advertisement
Editors Selection IGR 9-4
Epidemiology: Glaucoma as cause of blindness
Comment by David Friedman on:
55778 Long-term trends in glaucoma-related blindness in olmsted county, Minnesota, Malihi M; Moura Filho ER; Hodge DO et al., Ophthalmology, 2014; 121: 134-141
See also comment(s) by Kaweh Mansouri •Malihi and colleagues report on the population incidence of glaucoma blindness in Olmstead County near the Mayo Clinic comparing the time period from 1965-1980 to the time period from 1980-2009. Over 90% of the residents are White and therefore the findings do not apply to the entire population of people with glaucoma. Data for this community-based study come from an integrated system to gather medical records from the major healthcare providers in the area. Nearly all people living in the county receive healthcare at Mayo or one of its satellites.
The authors enrolled only newly-diagnosed patients and included primary open-angle glaucoma, pseudoexfoliation glaucoma, pigmentary glaucoma, and treated hypertension in the definition of 'glaucoma'. Gonioscopy is not mentioned in the paper. Blindness was based on visual acuity or visual field loss and had to be attributable to glaucoma. Cases with other causes of blindness were excluded, as were 'ambiguous cases'.
The incidence of bilateral blindness was 4.3% at 20 years in the more recent time period, half the previous estimate of 9.0%. Similarly, the incidence of monocular blindness decreased from 25.8% to 13.5% across the two time periods. This is a dramatic decline in blindness rates.
Why was there such a steep drop? Firstly, the finding may be real and better treatments may have led to better outcomes in patients diagnosed with glaucoma. Secondly, individuals with glaucoma may have been diagnosed earlier in the disease process in the latter period and this would have led to a lower rate of developing blindness. No comparisons of visual field loss are possible because automated perimetry was not available at the start of the study period (some were even diagnosed with tangent screens). The authors take pains to try to show that this lead-time bias is not present, but it is possible that earlier detection had a role in the drop in rates. A third possibility is better documentation of other causes of blindness. Charts from the earlier time period had less detailed recording of data and it is likely that other causes of blindness (AMD, vein occlusions, and so on) were not recorded as well in the initial time period. This would have biased the results to more 'glaucoma' blindness. Finally, it is not clear how changing definitions and the inclusion of treated ocular hypertension in the definition of glaucoma affected the results. It is possible that an excess of these ocular hypertensives in the latter time period would have resulted in a less diseased baseline population and this would have led to lower blindness rates.
Carrying out community-based research is not easy and charts are not always complete
In summary, carrying out community-based research is not easy and charts are not always complete. Despite that, the authors have done an excellent job of providing estimates for blindness rates among White patients with diagnosed glaucoma in Olmstead County. Fortunately for our patients, the rate of blindness appears to be substantially lower than previously reported!
Comments
The comment section on the IGR website is restricted to WGA#One members only. Please log-in through your WGA#One account to continue.Log-in through WGA#One
