Healio
Healio
July 15, 2006
5 min read
Save

Diagnosis: Screening for glaucoma

Glaucoma develops gradually over a number of years without any obvious symptoms. The most common form of glaucoma in the United States, primary open-angle glaucoma (POAG), is characterized by a loss of retinal ganglion cell axons and often goes unnoticed until irreversible loss of peripheral vision occurs.1 Recommendations published by the U.S. Preventive Services Task Force in 2005 state that good evidence exists to support the use of screening to detect increased IOP and early POAG in adults.2 Evidence also supports early treatment of high-risk adults with elevated IOP to reduce the number of patients who develop visual field defects.2

Findings from the Los Angeles Latino Eye Study (LALES)

Vertical cup:disc ratio (cutoff point >0.6) was the most accurate screening tool with a sensitivity of 92.3% and a specificity of 95.3%. We were surprised that cup:disc ratio was such an effective screening tool.
—Brian A. Francis, MD

Screening is the first step to early detection and may be particularly important in high-risk groups including African-Americans, Hispanics and people over age 60.3 At this year’s American Glaucoma Society (AGS) meeting, Brian A. Francis, MD, and colleagues presented a study on the predictive value of various glaucoma screening tests in 6,082 Latino individuals of primarily Mexican descent aged 40 years and older.4 The population-based sample reported on at the recent AGS meeting was drawn from the population of the Los Angeles Latino Eye Study.5 All individuals underwent the following diagnostic tests: Swedish Interactive Threshold Algorithm (SITA), Standard Humphrey visual field testing, frequency-doubling technology perimetry, IOP measurement, central corneal thickness and simultaneous stereoscopic optic disc photography. To reduce bias toward optic nerve parameters, the investigators required both a visual field defect and an optic nerve appearance consistent with glaucoma to make the diagnosis. The primary outcomes were sensitivity and specificity of the diagnostic tests. Sensitivity in this case refers to the proportion of people with glaucoma in whom the test result is positive and specificity is the proportion of people without glaucoma in whom the test result is negative.6 Vertical cup:disc ratio (cutoff point>0.6) was the most accurate screening tool with a sensitivity of 92.3% and a specificity of 95.3%. The cutoff point of 0.6 for cup:disc ratio was chosen because it represented the 95% cutoff point for the patient population. Screening parameters including IOP, corneal thickness and prior diagnosis had poor sensitivity and good specificity whereas Humphrey visual field analysis and frequency doubling technology perimetry had moderate sensitivity and specificity. The investigators concluded that although the vertical cup:disc ratio appears to be an appropriate screening tool in the Latino population, most glaucoma screening devices remain problematic. They remain optimistic that devices that can measure vertical cup:disc ratios or nerve fiber layer thickness will be helpful in glaucoma screening in the future.

Is the ISNT rule useful?

The diagnosis of POAG is not based on a single test, but rather on the finding of a combination of characteristic changes in the optic disc along with visual field defects.2 Although the determination of cup:disc ratios remains important, it is critical to recognize and understand other structural characteristics of glaucoma. Numerous clinical signs have been suggested to identify glaucomatous optic nerve head abnormalities. For example, when examining the neural rim, consideration of the “ISNT” rule has been suggested.7 The rim width of a healthy optic disc varies by quadrant with the inferior (I) area being the thickest, followed in descending order by the superior (S), nasal (N) and temporal (T) quadrants (see Figure). If observations of the neural rim do not follow this rule, the ophthalmologist may suspect early glaucomatous optic nerve damage.

Figure
Figure. Neural rim observations that do not follow the ISNT rule may indicate early glaucomatous damage.

