Conflict/competing interest: There are no commercial conflicts.
Killing two birds with one stone: the potential effect of cataract surgery on the incidence of primary angle-closure glaucoma in a high-risk population
Version of Record online: 3 AUG 2011
© 2011 The Authors. Clinical and Experimental Ophthalmology © 2011 Royal Australian and New Zealand College of Ophthalmologists
Clinical & Experimental Ophthalmology
Volume 40, Issue 4, pages e128–e134, May/June 2012
How to Cite
Chan, W., García, J. A., Newland, H. S., Muecke, J., McGovern, S., Selva, D., Aung, T. and Casson, R. J. (2012), Killing two birds with one stone: the potential effect of cataract surgery on the incidence of primary angle-closure glaucoma in a high-risk population. Clinical & Experimental Ophthalmology, 40: e128–e134. doi: 10.1111/j.1442-9071.2011.02607.x
Funding sources: Financial support was provided by an Independent Grant Scheme from Pfizer Australia.
- Issue online: 14 JUN 2012
- Version of Record online: 3 AUG 2011
- Accepted manuscript online: 13 JUN 2011 11:51AM EST
- Received 6 October 2010; accepted 15 May 2011.
- angle-closure glaucoma;
- blindness prevention;
- cataract surgery;
Background: To estimate the proportion of cataract surgery performed at various visual acuity and lens opacity thresholds that would coincidentally treat early angle-closure disease, and to estimate the effect of this surgery on the incidence of primary angle-closure glaucoma.
Design: Cross-sectional, population-based survey in Meiktila, Myanmar.
Participants: Total of 2076 inhabitants, 40 years of age and over were included.
Methods: Eyes with cataract-induced visual impairment, and primary angle-closure disease were identified. Analyses were stratified by various pinhole-corrected visual acuity and Lens Opacity Classification System III scores thresholds.
Main Outcome Measures: The dual role of cataract surgery in primary cataract treatment and primary angle-closure glaucoma prevention was estimated.
Results: Of 4153 eyes available for analysis, 261 eyes were either primary angle-closure suspect or primary angle closure; 975 eyes had a visual acuity of <6/18 and Lens Opacity Classification System III score ≥3 on the nuclear or cortical scales. Of these, 86 eyes had either primary angle-closure suspect or primary angle closure. If cataract surgery were performed on all 975 eyes, this would potentially prevent up to 86 cases of primary angle-closure glaucoma in this population; 8.82% (95% confidence interval 7.12–10.78%) of the cataract surgery would address the cataract and prevent primary angle-closure glaucoma. This would achieve a 38.46% (95% confidence interval 20.23–59.43%) relative reduction in the incidence of primary angle-closure glaucoma in the adult population.
Conclusion: In populations with a high prevalence of both visually significant cataract and angle-closure disease, quality cataract extraction can serve a dual role of visual restoration and reducing the incidence of angle-closure disease in the population: killing two birds with one stone.
Cataract and glaucoma are the leading causes of blindness in the world.1 Angle-closure glaucoma accounts for approximately half of the glaucoma blindness, and is particularly common in those of Asian descent.2 The adult population of Myanmar is no exception.3 In the Meiktila Eye Study, 8.1% of the population 40 years of age and over was blind by World Health Organization acuity-related definition; cataract and glaucoma were the cause for 64% and 17% of blindness, respectively.3 Primary angle-closure glaucoma (PACG), both acute and chronic, was visually devastating in this population.4 Nuclear cataract was as an independent predictor of angle-closure disease.4 Nuclear cataract has also recently been identified as an independent risk factor for angle-closure disease in two other Asian populations.5,6
The current classification of angle-closure disease is based on its perceived natural history: a gradual, but not necessarily relentless, progression of angle narrowing causing iridotrabecular contact, at first appositional and latter synechial, with eventual intraocular pressure elevation and the development of glaucomatous optic neuropathy.7 The classification also recognizes that this gradual process can be interrupted by sudden acute angle closure. Hence, in the early stages of angle-closure disease, provided long-term appositional closure has not irreversibly damaged the trabecular meshwork (a phenomenon which is widely believed to exist but lacks convincing evidence), and in the absence of extensive synechial angle closure, it would generally be accepted that removal of the lens would prevent iridotrabecular contact and essentially halt further progress of the disease. Whether or not cataract extraction and intraocular lens implantation opens all angles with appositional closure is unclear, but exceptions seem rare. Here, we estimate the potential impact of increasing cataract surgery rates (CSR) on reducing the burden of PACG on the adult population in Meiktila.
