Do IOP fluctuations matter? To the assiduous 10-year-long observers of glaucoma literature, the sheer number of publications on the subject would suggest that an all-important risk factor for glaucoma progression and/or conversion has been identified, challenging the value of mean diurnal IOP as glaucoma management parameter. The most discerning of those observers, however, may well wonder whether we are indeed witnessing a paradigm shift or whether this controversy results from the limitations of analyzing data collected for other purposes post-hoc, combined with the lack of an appropriate method for measuring 24-hour IOP fluctuations.

This is the controversy that Singh and Shrivastava (628) have chosen to discuss. In their relatively short, yet information-packed review paper, they start by reminding us how the mean diurnal IOP (still the standard measure of efficacy in glaucoma therapy) has been challenged ‐ though not replaced ‐ by the concept of a damage-threshold IOP and, more recently, by the somewhat ill-defined concept of IOP fluctuation, which they proceed to clarify by distinguishing between 24-hour 'fluctuation' and longer-term or inter-visit 'variation'. Then, they review critically eight of the most significant papers published in the last decade on the relation between IOP fluctuation/variation and glaucoma conversion or progression. From the eight reviewed papers, three concluded that IOP fluctuation was an independent predictor of progression, and another three that it was not. The remaining two papers presented mixed evidence, either in terms of patient subgroups (IOP fluctuation being a significant risk factor in patients with low, but not with high mean IOP) or in terms of substitute indicators for fluctuation (IOP standard deviation being positively, and IOP range being negatively correlated with progression).

At the present time, mean IOP should by no means be discarded as a key parameter in glaucoma management

Although the ratio of positive-to-negative conclusions in the sampled publications would point towards a perfect 'draw' in the controversy, Singh and Shrivastava highlight differences in depth and quality of analysis, which indicate what their intimate conviction is. The papers 'in favor' tended to be based on substitute indicators of fluctuation (e.g., cross-sectional variability in self-tonometry values over 5 days, or SD of IOP measurements over a very long period) or on suboptimal patient populations (e.g., that have undergone glaucoma surgery sometime during the observation periods), thus introducing severe limitations with regard to causality. Also, weaknesses or speculations in their analyses are highlighted. On the contrary, the papers 'against' have better-argued analyses and use indicators of fluctuation and data sets corrected for confounding factors such as IOP level or change in treatment regimen.

What is then to be concluded on the question elegantly summarized in the paper title? The authors refrain from taking sides, diplomatically stating that 'the data are certainly not overwhelming in one direction or another' and anticipate that such questions will remain unanswered until a continuous IOP-recording device becomes available. However, they do point out that the validity of a theory does not depend on the number of studies and publications, but on their quality, and they very correctly insist that at the present time, mean IOP should by no means be discarded as a key parameter in glaucoma management.

Biography:
George N. Lambrou, MD (Univ. of Strasbourg), MSc (Univ. of Paris-South), PhD (Univ. of Amsterdam) currently practicing in the Athens Institute of Ophthalmology with special emphasis on Glaucoma, is the Global Project Leader of the World Glaucoma Day and the Executive Vice-Chair of the WGA World Glaucoma Day Committee.