The appendix of the eye: better out than in


  • Conflict/competing interest: No stated conflict of interest.

We thank Dubey et al1 for their kind words and affirmation regarding the approach in our recent review,2 and we are grateful that they have made a further contribution to discussion of this topic.

Our review's primary purpose was to establish what is known with regard to the efficacy and appropriateness of clear lens extraction (CLE) in primary angle closure disease (PACD) (i.e. the evidence base), and constrain this within strict and current definitions as applied to PACD.

When cases do not fall into the areas for which we have evidence, they need to be managed based on what we do know, on extrapolation of the existing literature and (last but not least) on our ‘clinical experience’ and that of others. The views of experienced clinicians are usually at the core of any change in established practice. On the other hand, it is imperative that any systematic new intervention is best done on the basis of a research protocol or formal trial.

There remain areas in which we have no evidence because it has not (yet) been sought – the Rumsfeldian ‘known unknowns’– which is not to say that a treatment does not work but merely that we do not know that it does. The authors highlight one such area in the discussion surrounding CLE as a means of halting peripheral anterior synechiae (PAS) formation: although we might intuitively believe that the procedure works, we currently have – at least to our knowledge – no evidence that it does. Documentation of PAS can be difficult – so too, therefore, progression – and there is a risk of overestimation of PAS preoperatively which might exaggerate apparent surgical benefit.3 We might equally believe that once iris synechial adhesion has commenced, it will proceed unabated, so that genuine clinical equipoise does exist. We state again that cataract extraction in such a situation would not be controversial; the balance of risk versus benefit for CLE remains indeterminable.

If the problem on the other hand is one of uncontrolled IOP – particularly in the presence of glaucoma (PACG: primary angle closure glaucoma) – then the question posed is why one would not use the operation we have that deals with pressure. To answer this, we need to know at least the rates of trabeculectomy subsequent to CLE for treatment of primary angle closure (PAC)/PACG,4 as compared to the rate of cataract extraction (and other complications) following trabeculectomy for these angle closure conditions.5 Again, our information on this is incomplete; the presence of an early cataract makes the decision a bit easier.

With regards to the authors' comments on the risk of acute primary angle closure (APAC) in a fellow eye, we draw a clear distinction between prevention of APAC/PACG versus management of an acute attack of angle closure, particularly in relation to the respective effects of pilocarpine as prophylaxis in the presence of a patent LPI as compared to its use as treatment of APAC in the absence of an LPI. We annotated a trial that showed that after an LPI in a fellow eye, there were no cases of APAC reported.6 Extrapolation from this existing evidence (perhaps ‘unknown knowns’) suggests that the risk is about 1.4%, which is less than the risk of lens extraction, and the follow up is less expensive. So an APAC fellow eye that has had an LPI and open angles is ‘safe’ and requires only follow up. The risk in a fellow eye that does not open following LPI is about 5%, still well short of a certainty. It seems worthwhile, therefore, trying non-invasive laser peripheral iridoplasty (we prefer this to the ambiguous term ‘gonioplasty’7) or pilocarpine prior to any incisional surgery. If the person does have specific symptoms, an acute attack, develops PAS and/or raised IOP, then all these weigh in the decision to take out a clear lens. All we are recommending is that CLE is not a routine for fellow eyes, with or without LPI. And we are glad the authors agree.

When considering the use of CLE as treatment for APAC – as previously reported by these authors8– we agree that such an approach is obviously indicated on a case-by-case basis. We only question this as a primary approach and plead for prospective data in such cases. If their three cases did well, the upper limit of the confidence interval for complications remains around 99%. This again underpins our suggestion for prospective data or a trial for this scenario.

Ophthalmology is currently caught in a cleft stick. Although marketing and approaches to government bodies stress the highly specialized and complex nature of cataract surgery, clinical indications to remove a lens seem sometimes to border on the cavalier, as if the operation is now so trifling and we have become so proficient at it that the lens should be regarded as the merest nuisance on the path to perfect vision and anatomical correctness.

Clear lens extraction has a place in management of PACD. Nevertheless, it is an operation. It has a risk. As always, to ascertain benefit this risk must be compared to the natural history of the condition for which surgery is being proposed as well as to conservative measures. Where available, the evidence base should be deployed, in a clear-minded manner. In the management of PACD, we are glad that these authors concur, and thank them for their endorsement.