The methodology for the Meiktila Eye Study has been previously described.3 Briefly, the inhabitants 40 years of age and older from villages in the Meiktila District were selected by random, stratified cluster sampling. A comprehensive ophthalmic examination including presenting and pinhole Snellen visual acuity with an illiterate E chart, slit-lamp examination of the anterior segment, and dilated stereoscopic fundus examination was performed on each participant.
Intraocular pressure was measured with a Goldmann applanation tonometer (Haag-Streit, Koeniz, Switzerland) and anterior segment examination was performed at the slit lamp (Haag-Streit, Koeniz, Switzerland). Two experienced ophthalmologists (RJC and SM) performed the gonioscopy using a Sussman goniolens. Static gonioscopy was performed in dim illumination with minimal pressure on the cornea using a short slit beam; each quadrant was graded using a Scheie-type classification. If >90 degrees of posterior trabecular meshwork (posterior TM) was visible the pupil was dilated with tropicamide 1% and phenylephrine 2.5%. Eyes with ‘occludable’ angles were dilated with tropicamide 0.5% only and kept under observation for 4 h, or if not possible, they were not dilated. If either eye had evidence of previous acute angle-closure glaucoma (see definition below) then neither eye was dilated. Two experienced ophthalmologists (HSN and JM) performed the lens gradings after dilation using the Lens Opacity Classification System (LOCS) III.8 The agreement between the lens gradings was good (Fleiss kappa = 0.78).
Optic disc and retinal examinations were then performed using a 78 D lens and reference to standard disc images. The vertical cup:disc ratio and the presence of focal notching were recorded. The agreement between the two ophthalmologists was good for grading the angle (Fleiss' kappa = 0.78) and determining the cup:disc ratio (Fleiss' kappa = 0.72).
The investigators were aware of the studies hypothesis. Parameters utilized in this study were bilateral corrected visual acuity; LOCS III score; and status of primary angle closure (PAC) or primary angle-closure suspect (PACS), collectively known as primary angle-closure disease (PACD). PACS is defined as those eyes with irido-corneal angles traditionally described as ‘occludable’ in epidemiological studies (<90 degrees of posterior meshwork visible on gonioscopy) and considered at risk of angle-closure glaucoma. PAC is defined as any eye with PACS and peripheral anterior synechiae and/or elevated intraocular pressure (>97.5th percentile for the population [≥22 mmHg in this study]) and/or iris whorling or stromal atrophy, but without any glaucomatous optic neuropathy. Peripheral anterior synechiae was defined as an area of iridotrabecular contact which could not be broken during dynamic gonioscopy.
(a) Estimating the percentage of cataract surgery that would treat both cataract and prevent the progression to PACG – the ‘K2B%’ (killing two birds percentage).
We included only eyes in which the primary cause of visual impairment was cataract. We then stratified the eyes into three groups according to pinhole-corrected visual acuity: Group 1 eyes were <6/18, Group 2 eyes were ≤6/60 and Group 3 eyes were <3/60. Each main group was further subdivided into three subgroups according to the LOCS III score on the nuclear and cortical scales. Subgroups were recorded as 3–3.9, 4–4.9 and ≥5, representing the range of LOCS III score. Hence, the impact of cataract surgery stratified by both visual acuity and degree of cataract could be assessed.
