Glaucoma Watch prev next

Dec 17 2008

Undiagnosed Open Angle Glaucoma: The Thessaloniki Eye Study

posted by: Suraj Afshar, OD

This cross-sectional, epidemiological study of chronic eye disease in the elderly population of the Greek city of Thessaloniki is one of the largest in Europe and one of the few to identify factors associated with undiagnosed glaucoma. The study was done by associates of the Department of Ophthalmology of the Aristotle University of Thessaloniki, led by Fotis Topouzis, MD, PhD.

Population/ Methods: Initial recruitment of 5000 people was randomly selected from the 321,000 population of Thessaloniki. Of 3617 eligible participants, 2554 people (60 years of age or older) chose to participate in the study.

A short phone questionnaire, which included information regarding when the subject last saw an eye doctor, was administered to a random sub-group of 66% of the participants. Comprehensive history and examination were administered to all study participants.

Definitions of Glaucoma: The study utilized two definitions of glaucoma. The first definition strictly defined glaucoma based on criteria that required both structural and functional damage irrespective of IOP. Structural damage was defined by thinning or notching of the disk rim or C/D asymmetry >20%. Functional damage was confirmed with confirmed threshold glaucomatous VF defect. A second definition permitted classification of glaucoma on clinical judgment even though the strict criteria were not fulfilled (such as unreliable or unusable VF, biomicroscopy, or ophthalmoscopy or structural changes without functional confirmation).

Subjects were additionally classified as POAG when presenting with open chamber angle without secondary causes or PEXG when presenting with the presence of pseudoexfoliative material of the crystalline lens.

Analysis: Regression analyses were performed to assess potential factors associated with undiagnosed glaucoma. Included were: age, gender, family history, cataract surgery, monocular VA / IOP / vertical c-d ratio / AGIS VF score differential, and type of OAG (POAG / PEXG). Limited analyses also focused on respondents to the short questionnaire regarding time since last doctor visit.

Results: Using definition 1, the incidence of OAG (POAG and PEXG) was 3.8% (96 out of 2554) with 52.1% of those (50 out of 96) previously undiagnosed. The incidence of OAG (POAG and PEXG) using definition 2 was 5.5% (141 out of 2554) with 50.4% of those (71 out of 141) previously undiagnosed. The undiagnosed OAG findings were significantly more likely in males, in lower vertical c/d ratios, and those without a confirmatory family history. Notably, a lower percentage had not been seen by an eye doctor in the past year.

The incidence of undiagnosed POAG was higher than undiagnosed PEXG using each study definition.

Analysis also revealed a 3-4 times increased risk of undiagnosis of POAG when compared to PEXG both when including or excluding the variable of time from last doctor appointment.

Conclusions: The Thessaloniki Eye Study is certainly one of the most comprehensive and largest studies in Europe. Remarkably the study found the percentage of undiagnosed glaucoma cases very similar to that found in other large studies such as Baltimore (50% among whites), Blue Mountain (51%), Barbados (51%), Roscommon (49%), and Rotterdam (53%). Despite differences in population, definitions of glaucoma and methodologies in examination, the percentage of undiagnosed glaucoma remains the same.

Interestingly, POAG had a higher rate of being undiagnosed as compared to PEXG by 22%. As discussed by the authors, the lower rate for PEXG may be attributed to higher incidence of cataract and IOP. The diagnosis may result from modification of clinical assessment and higher level of suspicion coupled with more frequent evaluations.

For eye care professionals, the study highlights factors associated with lack of diagnosis and urges a refinement of strategies to improve glaucoma detection.

Citation: Fotis Topouzis, Anne L Coleman, Alon Harris, Archimidis Koskosas, Panayiota Founti, Gordon Gong, Fei Yu, Eleftherios Anastasopoulos, Theofanis Pappas, and M. Roy Wilson. “Factors Associated with Undiagnosed Open-Angle Glaucoma: The Thessaloniki Eye Study.” American Journal of Ophthalmology, February 2008 (Vol 145, No. 2), pp327-335.

Eric J. Conley, OD - on Jun 26, 2009

In reading the details found within the journal article highlighted above, it is evident that significant efforts must be made to educate the population better as to the risks of glaucoma and the need for regular eye evaluations. The results here are not surprising (POAG undiagnosed more often the PXG) in that noticed PXE findings would certainly focus any practitioner to 'more' carefully (as a result of introduced bias) evaluate the patient for glaucoma factors (VF, HRT, etc) that might not otherwise occur in the examination. In my experience, a significant number of clinicians, both OD and OMD alike, are failing to make the appropriate open-angle glaucoma diagnosis secondary to one or more of the following reasons: poor understandings of the physical (NFL, ONH, GCC, etc.) glaucomatous changes in the eye under glaucoma stress, poor stereoscopic Slit-lamp examination techniques/equipment (using a 90D v. 60D lens) and/or reduced examination times per patient encounter. So perhaps the more important question that should be addressed is how often existent glaucoma is not diagnosed despite the patient visiting an eyecare provider regularly. A paper by Wong et al (referenced in the Topouzis et al article above) attempted to address this very point. Unfortunately, its findings suggest that a visit to an OD rather than OMD leads to increased non-diagnosis. Of course, the comparison likely does not apply across all continents as prescriptive authority for OD's has significant variability and as such the ocular examination therefore also varies (dilated vs. undilated, etc.) We find similar disparities between OD and OMD outcomes in the Barbados Eye Study which also suggested this disparity in under-diagnosis when evaluated by ODs (not permitted by law to dilate their patients) To conclude, I believe that we can take home several points from these articles to benefit our patients: 1) The Thessaloniki Eye Study noted that undiagnosed patients were more likely to have smaller c/d ratios -- Missed diagnosis in glaucoma happens when clinicians fail to measure the size of the overall disc. Size matters, possibly more than any other factor, when evaluating C/D ratio relevance. Smaller nerves should have a smaller C/D, Average should be average (0.4) and larger nerves should have larger cupping. See the Glaucoma Handbook by Anthony Litwak OD 2) Population based studies have recognized a disparity in the ability of ODs to recognize Glaucoma vs. their OMD counterparts. Here is the caveat though, in these studies, the ODs did not have prescriptive authority to dilate their patients. The take-home message is that every new patient requires a dilated fundus exam with careful nerve examination preferably with a 60D, and if necessary up to 78D, lens. An undilated, monocular viewpoint will almost always result in C/D judgement errors (under-estimations) due to missed sloping contours and unreliable color ques. Undilated examinations were very likely the reason for the disparity contributing to non-diagnosis among the OD groups in these studies. Detection of glaucoma by eye health professionals is an area we can certainly improve upon with continued education and technological advancements. Please feel free to respond to this message to discuss the points further. Best regards, Eric J. Conley OD Assistant Professor and Director of Continuing Education Illinois College of Optometry 3241 S. Michigan Ave Chicago, IL 60616