We read with interest the paper by Ang et al about the problem of bullous keratopathy following laser iridotomy in a Japanese hospital.1 Corneal decompensation is a serious complication causing visual morbidity, and there is cause for significant concern if the incidence after laser iridotomy is indeed truly as high as suggested in this study. We have some concerns about the presentation and interpretation of the data in this report, and ultimately the conclusions that were drawn from it.

The authors present the number and proportion of all penetrating keratoplasties performed for bullous keratopathy following laser iridotomy in Kyoto, Singapore and Nottingham (UK). Ang reported that these accounted for 20% of all penetrating keratoplasties in Japan, compared with 6.5% in Singapore and 0% in Nottingham. However, one cannot interpret these figures without knowing the total number of iridotomies being performed in the catchment area of each hospital. One would expect a higher number of laser iridotomies in Asian nations, and hence the lower rate of complications from laser iridotomy in Nottingham is not surprising.

Endothelial cell loss in an eye following acute angle closure has been well documented,2 3 and this by itself can be a potential risk factor for further endothelial cell loss and bullous keratopathy. Forty-one per cent of cases of bullous keratopathy in this series had a previous episode of acute angle closure. It is not stated how many of these eyes developed chronic glaucoma subsequently or underwent glaucoma or cataract surgery, all of which are risk factors for bullous keratopathy.

The authors have noted that in Japan, 59% of eyes that had corneal decompensation underwent prophylactic iridotomy in an asymptomatic fellow eye. This leads us to believe that the risk of endothelial failure is more likely related to the specific laser technique used. As stated, the technique predominantly adopted in Japan is that of solely argon laser iridotomy. Sequential argon-YAG laser iridotomy is the standard technique in Singapore National Eye Centre.4 This technique delivers a significantly lower laser energy to the eye than argon iridotomy alone.5 In the UK, Nd:YAG pulsed laser is the standard technique of performing laser iridotomy, with solely argon laser iridotomy being very rarely performed. It would clearly have been very instructive to compare the average laser energies delivered in the three centres, but this information was not provided.

This report of high rates of corneal endothelial failure after prophylactic laser iridotomy is at odds with our clinical experience in London. Approximately 400 laser iridotomies are performed each year at Moorfields Eye Hospital in London. Nd:YAG is used for Caucasian patients, and sequential argon:YAG iridotomies performed in Asian and African people. We have encountered no cases of endothelial failure in prophylactic treatments, and only two among people with advanced angle-closure disease.

Angle-closure glaucoma is one of the leading causes of preventable blindness worldwide. Robust research suggests that laser iridotomy or iridectomy is a safe and highly effective method of preventing loss of vision in most cases. We have reservations about the authors’ generalisation of the high rates of corneal decompensation following laser iridotomy and their conclusion that this is a growing problem in Asian countries. We believe the data point to a localised problem with laser technique. The precise magnitude of the risk attributable to the various techniques used to perform laser iridotomies cannot be assessed from these data.