The eyes in the nine subgroups represent visually impaired eyes that would potentially benefit from cataract surgery. We then identified eyes with PACD in these subgroups. Within each subgroup, the percentage of cataract surgery that would treat both cataract and prevent the progression to PACG or the K2B% was calculated by dividing the number of eyes with PACD by the total number of eyes in the subgroup.
(b) Estimating the reduction in the population burden of PACG by cataract surgery – the ‘Burden Relief percentage (BR%)’.
We first estimated the number of eyes with PACD that are projected to progress to PACG for the entire study population and for each subgroup. As there are no available data of progression rate in this population, we applied progression rates from another population to our study population.9,10 We used 10-year progression rates of 6.26% and 28.5% for PACS and PAC to progress, respectively, to PACG.11 To estimate the reduction in the population burden of PACG by cataract surgery or the BR%, we divided the number of eyes with PACD that progresses within each subgroup by the number of eyes with PACD that progress in the population.
Finally, we calculated the number of eyes estimated to get cataract surgery based on the current CSR in Meiktila. These patients are grouped under the subgroup ‘Current CSR’ to allow comparison between the K2B% and BR% of current cataract surgery coverage with the other subgroups. CSR of 4.18%, 9.39% and 13.47% were applied to cataractous eyes (LOCS III ≥ 3) with visual acuity of <6/18, ≤6/60 and <3/60, respectively, to obtain the number of eyes in the ‘Current CSR’ group.12
We made the following assumptions for our model:
- 1All PAC and PACS eyes are at risk of progressing to PACG.
- 2The risk of progression to PACG is the same for all eyes with PACD and is independent of cataract and visual acuity status.
- 3The progression rate of PACD to PACG in other populations is similar to this population and directly applicable to this population.
- 4Cataract surgery will afford 100% prevention against PACG developing in eyes with PACD.
Both K2B% and BR% (both proportions with binomial distributions) and their 95% confidence intervals were calculated using a commercially available statistical software package (SPSS for Windows, version 10.1, SPSS Inc., Chicago, IL, USA).
A total of 2481 subjects were eligible, 2076 participants (4152 eyes) were examined and included in analysis. There were 261 eyes from 148 subjects identified with angle-closure disease; 217 were PACS and 44 had PAC. We excluded 329 eyes with poor visual prognosis secondary to other ocular pathology.
(a) The K2B%
The total number of clinically significant cataractous eyes (LOCS III score of ≥3) stratified according to pinhole-corrected visual acuity are presented in Table 1. Table 2 displays the eyes stratified into nine subgroups according to pinhole-corrected visual acuity and LOCS III score. The visual acuity thresholds are cumulative: blind subjects are also included in the moderate low vision subgroup and similarly moderate low vision and blind subjects are included in the low vision subgroup.
|Best visual acuity||Low vision||Severe low vision||Blind|
|Number of eyes||975||511||210|
|Number of eyes with PACD (PACS/PAC)||86 (65/21)||44 (31/13)||8 (3/5)|
|Best visual acuity||Low vision||Mod. low vision||Blind|
|Number in Subgroup||191||378||408||55||141||315||9||29||172|
|Number with PACD in subgroup (PACS/ PAC)||11 (11/0)||35 (27/8)||40 (27/13)||4 (4/0)||13 (11/2)||27 (16/11)||0 (0/0)||1 (0/1)||7 (3/4)|
There were 975 eyes with clinically significant cataractous eyes (visual acuity <6/18 and LOCS III of ≥3). In total, 76 eyes (7.69%) were estimated to get cataract surgery based on current cataract surgery coverage in Meiktila and included in the ‘Current CSR’ subgroup. The K2B% of the various subgroups are presented in Figure 1.