Figure courtesy of Robert Weinreb, MD

Gadi Wollstein, MD, and colleagues presented their findings of a study that assessed the validity of the ISNT rule in a sample of 96 eyes of 53 consecutive healthy volunteers and 33 eyes of 25 patients with moderate glaucoma.8 All study participants underwent standard visual field testing, stereoscopic disc photography, Heidelberg retinal tomography (HRT) and optical coherence tomography (OCT). Three glaucoma experts independently evaluated the disc photographs for rim thickness by quadrant. Two out of the three graders agreed that the ISNT rule held true in most of the healthy eyes and quantitative OCT retinal nerve fiber layer (RNFL) thickness data confirmed these results. The specificity and sensitivity of the ISNT rule to differentiate between healthy and glaucomatous eyes with respect to overall consensus among graders were 33% and 100%, respectively. However, specificities varied widely among graders (21% to 72%). HRT had a specificity of 25% and a sensitivity of 97% and OCT RNFL data had 43% and 88%, respectively. Although use of the ISNT rule as a diagnostic tool showed high sensitivity to the optic disc abnormality in patients with moderate glaucoma, the low specificity indicates that healthy eyes may not adhere to the ISNT rule. Therefore, this rule is appropriate as a tool to indicate potential optic nerve damage secondary to glaucoma.

The relationship between RNFL and visual fields

It has been proposed that the early detection of structural changes may aid in the identification of patients with glaucoma prior to the appearance of visual field defects. To that end, the structure-function relationship of RNFL to visual field sensitivity was examined in 143 glaucomatous eyes, 193 suspected glaucomatous eyes and 137 normal eyes from a total of 312 patients in a retrospective cross-sectional study.9 OCT was used to measure RNFL within 3 months of visual field testing and the correlation of mean RNFL thickness to mean deviation of the visual field was calculated for each population of patients.

OCT detected early RNFL loss in both glaucoma suspects and patients; however, mean RNFL thickness was significantly thinner in glaucoma patients (64.9 ± 19.5 µ) compared with RNFL thickness in glaucoma suspects (86.5 ± 16.5 µ) or normal patients (96.1 ± 13.8 µ; P < .0001). Mean deviations of the visual field were also significantly lower in glaucoma patients (-8.5 ± 3.9 dB) compared with those in glaucoma suspects (-2.9 ± 2.6 dB) and normal patients (-1.4 ± 1.2 dB; P < .001). Mean RNFL thickness was more highly correlated with mean deviations of visual field testing in glaucoma patients (r=0.70, P < .0001) than in patients suspected of having glaucoma (r=0.29, P < .0001) or normal patients (r=0.07, P=.43).

These results support the hypothesis that RNFL thinning occurs prior to the development of visual defects as measured by perimetry, supporting the use of OCT as a screening tool.

References

  1. Lasik Institute resources page. Lasik Institute web site: Available at: http://www.lasikinstitute.org/Glaucoma_detected.html. Accessed April 28, 2006.
  2. Agency for Healthcare Research and Quality (AHRQ) [database online]. Rockville, MD: U.S. Preventive Services Task Force (USPSTF). Screening for glaucoma: recommendation statement. 2005.
  3. Glaucoma Facts and Stats. Glaucoma Research Foundation Web site. Available at www.glaucoma.org/learn/glaucoma_facts.html. Accessed April 28, 2006.
  4. Francis BA, Varma R, Lai M-Y, Azen SP, Chopra V. Predictive value of glaucoma screening tests in the population-based Los Angeles Latino eye study. Paper presented at: Annual Meeting of the American Glaucoma Society; March 3, 2006; Charleston, SC.
  5. Fraser-Bell S, Wu J, Klein R, Azen SP, Varma R. Smoking, alcohol intake, estrogen use and macular degeneration in Latinos: the Los Angeles Latino Eye Study. Am J Ophthalmol. 2006;141(1):79-87.
  6. Centre for Health Evidence resources page. Centre for Health Evidence Web site. Available at: http://www.cche.net/usersguides/diagnosis.asp.
  7. Weinreb RN. Compliance with medical treatment of glaucoma. J Glaucoma. 1992;1:134-136.
  8. Wollstein G, Pan Y, Ishikawa H, Kagemann L, Gabriele M, Schuman JS. Is the ISNT rule useful for glaucoma detection? Paper presented at: Annual Meeting of the American Glaucoma Society; March 4, 2006; Charleston, SC.
  9. Gupta A, Chaku M, Hughes BA, Juzych MS, Kim C. Relationship of retinal nerve fiber layer to visual field in glaucoma, glaucoma suspect and normal patients. Paper presented at: Annual Meeting of the American Glaucoma Society; March 4, 2006; Charleston, SC.