Statistics from Altmetric.com
A geographical approach to normal IOP interpretation
As the demographic transition leads to progressive population ageing globally, this phenomenon coupled with decreases in the burden of infectious disease and subnutrition in the developing world will directly impact on the epidemiology of eye disease. It is expected that while age related cataract will continue to be the leading cause of visual impairment globally, the prevalence of other age related conditions such as primary open angle glaucoma (OAG), macular degeneration, and diabetic retinopathy will increase.1 Therefore, age related eye disease will grow in importance, both from clinical and public health perspectives. In spite of this, wide gaps still exist in our knowledge about the epidemiology of ocular disease in many regions across the globe. Ongoing events in the Middle East focus the critical spotlight of world opinion on this region. It is therefore refreshing to learn that an enterprising group of researchers has successfully completed a major epidemiological study of eye disease, which promises to provide much welcomed new and important information (this issue of BJO, p 652). As a result of this body of work, we now have access to the first representative population based data on intraocular pressure (IOP) distribution in a Middle Eastern population.
OAG and elevated IOP are often considered to be inextricably linked, but while up to two thirds of individuals with OAG in population based studies have elevated IOPs (that is, 21 mm Hg or higher), relatively few with an elevated IOP have or will in fact develop OAG.2,3 Elevated IOP is a major risk factor for OAG but other risk factors for these conditions differ, suggesting that the underlying aetiological mechanisms could be different.4,5 Major advances have been made in clarifying relations between IOP and OAG, such that OAG is now defined as a group of diseases characterised by optic neuropathy, irrespective of IOP.6 Why then should we wish to learn more about IOP distributions in yet another population? In order to comprehensively manage any disease process, one must be armed with basic epidemiological knowledge about the frequency and distribution of the disease and related risk factors in order to understand the pathophysiology, plan clinical treatment strategies and, at another level, develop optimal public health approaches. The existence of similar risk factors in differing populations aids our understanding of disease aetiology, and it is hoped that the authors will rise to the challenge of presenting truly comprehensive data from their population.
The real potential of the Tehran Eye Study lies in the future, as the researchers follow their cohort prospectively
Compared to most published reports, the Tehran Eye Study population was relatively young (28% aged less than 20 years), and as a result mean IOP was correspondingly lower (14.5 (SD 2.6) mm Hg). Given the obvious age differences, perhaps the comparison of IOP distributions with other epidemiological studies should have been more appropriately restricted to older participants. This is relevant as those aged 40 years and older were noted to have a value for the upper limit of normal IOP (mean IOP (+2 SD) of 20.8 mm Hg compared to 23.9 mm Hg in the Baltimore Eye Study.7 The Tehran Eye Study also confirmed associations between IOP and older age, iris colour, and myopia. Describing the wide variation in IOP distributions across different populations and geographical locations, the authors recommend the use of a “geographical approach to normal IOP interpretation.” Several factors may partly explain this variation, including population demographic factors such as age, sex, race; distribution of disease and physical characteristics including blood pressure, pulse rate, diabetes, obesity, myopia, iris colour, nuclear sclerosis, family history of glaucoma; lifestyle factors such as use of alcohol and smoking; and seasonality, as well as probable genetic factors.2,4,5,8–12 In other words, the paradigm of nature in conjunction with (rather than versus) nurture, and the likely gene-environmental interactions encompassed therein, would explain this variability. Unfortunately the “geographical approach” has not been defined and remains a potential source of controversy. The caveat emptor that caution must be used in interpreting risk factor data based on any form of “geographical approach” must be invoked, given the possible misinterpretation of associations as a result of the ecological fallacy.13 Such an approach would be reasonable if it implies that IOP should be defined according to population specific norms or outcomes,14 given that IOP varies according to the population distribution of risk factors. The traditional IOP cut point of 21 mm Hg may not necessarily be applicable to all populations,2,14 and in the Tehran study fewer than 0.5% of participants had IOP levels which exceeded this value. A useful objective of any “geographical approach” could then be investigations to define population specific IOP treatment thresholds, an outcome likely to have both clinical and public health relevance.
The clinical significance of therapies to reduce IOP has now been underpinned by evidence demonstrating that such interventions decrease visual field loss in people with OAG.15,16 However, conflicting data now exist on their effectiveness to prevent OAG in people with ocular hypertension.17,18 Nevertheless, data on IOP distributions are necessary to inform public health planning on the prevention of associated visual loss and this will become increasingly relevant as populations age. The real potential of the Tehran Eye Study lies in the future, as the researchers follow their cohort prospectively. Every effort must therefore be made to support their achievement of this objective, given the potential to document the natural history of major eye diseases and further elucidate key risk factors for visual loss.3,19,20
Note in Proof
A geographical approach to normal IOP interpretation