(b) The BR%
Table 3 presents the number of eyes with PACD that are projected to progress into PACG for the entire population and each subgroup. The BR% of the various subgroups according to best corrected visual acuity and range of LOCS III score are presented in Figure 2.
|VA & LOCS score thresholds||No. of PACS projected to progress/ No. of PACS||No. of PAC projected to progress/ No. of PAC||Total No. of eyes projected to progress to PACG|
|<6/18 & ≥3||4.08/65||5.99/21||10|
|<6/18 & ≥4||3.39/54||5.99/21||9|
|<6/18 & ≥5||1.70/27||3.71.13||5|
|≤6/60 & ≥3||1.94/31||3.71/13||6|
|≤6/60 & ≥4||1.70/27||3.71/13||5|
|≤6/60 & ≥5||1.00/16||3.14/11||4|
|<3/60 & ≥3||0.19/3||1.43/5||2|
|<3/60 & ≥4||0.19/3||1.43/5||2|
|<3/60 & ≥5||0.19/3||1.14/4||1|
Table 4 summarizes the point estimate with 95% confidence interval of K2B% and BR% for each of the subgroups. After stratifying for visual acuity, cataract surgery on cataractous eyes shows a proportional decrease of K2B% and BR% with worsening visual acuity. However in the subgroup analyses, within individual band of visual acuity, the K2B% increases as the severity of LOCS increases.
|VA & LOCS score thresholds||K2B% (95% CI)||BR% (95% CI)|
|Current CSR||7.89 (2.95–16.40)||3.85 (0.01–19.64)|
|<6/18 & ≥3||8.82 (7.12–10.78)||38.46 (20.23–59.43)|
|<6/18 & ≥4||9.57 (7.6–11.84)||34.61 (17.21–55.67)|
|<6/18 & ≥5||9.83 (7.11–13.14)||19.23 (6.55–39.35)|
|≤6/60 & ≥3||8.61 (6.33–11.39)||23.08 (8.97–43.65)|
|≤6/60 & ≥4||8.77 (6.34–11.75)||19.23 (6.55–39.35)|
|≤6/60 & ≥5||8.57 (5.72–12.22)||15.38 (4.36–34.87)|
|<3/60 & ≥3||3.81 (1.66–7.37)||7.69 (0.95–25.13)|
|<3/60 & ≥4||3.98 (1.73–7.69)||7.69 (0.95–25.13)|
|<3/60 & ≥5||4.07 (1.65–8.21)||3.85 (0.01–19.64)|
We have used data from the Meiktila Eye Study to model the percentage of PACD that would be ‘captured’ at various acuity-based and lens opacity-based thresholds for cataract surgery. We found up to 38.46% of total PACG burden reduction when cataract surgery is provided to all clinically significant cataractous eyes (Pin-hole acuity < 6/18 and LOCS III ≥ 3). Both the K2B% and the BR% were reduced with decreasing visual acuity. The relatively low proportion of blind eyes with either PAC or PACS was because a considerable proportion of these eyes had already progressed to PACG, which was not included in the numerator. Also within each of the visual acuity groups, the K2B% increased with increasing severity of LOCS III score. Our results suggest that: (i) providing cataract surgery at a lower clinical threshold (better visual acuity); and (ii) targeting patients with higher LOCS III score, we can optimize the dual role of cataract surgery and lead to a greater reduction in prevalence of PACG in this region.
