advertisement

Editors Selection IGR 17-3

Progression: Natural history of OAG

Comment by Kouros Nouri-Mahdavi on:

25002 Natural history of open-angle glaucoma, Heijl A; Bengtsson B; Hyman L et al., Ophthalmology, 2009; 116: 2271-2276

Find related abstracts

Heijl et al. (1529) have provided us again with important data from the Early Manifest Glaucoma Trial, this time with regard to the natural history of untreated open-angle glaucoma. The results are based on 118 out of 126 patients (94% retention) from the original EMGT cohort who remained untreated for at least six years or until progression was established. Rates of progression (expressed as decibels/year) and frequency of progression (based on the well-known EMGT criteria) were compared among high-tension (HTG), normaltension (NTG), and exfoliative glaucoma (PXG) groups. The authors reported that the NTG group had the slowest progression rate (median: -0.22 dB/year) and the least frequency of progression (56%) at the end of six years, while the PXG group demonstrated the highest rate (median: -1.13 dB/year) and frequency (93%) of progression with HTG eyes falling in between (median progression rate of -0.46 dB/year and 74% progression at six years). This is despite the fact that PXG eyes had the same baseline IOP as POAG eyes. Patients older than 68 years (the median age in the study cohort) progressed much faster compared to younger patients. A significant amount of variability was observed within the three groups meaning that we will not still be able to predict with great confidence the future rates of progression in individual patients according to the diagnosis.

While the data are consistent with the scarce available evidence in the literature on the natural history of untreated glaucoma, a few things should be kept in mind before the results are generalized to similar patient populations. The definition of NTG vs. HTG is quite arbitrary in general and was based on two baseline IOP measurements in EMGT. No formal IOP curve was done. So one would expect that there would be some overlap between the two diagnostic groups. The authors looked at progression rates in univariate analyses although, to their credit, they did look at the distribution of two other potential baseline risk factors for progression, baseline IOP and age. It would have been interesting to compare the rates in a multivariate model where the rates are adjusted for follow-up IOP and baseline field damage (the other risk factors found in previous EMGT studies) in addition to baseline IOP and age to see whether the rates of progression would still be the same. That said, however, I would expect a fairly high degree of correlation between the baseline and follow-up IOPs. Also, the number of PXG eyes was small potentially limiting the power of such analyses.

Eyes with PXG are more likely to develop or have cataract progression or surgery, which could confound detection of glaucoma progression in trend analyses. However, given the results are consistent between the MD trend analysis and EMGT criteria, which at least partially correct for the influence of media opacity, this does not seem to be a major issue.

The EMGT database could also be used to answer the question whether, once adjusted for IOP, the progression rates are really different between HTG and NTG patients or we are just dealing with the same disease which progresses more slowly at lower pressures since there is less of the main risk factor for the disease. The last issue that needs to be considered is the fact that neither the MD trend analysis nor the EMGT criteria are location-sensitive. The functional impact of glaucoma is dependent on the location of field defects. Although NTG eyes may be progressing at a slower pace, given the existing evidence that the paracentral regions of the field are more likely to be involved in such eyes, the impact of progression may be different compared to HTG eyes.

I would like to commend the authors for providing us with invaluable information that will keep us thinking for a while and which will likely not be reproduced in our lifetime.

• ← Previous
• Next →