea for assistance with statistical consultation.
Table 1. Participant demographics
IOP indicates intraocular pressure; MD, mean deviation from automated visual field exam; CCT, central corneal thickness.
Table 2. Central corneal thickness change rate according the type of operation and to the presence of acute angle closure attack
CCT indicates central corneal thickness; PI, laser peripheral iridotomy.
* Repeated Measures Linear Mixed Model: Duration as continuous variable & Operation
**Repeated Measures Linear Mixed Model: Duration as continuous variable & angle closure attack
Table 3. Factors associated with central corneal thickness change rate in primary angle closure patients
IOP indicates intraocular pressure; MD, mean deviation from automated visual field exam; PI, laser peripheral iridotomy; CCT, central corneal thickness.
*Repeated Measures Linear Mixed Model