For example, the Blue Mountains Eye Study showed that disc haorrhages commonly occur in the general population, not just among people with glaucoma. There is also overlap in clinical test findings between those with glaucoma and those without. The population average intraocular pressure (IOP) of 15.5 mmHg is not very different to the average IOP of 17.5 mmHg among those who develop glaucoma.Glaucoma occurs just as commonly with intraocular pressures under 21 mmHg as it does with pressures above 21 mmHg. You can even have manifest glaucoma without any visible disc cupping, yet be perfectly normal with a cup-to-disc ratio of 0.7 or more.Because no single clinical test or test finding can give us a glaucoma diagnosis, there is considerable uncertainty in identification of the disease. A number of studies have confirmed that more than 50% of glaucoma rains undiagnosed, even among those who have recently had an eye examination.So how are we going to improve the detection rate of this potentially blinding eye disease?{{quote-A:R-W:400-Q: Given a glaucoma prevalence of only 3% in the general population, routine OCT screening is twice as likely to produce a wrong diagnosis than actually detecting glaucoma }}In an admirable but misguided attpt to improve detection rates, there has been a push by a major optometry chain to introduce routine optical coherence tomography (OCT) scans.I understand that staff perform the scans on all comprehensive consultations prior to the patient seeing the optometrist. A yellow or red sector flags the patient as potentially having glaucoma. Threshold visual fields are then performed, again prior to seeing the optometrist.The critical probls with this strategy are the numerous errors that can occur with OCT scan capture and interpretation, particularly when instrument use is delegated to unqualified and under-trained non-ophthalmic staff.Artefacts due to poor head and eye alignment, tilted retinal scans, tear film issues and cataracts all affect scan quality and the resultant thickness data produced. This faulty data is then compared to the instrument’s normative database, and any variance greater than 95% or 99% is flagged in yellow or red respectively.Humans are highly varied and many eyes without glaucoma will show red or yellow sectors due to factors such as high myopia, tilted discs, optic nerve drusen or simply being atypical. The OCT normative database is also limited in the number of subjects, age range and ethnicity.Due to these factors, OCT has a glaucoma false-positive rate of up to 25% (also known as “red disease”), with an average detection sensitivity of only 80%.Given a glaucoma prevalence of only 3% in the general population, routine OCT screening is twice as likely to produce a wrong diagnosis than actually detecting glaucoma. Of a theoretical 100,000 OCT scans performed on patients over the next year, only 3000 patients would truly have glaucoma. About 2500 cases would be correctly detected, while almost 5000 people would be falsely identified.I have spoken with ployee optometrists who’ve expressed concerns about this program, the OCT scan artefacts, unreliable visual field results, and the replacent of clinical decision making by rigid protocols. Test defects cannot be ignored and usually require the patient return for further testing or, more commonly, be unnecessarily referred to ophthalmology at considerable cost to the patient and health care syst.One of the tenets of medicine is “first do no harm”. Routine testing such as PSA blood tests for prostate cancer have rightly been questioned as they oftentimes cause greater harm than good in normals. Routine OCT testing falls into the same category.So what is the astute clinician to do when fishing for glaucoma? Bait your line with a comprehensive history, thorough stereoscopic evaluation of the discs, detailed ocular examination for secondary risk factors, careful gonioscopy, then judicious ordering of ancillary testing such as threshold visual fields and OCT.Glaucoma is diagnosed by the skill, expertise and acumen of the clinician – not by a machine.
 |
Name: Graham LakkisQualifications: BScOptom (UniMelb) GradCertOcTher FACOWorkplace: Lakkis Optometry, University of Melbourne Department of OptometryPosition: Principal Optometrist, Senior FellowLocation: VictoriaYears in the profession: 31 |
