Rethinking laser iridotomy in angle-closure disease

Till recent times, one of the important procedures in glaucoma was performing laser iridotomy (LI) for primary angle-closure disease (PACD). While there are no records from India, there definitely was a trend of increasing number of LI being done, than ever before. It was considered to benefit the entire spectrum of PACD. So much was the passion for doing iridotomies, that the focus of ophthalmologists, especially the glaucoma specialists, shifted from managing high IOP to performing iridotomies.

This, however, changed (or should have) with numerous randomized controlled trials, in populations with a high prevalence of PACD, that showed how little LI changes the natural course of the disease. One of these (not so recent) trials showed that the efficacy of LI in primary angle-closure (PAC) was not much,[1] while two other recent trials showed how less it benefitted the primary angle-closure suspects (PACS),[2,3] thus, arose the question of the efficacy of LI in PACD.

In India, there are about 21 million people with PACS[4] and these numbers are much higher in the Chinese population. The Zhongshan Angle-Closure Prevention (ZAP)[2] as well as the Asymptomatic Narrow Angle Laser Iridotomy Study (ANA-LIS)[3] showed that actually, very few PACS eyes go on to develop high IOP, acute angle-closure attacks, or primary angle-closure glaucoma (PACG) without LI. In the ZAP trial,[2] only five (0.56%) PACS eyes of 889 eyes (that did not undergo LI) had raised IOP >24 mmHg and five eyes (0.56%) had an acute angle-closure attack, over a 6-year follow-up. In the ANA-LIS,[3] only 18 PACS eyes (3.7%) of 480 (that did not undergo LI) had raised Intraocular pressure (IOP) >21 mmHg and only two eyes (0.4%) of 480 had an acute angle-closure attack, over a 5-year follow-up. Which of the PACS or PAC eyes progress (with or without LI) is another story that has received much attention.[5] Shallower anterior chamber depth (ACD) eyes (which is generally associated with non-pupillary block mechanism) were found to be more at risk of progression. These eyes with thicker lens or increased lens vault, at the very outset, are unlikely to benefit from an LI. Interestingly, both these trials demonstrated that gonioscopy is only weakly predictive of angle-closure outcomes and fails to identify individuals at risk for disease progression.[6]

As these trials got published, arguments were put forth to protect and retain the so-called innocuous LI in the armamentarium of glaucoma specialists. After all, they (glaucoma specialists) were the only ones to put in the gonioscope and decide to laser or not to laser. One of these arguments was that these trials were conducted in a relatively younger population (by Western standards); around 60 years of age, wherein the risk of PAC or an acute attack itself was lower. But to counteract this argument, is the fact that a large number of patients by the age of 70 years would undergo a lens extraction which would probably be more beneficial than an LI.[7] In this context, one should not forget that in India, there exist a large number of cases with angle-closure disease, presenting less than 40 years of age, in whom the mechanism of pupillary block (for which LI is classically beneficial) is relatively less common.[8] Another argument put forth was that these trials selected patients from the general population (and were not truly hospital-based) and had deeper average ACD that were unlikely to progress to PAC anyways but the average ACD in these trials (on anterior segment optical coherence tomography (ASOCT)) was 2.2 mm which is still shallower than the general population. To counteract this argument, very shallow ACD is generally a non-pupillary block mechanism. What is clear and has always been is that the LI only relieves a relative pupillary block (RPB), which accounts for only 30–40% of cases of PACD.[9] It thus needs a clear insight to understand which eyes have a RPB as the predominant mechanism and will thus benefit from an LI. This is what needs to be taught in glaucoma training centers, rather than: “shoot” if you find an occludable angle. As these trials were conducted in high-risk population with PACD, it was also assumed that the results would not be applicable to populations (Caucasians) where RPB was the predominant mechanism. The argument still boils down to, doing an LI only when you find a RPB.