The concept that increasing rates of cataract surgery at the population level can reduce the incidence of angle closure is supported by recent empirical evidence.13,14 Keenan et al. recently analysed patient records in England and reported that the incidence of PACG has started to decline in recent years, after a decade of increasing cataract surgery.14 They concluded that this trend supports the hypothesis that increasing rates of cataract surgery at the population level may reduce the population incidence of acute angle closure.8 Hu et al. recently reported a similar trend in Taiwan.13
Our model is an oversimplification of reality and has a number of limitations: (i) We have taken cross-sectional data and used it to model changes over time. (ii) To model changes over time, we have assumed that if cataract extraction is performed at an early stage of PACD that it is 100% protective against progression to PACG. In other words, we are assuming that if being phakic is considered as the ‘exposure’, then the population attributable risk percentage of phakia is 100%; that is, the risk of conversion of PACD to PACG in pseudophakes is 0%. However, we believe that before synechial closure has occurred that removal of the lens will open the angle in the vast majority of cases and that a 100% prevention assumption is reasonable. There are no data in the literature regarding the efficacy of cataract surgery in preventing PACD progression, even adopting a more conservative, albeit arbitrary, value of 70% efficacy, we will still be achieving a BR% rate of up to 28.85%. (iii) Information about the incidence of actual progression of PACD is scarce, but Thomas et al. reported that over the course of 5 years, 22% of PACS progresses to PAC9 and 28.5% of PAC progressed to PACG.10 This was extrapolated to give the 10-year progression rate used in this study.11 The incidence of progression over longer periods has not been reported, but the incidence of progression does not change the relative protection afforded by cataract surgery, the K2B%, assuming that progression is not influenced by cataract severity. But this assumption may not be valid. We have assumed that those subjects with PACD that did not meet the cataract surgery thresholds had the same incidence of progression to PACG as those that were ‘captured’ by cataract surgery. However, this may significantly underestimate the actual risk reduction afforded by cataract surgery. Nuclear cataract was an independent predictor of PACD in this population; hence, the patients with significant cataract may form a large portion of those with PACD that progress to PACG. (iv) We have constructed the model with the eye as the unit of analysis rather than the individual. This was a simplification to avoid the confusion arising when both eyes have PACD but only one eye met the cataract surgery thresholds. However, the restoration of normal vision after successful cataract surgery in one eye coupled with the elimination of risk of visual impairment from PACG in that eye, assuming no other blinding pathology in that eye, would leave the individual classified as not visually impaired regardless of the fate of the fellow eye.
Another limitation of our study is in Meiktila, the mean age of patient undergoing cataract surgery is 67.2 years of age (private communication with PA Athanasiov8) is significantly higher compared to the mean expected life expectancy of 54 for the country.15 This reduces the impact of cataract surgery on PACD progression as it takes years for PACD to progress into PACG. The Meiktila District was chosen for logistical reasons, not randomly, and may not be representative of neighbouring regions within the Mandalay Division of central Myanmar; however, we have no reason to believe that this is the case.
Currently the PACD pyramid in Meiktila is top heavy; on a per-eye basis prevalence of PACG, PAC and PACS are 1.9%, 1.05% and 5.2%, respectively.4 With time, we believe that increasing the CSR provided at a lower visual acuity threshold would reshape the PACD pyramid and reduce the prevalence of PACG in the population. As PACG accounts for 84% of the glaucoma-related blindness in this population,4 reducing the incidence of PACG by cataract surgery would directly influence the number of PACG-related blindness. Eventually the BR% of cataract surgery would decrease and plateau as the prevalence of pseudophakia in the population increases. Until then, we believe that increase provision of good quality cataract surgery could act as a parallel step in preventing PACG in this population.
Although we do not have postoperative outcomes of cataract surgery for this population, we know that in developing countries, the rate of poor visual outcome (visual acuity <6/60) after cataract surgeries range from 7% to 40%.16 If quality of cataract surgery is poor, or is compromised in the pursuit of increased quantity, the number of iatrogenic poor visual outcomes will far outnumber any potential gain in preventing PACG from advocating higher CSR. To prevent this, overall quality of cataract surgery and services to address postoperative complications and refractions must also be improved. Only with both improved quality and quantity of cataract surgery would the full impact of cataract surgery in addressing cataract and PACG be realized in this population.
In conclusion, assuming cataract surgery will afford 100% prevention against PACG developing in eyes with PACD; by increasing high-quality cataract surgery coverage, particularly to thresholds of moderate visual impairment, and particularly in regions where PACD is prevalent, that the incidence of PACG will decrease. Hence, we can kill two blinding birds with one stone.
- 15WHO Statistical Information System (WHOSIS). World Health Statistics 2010. Geneva: World Health Organization, 2010. Accessed April 2010. Available from: http://www.who.int/whosis/whostat/EN_WHS10_Full.pdf