At the completion of this article, the reader should be able to improve their management of patients with glaucoma. Including:
• Refine their use of normative data base to reliably assess glaucoma
• Prioritise the accumulation of data points in progression analysis before making therapeutic changes to treatment
• Recognise the impact of neurodegeneration in the assessment of glaucoma with visual fields
• Understand the role of laser iridotomy in the treatment of patients with angle closure glaucoma
A successful optometric practice in the 21st century depends on the clinical command of an ever-advancing range of diagnostic technology. While so much about glaucoma care is heading in the right direction, an over-reliance on technology and a lack of corroborating clinical evidence can lead to errors, DR JOSEPH SOWKA explains.
Dr Joseph Sowka
OD, FAAO, Diplomate
Glaucoma can be diagnosed by observations of characteristic changes in the optic disc and retinal nerve fibre layer, abnormalities in threshold perimetry, alterations in structure demonstrated on optical coherence tomography (OCT) and assessment of risk factors such as intraocular pressure (IOP) and family history of the disease. Diagnosing and managing patients with glaucoma can be a challenging task.
Therapeutic intervention, on the other hand, is generally straightforward; that is: reduction of IOP with medicines, lasers, and/or surgery. However, errors in diagnosis and therapeutics can make glaucoma management an arduous task. Take care not to make these common errors.
Number 1: the OCT is wrong
There are several issues in imaging that make OCT assessments for glaucoma very suspect and even misleading. A relatively limited normative database (against which the patient’s measurements are compared), signal quality, blinks and saccades, segmentation errors, media opacities, and an abnormal axial length can all contribute to induced false measurements on an OCT.1-3
When interpreting an OCT printout, ensure the quality score (as indicated for each specific proprietary device) has been met at a minimum. Look to see that there is proper illumination and clarity of focus and the optic disc image is properly centred with no missing data.
Inspect the scan for signs of eye movement. Look to see how the device has segmented the individual layers to ensure that no artificial errors have been introduced. Posterior vitreous detachments and other vitreal issues may confuse the device and make it seem that it is measuring tissue that really isn’t there.
Finally, if using any macular scans or ganglion cell analysis measurements, ensure that there is no concurrent macular disease. If there is, then do not use this potentially misleading information.
Number 2: treating ‘red disease’
Most OCT printouts colour code results as to degree of statistical significance. Common coding uses green to connote the patient’s measured data to be within 95% confidence interval, red to indicate when findings would occur normally in just 1% of the population, and yellow to indicate all intervening values with borderline significance.
In that, each OCT manufacturer employs a relatively limited normative data base to compare against. There commonly will be situations where a patient’s measured data falls outside the device’s normative data base, yet the patient may be completely healthy and normal. Just because a patient’s measured information falls outside the 99% level doesn’t mean that there is disease present. In this instance, much of the printout will be coded as abnormal in red, yet there is truly no disease present. This is commonly referred to as ‘Red Disease’.4
The use and overemphasis of imaging technology to the exclusion of additional clinical findings and assessment of risk will put patients in peril. All imaging technology must be interpreted in context with other clinical findings and when the OCT results do not fit with known correlates of glaucoma, the results should be interpreted with caution.
Number 3: not treating ‘green disease’
Similar to Red Disease discussed above, there may be instances where patient data may fall within the OCT normative confidence interval with results printed in green, yet have clinically, ophthalmoscopically visible damage to the retinal nerve fibre layer (RNFL) and functional loss on threshold perimetry. This commonly occurs when inspecting the Quadrant and Clock Hour Graphs on OCT. When the RNFL analysis is divided into four quadrants or 12 clock hours, it must be remembered that considerable area is being averaged to give these sector values.
A focal RNFL defect may be present ophthalmoscopically, but when averaged in with adjacent healthy tissue on OCT, the value may fall within the device’s normative data base.5 Thus, everything is printed in green, giving a false sense of security in an eye that truly has structural damage. Thus, it is important to weigh the OCT results against the optic disc photographs and clinical examination to ensure that ‘Green Disease’ is not missed.6
Number 4: changing therapy based upon one bad finding
IOP measurements and visual field results can be variable, especially when one considers patient compliance with medications and the psychophysical responses in threshold perimetry. Patients often overstate adherence to medical therapy. Even when not trying to be intentionally misleading, many patients may not correctly remember if they used their medication properly immediately before the examination. Medicines don’t fail overnight.
A medically-adherent patient will not have an IOP of 15 mm Hg on one visit and 30 mm Hg on the next visit due to medicine failure or progressing trabecular dysfunction. There will be a slow, progressive upward drift of IOP in cases where medicines are failing to control IOP. Always insist at least two IOP readings above target (and preferably three) before making any therapeutic changes. Similarly, visual field changes occur frequently, but shouldn’t be considered progression unless the change is verified in a subsequent (and preferably two) visual field. Over 80% of abnormal visual fields noted in the Ocular Hypertension Treatment Study were not verified on repeat testing. Always look for a sustained decrease in visual field results before changing therapy.