In contrast to PACS, there are few trials that look into the efficacy of LI in the PAC subset of eyes. Yip et al.,[1] in a randomized controlled trial, found that screening with a central ACD and performing LI in PAC eyes did not reduce the 6-year incidence of PACG. The Effectiveness of early lens extraction for the treatment of primary angle-closure glaucoma (EAGLE) study[7] reported that 65% of PAC/PACG eyes require medical therapy after an LI, which is indicative that LI alone is not sufficient to prevent disease progression in these eyes. Unfortunately, PAC as a diagnostic group is not well defined, as the presence of peripheral anterior synechiae (PAS) alone is sufficient to make the eye jump into the category to PAC from PACS, and this identification of PAS on gonioscopy in one or two quadrants is highly subjective and generally biased. Despite insufficient high-quality data, it is still advised to perform LI in PAC eyes as their risk to PACG seems higher. However, appropriate timing of LI and the eyes which will actually benefit from the procedure have not been studied.

The proportion of PACS to PAC to PACG in Asian eyes is around 10 to 2 to 1.[10] Hence, PACG is less common in the general population. But once a glaucomatous optic neuropathy sets in (PACG), the indication for doing an LI becomes less clear especially as all these eyes need medical/surgical therapy.[11] By the time a patient comes with PACG, he/she also has enough iris atrophy, that in any case would prevent an RPB. Moreover, such eyes would have extensive damage to their trabecular meshwork that cannot be undone by an LI. In such cases, an LI would only give false hope of doing something, when the eye is already in the stage where progression could be very fast if appropriate medical or surgical therapy is not initiated. Nolan et al.[12] in their study concluded that in eyes with angle-closure and established glaucomatous optic neuropathy, LI would fail in half the eyes. Their criteria for failure were stringent, either need for surgery or a drop in visual acuity to <3/60. One meta-analysis found that LI is not effective in preventing visual field loss in PACG eyes; in fact, the adverse effects were more in LI-treated eyes than those that were not treated.[13]

Another question general ophthalmologists want to know is, whether they need to do an LI before cataract surgery for an eye with a shallow anterior chamber for fear of inducing an acute angle-closure attack with dilation or having a malignant glaucoma intra-operatively. In a study among Chinese patients, only three (0.6%) developed an acute angle-closure attack of 471 PACS participants who were dilated.[14] Also, there is nothing to suggest that an LI will prevent a malignant glaucoma, though there are reports to suggest that the LI itself can cause a malignant glaucoma.[15,16]

The adverse effects of LI cannot be ignored, and after all, the risks need to be weighed against the potential benefits. Although neither the ZAP nor the ANA-LIS reported any adverse effects of LI, visual disturbances are well-known and problematic when the LI is conducted as a prophylactic procedure.[17,18,19] The risk of long-term corneal decompensation also remains in certain ethnic groups.[20] The increased risk of cataract progression with LI has been found in three of the four studies published over the last 20 years.[21,22,23,24] One-fourth (25%) of fellow eyes of acute angle-closure patients were found to have progression of their cataract after a prophylactic LI, in a study from Singapore.[21] One-fifth (20%) of the participants from Mongolia who underwent LI for PAC had progression of their cataract and decline in their vision, though the authors concluded that LI was not independently associated with cataract progression.[22] A large study from South India[24] found 38% of cases having cataract progression when LI was conducted for PACS, increasing the odds for cataract progression to 1.7. In contrast, participants in the ZAP trial were not found to have significant cataract progression after LI for PACS, though they found that LI-treated eyes had higher average nuclear cataract grades.[23] So, is LI more effective causing cataract progression than preventing disease progression in PACD? It is time to rethink….

About the author

Prof. Viney Gupta

Prof. Viney Gupta is currently Professor Glaucoma Services at Dr R.P Centre for Ophthalmic Sciences, AIIMS, New Delhi. He is currently heading the Glaucoma Services Unit 2, at Dr R.P Centre for Ophthalmic Sciences. He did his fellowship training in Glaucoma at the Royal Victoria Eye and Ear Hospital, Australia. He is also an MBA from Delhi University. He has published more than 220 papers in peer reviewed journals and has been the Chief guide/co Guide for more than 50 Post graduate thesis in Glaucoma. He is on the Editorial board of IJO as an Associate Editor, as well as on the Editorial boards of AJO Case reports and EYE. He has a keen interest in Congenital and Juvenile Glaucoma and Glaucoma Genetics.