Primary angle-closure glaucoma (PACG) is an important, preventable cause of visual loss. PACG affects 20 million people, and has blinded over 5 million globally [1]. Although it is well established that Asian people are at greater risk than white people of European origin [2, 3], an estimated 1.6 million white Europeans, 581,000 white US citizens and 130,000 white Britons have visual field loss from PACG [4]. The results of two major clinical trials have transformed the evidence-base informing management of PACG. The EAGLE trial clearly demonstrates that anyone with PACG, and those with an intraocular pressure (IOP) > 30 mmHg resulting from primary angle-closure (PAC), should be offered clear lens extraction as the first intervention. This offers better disease control (better pressure control with less medication), better quality of life, and is more cost effective when compared to the standard care of laser peripheral iridotomy (LPI) [5].

In the context of the large numbers of people affected, it is not surprising that, in 2005, 75% of UK consultant ophthalmologists when questioned said they would offer LPI as a preventive treatment to patients at risk of PACG [6]. However, this strategy is probably a well-meaning extrapolation of the unquestioned importance of performing LPI in patients suffering acute angle-closure (AAC) [7]. However, the strategy of offering prophylactic LPI was not based on evidence. Indeed, one large randomised controlled trial in Mongolia showed no benefit for a package of screening and prophylactic LPI [8]. Furthermore, current policy strongly advises community optometrists to refer all patients who may be at risk of PACG to see an ophthalmologist [9]. In this context, “at risk” is defined as a limbal chamber depth grade of </=25% of peripheral corneal thickness [10]. These people are assumed “primary angle-closure suspects” (PACS) [11].

The Zhongshan Angle-closure Prophylaxis (ZAP) study provides the first clear insights into the natural history of PACS, and of the benefits of prophylactic LPI, in one of the highest risk populations on earth—Chinese people over the age of 50. The trial showed that, at the end of the planned 3-year follow-up period, there was no detectable benefit from prophylactic LPI (36 month hazard ratio (HR) = 0.90, 95% CI 0.44–1.85; p = 0.777). The reason for this finding was an exceptionally low rate of incident PAC or PACG. The study was extended for a further 3 years, at which time, it was found that LPI halved the risk of incident PAC (72 month HR = 0.52; 95% CI 0.30–0.91; p = 0.023). There were no incident cases of PACG over the 6 year follow-up. There were only 5 untreated eyes and one treated eye that suffered AAC. In eyes that had not undergone prophylactic LPI, this equated to a risk of 1.1 eyes per 1,000 years. There were 5 untreated and 3 treated eyes that were found to have a sustained IOP elevation (>21 mmHg). The bulk of PAC disease identified in the ZAP trial were defined by peripheral anterior synechiae (PAS), affecting 15 treated and 30 untreated eyes [12]. It is important to recognise that, while elevated IOP does present a measurable risk for visual field loss [13, 14], the risk of sight loss associated with PAS is unknown.

With new data from the EAGLE and ZAP trials, the Royal College of Ophthalmologists approved a proposal from the authors to draw up guidelines for management of PACG and PAC in the UK. The evidence synthesis and writing of draft guidelines were completed in March 2020. The events of the following months have brought UK healthcare under pressure not seen since the foundation of the National Health Service. The huge backlog for routine care makes it vital that clinical capacity be used for maximum benefit. The results of the LiGHT trial brings selective laser trabeculoplasty (SLT) clearly into the repertoire of treatments that should be offered to the many patients with ocular hypertension and early primary open-angle glaucoma [15]. There is an opportunity cost for glaucoma laser treatment capacity. Offering prophylactic LPI which appears to be of marginal benefit in a high-risk Asian population, or offer SLT, which is proven in terms of its medical performance and health economics. SLT as first treatment has a 97% probability of being more cost effective than eye drops first at a willingness to pay of £20 000 per quality-adjusted life-year gained. In contrast, preliminary calculations for prophylactic LPI suggest a 7% probability of being cost effective at a willingness to pay £20 000 per quality-adjusted life-year in the UK (Ramjee, Foster, currently unpublished).

Looking at the performance of prophylactic LPI from the perspective of numbers needed to treat (NNT), the ZAP study authors calculated that, in a high-risk Chinese population, the NNT to prevent one case of PAC over 6 years was 44. Making cautious extrapolation to prevention of glaucoma, the NNT is 126 treated to prevent one case of glaucomatous field loss over 10 years. In the UK, the NNT’s will be larger, probably by a factor of 2–3, if one takes the ratio of AAC occurring in Caucasians to East Asians [16]. To quote from the ZAP trial manuscript, “LPI should only be offered to those with the (very) highest risk of PACG”.

Furthermore, the impact LPI has on patients should not be underestimated. In a focus group, patients unanimously reported great anxiety before LPI, and reported pain (see Table 1).

Table 1 Patient experience with laser peripheral iridotomy.
Full size table

Operationalising this recommendation in the NHS in 2021 requires a pragmatic balance of caution, and a rational appraisal of facts. Not only does prophylactic LPI offer a poor return on time, effort and expenditure, it diverts resources from other more cost-effective interventions. Furthermore, only a proportion of patients referred for assessment will have been correctly classified as PACS, with the false positive referrals unnecessarily absorbing further time and manpower.

In this context, we reinforce the message that prophylactic LPI should only be offered to those individuals at highest risk. Table 2 lists the characteristics that capture this level of risk. We see this recommendation not as a final, definitive policy for PACG management in the UK, but as an important stage in its evolution.

Table 2 “PACS PLUS” - Criteria for referral of people with suspected occluded angles to Hospital Eye Service or secondary care provider.
Full size table