Number 5: insufficient pre- and post-treatment IOPs
Unless a patient presents with very high IOP (e.g., above 45 mm Hg) or has advanced disease (with loss of central visual acuity or relative afferent pupil defect in an eye), there is generally no need to rush to treat chronic open angle glaucoma. It’s beneficial to get several IOP readings (at least two and preferably three) before initiating treatment of any kind. At one visit, the patient may be exhibiting a peak IOP or a trough reading. Knowing the range is very important.
Similarly, one should never prematurely judge efficacy of treatment based upon the IOP reading immediately after initiation of therapy. Even if the first IOP measurement after initiating therapy isn’t impressively lowered, consider leaving therapy unchanged and check at least one more time before deciding if a medication is truly efficacious or not.
In the example in Figure 5, there was an abrupt IOP drop after the initiation of therapy. However, it is notable that there are several pre-treatment IOPs that are nearly identical to the post-treatment IOPs, indicating that the prescribed medication, while overall effective, doesn’t consistently give the robust pressure reduction initially seen.
Number 6: not recognising a neurologic visual field
One of the most insidious situations in eyecare occurs when a patient with glaucoma manifests a neurologic disease concurrently. While glaucoma causes arcuate visual field defects that respect the horizontal meridian and neurologic conditions cause hemianopic defects that respect the vertical meridian, such patterns can get lost within the same patient.7,8
There are two ways to discern these differences. First, a glaucomatous visual field can be predicted by the optic disc and RNFL appearance. When the field loss is greater than expected and, often in an area not anticipated based upon the optic disc appearance, one should look for the neurogenicity by examining carefully both visual fields.
Additionally, the Greyscale printout is exceptional at identifying visual field defects which respect the vertical meridian while the Pattern Deviation can be quite poor. Further, while glaucomatous and neurologic damage can occupy the same quadrant, neurological defects may actually manifest a deeper scotoma within a glaucomatous defect.
Figure 6 shows a 74-year-old woman previously diagnosed with glaucoma. She had optic nerve and RNFL damage consistent with glaucoma. However, observation of the Greyscale showed bitemporal visual field defects that respected the vertical meridian. The fields were repeated and the pattern persisted. Ultimately, she was diagnosed with a pituitary macroadenoma and scheduled for neurosurgical intervention.7
Figure 7 is a 65-year-old female also previously diagnosed with glaucoma. Her visual fields showed bilateral inferior defects. The left visual field matched extreme superior disc damage in that eye. In the right eye, her superior field defect matched optic disc and RNFL damage, but there was no structural abnormality to explain her inferior visual field loss. Most notable was the fact that the right inferior visual field defect stopped at the vertical meridian on the Greyscale in the right eye. While the left eye had a significant inferior arcuate scotoma, it was notable that the left inferior nasal defect was absolute and much deeper that the remainder of the field loss. This led to the observation that she had not only glaucomatous arcuate visual field defects, but also a superimposed right inferior quadrantanopsia. Subsequent neuroimaging revealed an ischemic cerebral infarct.
Number 7: thinking laser iridotomy is best manner to treat angle closure
In Figure 8, this 74-year-old woman presented with blurred vision. She is highly hyperopic (spherical equivalent of +5.00 diopters) and correctable to 6/19 OD and 6/6 OS with marked nuclear cataracts, right being worse than left. Her IOP is 30 mm Hg OD and 25 mm Hg OS. Gonioscopically, she is open to anterior trabecular meshwork nasally and temporally in both eyes with no visible angle structures in any other quadrants. Indentation gonioscopy demonstrates no peripheral anterior synechiae. Anterior segment OCT shows dramatically occludable angles. She was diagnosed with primary angle closure in each eye, prescribed bimatoprost 0.01% QHS OU and referred, not for laser iridotomy, but lens extraction. Upon surgical consultation, her IOP is lowered to 17 mm Hg OU and she underwent uncomplicated cataract extraction with resolution of her angle closure.
This patient’s management is well supported by The Effectiveness in Angle Closure Glaucoma of Lens Extraction (EAGLE) study, which was a prospective, randomised clinical trial, indicating that lens removal is a viable option for chronic angle closure. It compared the safety and effectiveness of laser iridotomy and subsequent medical therapy to clear lens extraction for patients with newly diagnosed chronic angle closure glaucoma. It was seen that patients undergoing phacoemulsification lens extraction needed far fewer IOP controlling meds than those undergoing iridotomy and had a comparatively insignificant need for additional trabeculectomy to control IOP.9
Figure 9 shows a 30-year-old woman with moderate hyperopia (+2.00 diopters spherical equivalent) and normal IOP (20 mm Hg OU) who has been diagnosed in 2018 with primary angle closure and had undergone laser iridotomy. While the procedure was performed successfully, the status of her angles did not significantly change and she was subsequently diagnosed with plateau iris syndrome. At that point, the options suggested to her included chronic pilocarpine use, laser iridoplasty, and clear lens extraction, all of which were deemed poor choices for a young woman.
