Compass New Grid With 24 Center Points Boosts Glaucoma Testing Specificity

Female ophthalmologist wearing face mask checking eyesight of elderly man looking into tonometer. Woman ophthalmologist examining eye of a senior patient in her clinic.
Technology detects central visual field defects with higher specificity and lower sensitivity than previous equipment.

The most thorough screening for early glaucoma may well be to administer 2 visual field (VF) grids per eye; 24-2 for examining peripheral regions, and 10-2 to find any initial central visual field defects (CVFD). A study in Ophthalmology Glaucoma reviewed another option — the Compass New Grid (NG). That technology’s framework is similar to the Humphrey Visual Field Analyzer 24-2, in that it covers a 24˚ area in 65 locations; but, it also assesses 24 central points instead of an equidistant network of 12, according to a study published in Ophthalmology Glaucoma.

Compass automated fundus perimeter (Centervue SpA) automated fundus perimeter  employs an ophthalmoscope and eye tracker for fixation stability and improved stimuli delivery. Compass perimeter functions using the NG grid, as well as 24-2 (24 degrees in 54 locations), and 10-2 (10 degrees in 68 locations). Previous studies have compared Humphrey Visual Field Analyzer 24-2 with Compass 24-2, and suggest the 2 show comparable sensitivity and nearly equivalent defect discrimination, the current analysis explains.

This cross-sectional investigation of 123 eyes was conducted at the University of California San Diego in La Jolla, Calif., with participants enrolled in Diagnostic Innovations in Glaucoma Study (DIGS). The 3 study cohorts included 32 patients diagnosed with early glaucoma, 17 in a glaucoma suspect group, and 14 healthy individuals. All participants underwent a comprehensive ocular evaluation comprising slit-lamp biomicroscopy, intraocular pressure measurement, and other tests. Compass 10-2 and NG automated perimetry were performed approximately 1 week apart, and Spectralis (Heidelberg Engineering GmbH) macular optical coherence tomography (OCT) within 1 year.

For 62 eyes with no OCT-identified macular damage, Compass 10-2, using pattern deviation (PD) plot, assessed superior hemifield CVFD at a specificity of 0.36, and Compass NG showed a specificity of 0.81. For inferior hemifield CVFD — also using PD — 10-2 had a specificity of 0.52 and NG a specificity of 0.86. Alternatively, for 46 eyes with OCT-imaged macular damage, sensitivity for revealing a superior defect was 0.78 in 10-2 grid, and 0.48 with NG. Sensitivity in the inferior hemifield was 0.52 via 10-2, and 0.20 with NG.

Agreement between Compass 10-2 and NG to find early glaucomatous trouble spots in the central 10 degrees was “fair to moderate” using either PD or total deviation (TD) plot, the investigation explains. “Although the Compass NG detected fewer central visual field defects than the 10-2, it did detect CVFDs that were not observed in the Compass 24-2 test in early glaucoma patients,” according to the research.

The mean time it took participants to complete the grids was similar: 7.60 minutes for 10-2, and 7.55 minutes for NG (P =.35). This study was limited by a relatively small sample. Also, it is possible that “false positives” may have actually been true positives the OCT did not yet pick up. 

The analysis speculates that Compass 10-2 may be more helpful to detect early glaucoma in a young patient because it has higher sensitivity than NG. “In contrast, for glaucoma screening of a general population, a clinician might consider the use of NG because of its higher specificity,” the investigators report.

Disclosures: Some of the study authors have declared affiliations with biotech, pharmaceutical, medical device companies, and/or private or public research organizations. Please see the original reference for a full list of authors’ disclosures.


El-Nimri NW, Penteado RC, Bowd C, et al. Agreement between Compass Fundus Perimeter New Grid and 10-2 Testing Protocols for Detecting Central Visual Field Defects. Ophthalmol Glaucoma. Published online July 20, 2021. doi:10.1016/j.ogla.2021.07.006