advertisement

Editors Selection IGR 11-2

Comments

Comment by Ravi Thomas on:

26370 Transscleral diode laser cycloablation in patients with good vision, Rotchford AP; Jayasawal R; Madhusudhan S et al., British Journal of Ophthalmology, 2010; 94: 1180-1183

Find related abstracts

Based on Sackett's¹ recommendations for reading the literature with the objective of application to patient care, I paraphrase a description of this paper: 'it is a preliminary study not designed to provide a definitive answer, but to suggest that pursuing a definitive answer is worthwhile.' As a glaucoma specialist it is 'pertinent to the problems of my patients, (which is why I feel compelled to glance at it) but it does not offer tested solutions to these problems, (which is why I shouldn't actually read it).'

For a formal analysis, I use the method recommended by Riegelman.² Each letter of the (modified) acronym: SAARIE (Study, Assignment, Assessment, Results, Interpretation and Extrapolation) has a comprehensive checklist, but my word limit does not permit a complete analysis.

A positive feature is that the objective of the study is clearly stated. The design, however, is not appropriate for the question (of harm) that is being asked. None of the other check-lists I examined under 'SAARIE' are satisfied: examples include the known deficiencies, and hence inappropriateness of Snellen's acuity (in this case unmasked), as a valid measure of the primary outcome and the problems of (unmasked) IOP measurement. The authors would argue that a chart review such as theirs precludes such scientific niceties, but that is precisely the problem for applicability of the findings; certainly comparison of the results to those of well-conducted randomized trials is not justified. Additionally, no confidence intervals have been reported. Why do journals make strong recommendations about confidence intervals but apply them only selectively?

From my 'poor mission hospital doctor' (PMHD) perspective there are other positive aspects: I sincerely congratulate the authors for using this controversial indication on their own patients, rather than in a tame developing country institute with a rubber stamp for an ethics committee.

The enlightened ethics committee that allowed this study is a cause for optimism: I know for sure that in the two institutes that I directed during my PMHD life I would have faced enormous difficulty in obtaining approval for such an intervention; or for that matter in obtaining informed consent from patients for this controversial indication. The authors do acknowledge the limitations and predictably suggest further appropriately designed research, a statement that I concur with, although, as an aside, unless harm is the primary outcome measure and is common, which it seems to be, RCT's may not be the best design to address this.²

This study indicates that cyclodiode treatment can be effectively used in eyes with good vision. In fact, uncritical use of such a recommendation carries a real potential for harm

I must emphasize that the limitations of study design and data do not support their concluding recommendation: 'In the absence of comparative data this study indicates that cyclodiode treatment can be effectively used in eyes with good vision […]. 'In fact, uncritical use of such a recommendation carries a real potential for harm. It is natural for the PMHD in me to consider cyclo-ablation in patients with good vision, or even as a primary procedure, but in the light of existing knowledge a more conservative and meaningful approach would be to recommend that such treatment currently be undertaken only in the context of prospective clinical research designed to provide the much needed 'tested solution'.

• ← Previous
• Next →