She had laboured under the admonition that she could never be dilated. In 2022, she presented for eyecare and had not been dilated for four years. After a discussion of risks and benefits, she was reappointed for an early morning dilation. She returned to the office throughout the day for periodic evaluations until her dilation wore off. Throughout this time, her IOP never increased by more than 1 mm Hg in either eye. She was educated about signs and symptoms of acute angle closure and it was agreed that all future dilated examinations would follow a similar pattern.
It has come to light that possibly laser iridotomy may not be as necessary as once thought in patients with potentially occludable angles. A study performed in China examined 889 angle closure suspects where one eye received laser iridotomy and the other was observed without treatment. At 72 months, angle closure, as defined by IOP > 24 mm, development of at least one clock hour of peripheral anterior synechia, or an acute attack, was very infrequent. These endpoints occurred in 4.19 per 1000 treated eyes/yr and 7.97 per 1000 untreated eyes/yr (19 treated eyes and 36 untreated eyes). Acute angle closure occurred in only five untreated patients and one laser iridotomy treated patient.
While prophylactic laser iridotomy statistically significantly reduced incidence of development of angle closure glaucoma, the authors determined that laser peripheral iridotomy was not justified.10 In a subsequent analysis of these patients followed over time, it was seen that the incidence of acute angle closure attacks after pharmacologic dilation was low even in this high-risk group and prophylactic laser iridotomy was not recommended.11
Conclusion
Glaucoma diagnosis and management can be quite challenging. It is important to be aware that there are common errors that can make management much more challenging. It behooves the careful practitioner to keep these possible errors in mind when diagnosing and treating patients with glaucoma.
References:
1. Suwan Y, Rettig S, Park SC, et al. Effects of circumpapillary retinal nerve fiber layer segmentation error correction on glaucoma diagnosis in myopic eyes. J Glaucoma. 2018 Nov;27(11):971-975.
2. Giani A, Cigada M, Esmaili DD, et al. Artifacts in automatic retinal segmentation using different optical coherence tomography instruments. Retina. 2010 Apr;30(4):607-16.
3. Miki A, Kumoi M, Usui S, et al. Prevalence and Associated Factors of Segmentation Errors in the Peripapillary Retinal Nerve Fiber Layer and Macular Ganglion Cell Complex in Spectral-domain Optical Coherence Tomography Images. J Glaucoma. 2017 Nov;26(11):995-1000.
4. Chong GT, Lee RK. Glaucoma versus red disease: imaging and glaucoma diagnosis. Curr Opin Ophthalmol. 2012 Mar;23(2):79-88.
5. Hwang YH, Kim YY, Kim HK, Sohn YH. Agreement of retinal nerve fiber layer defect location between red-free fundus photography and cirrus HD-OCT maps. Curr Eye Res. 2014 Nov;39(11):1099- 105.
6. Sayed MS, Margolis M, Lee RK. Green disease in optical coherence tomography diagnosis of glaucoma. Curr Opin Ophthalmol. 2017 Mar;28(2):139-153.
7. Thomas R, Shenoy K, Seshadri MS, et al. Visual field defects in non-functioning pituitary adenomas. Indian J Ophthalmol. 2002 Jun;50(2):127-30.
8. Tanaka K, Wada I, Suenaga T. Incongruous, incomplete, homonymous hemianopia due to an infarction localized to the lateral geniculate body. Rinsho Shinkeigaku. 2017 Oct 27;57(10):595-598.
9. Azuara-Blanco A, Burr JM, Cochran C, et al. Effectiveness in Angle-closure Glaucoma of Lens Extraction (EAGLE) Study Group. The effectiveness of early lens extraction with intraocular lens implantation for the treatment of primary angle-closure glaucoma (EAGLE):The Lancet. Volume 388, No. 10052, p1389–97.
10. He M, Jiang Y, Huang S, Chang DS, Munoz B, Aung T, Foster PJ, Friedman DS. Laser peripheral iridotomy for the prevention of angle closure: a single-centre, randomised controlled trial. Lancet. 2019 Apr 20;393(10181):1609-18.
11. Friedman DS, Chang DS, Jiang Y, Huang S, Kim JA, Munoz B, Aung T, He M, Foster PJ. Acute Angle-Closure Attacks Are Uncommon in Primary Angle-Closure Suspects after Pharmacologic Mydriasis: The Zhongshan Angle-Closure Prevention Trial. Ophthalmol Glaucoma. 2022 May 11:S2589- 4196(22)00082-5.
