advertisement

---

Editors Selection IGR 19-4

Visual Function Tests: Pre-attentive visual search

Comment by George Lambrou on:

21523 Diagnostic sensitivity/specificity of preattentive vision tests in glaucoma, Loughman J; Davison P; Flitcroft I, Optometry and Vision Science, 2008; 85: 543-546

Find related abstracts

---

Differential light threshold perimetry ‐ erstwhile as Goldmann perimetry and more recently as standard automated perimetry (SAP) ‐ has been the gold standard glaucoma diagnostic test for decades. Yet, it is widely thought that other ways of exploring the peripheral visual pathways may provide faster, if not more reliable tools for earlier detection and/or more convenient follow-up of glaucomatous functional damage than the 15-minute-per-eye, psychophysical-skill dependent SAP. Hence the flourishing of alternative visual tests in recent years, of which high-pass resolution, SWAP and frequency-doubling technology perimetry are the better-known and more studied examples. Pre-attentive visual search (PAVS) is such an attempt. In spite of its somewhat misleading name (which implies that the subject 'searches' for a visual stimulus) it is based on the parallel processing of information from the entire visual field that results in detection of a peripheral stimulus without serial (foveal) attention or search. In the implementation of the test described by Loughman et al. (1026), the subject is required to detect, out of an array of 120 similar targets (7 min of arc boxes displayed on a computer monitor), a single target that differs from the others by a basic feature, such as flicker, motion or orientation. To compensate for variation in swiftness of reaction, the authors calculated PSI, a performance index taking into account the subject's reaction time.

The excellent results of preattentive visual search need confirmation through larger patient samples and appropriate study designs

As implemented, the test takes between one (for normals) and six minutes (for advanced glaucoma patients) per eye. To evaluate its diagnostic value, the authors examined 123 patients, of which 41 were patients with early glaucoma, 41 were glaucoma suspects and 41 were age-matched normals (the diagnosis had been made by experienced glaucoma specialists). They then used the results from the confirmed glaucoma patient and normal groups to calculate the sensitivity and specificity of the test at difference cutoff values and to draw the ROC curves. Based on this relatively small sample of patients, the diagnostic value of the test was remarkable: the specificities and sensitivities of the PSI index for flicker, displacement and orientation of the target exceeded 95% (and often 99%) while the area under the curve exceeded 0.98 for flicker and displacement reaching a peak 0.997 for the orientation test. If confirmed through larger patient samples, these results are impressive, as they exceed those of traditional function testing like SAP, SWAP or FDT (and ‐ by far ‐ the diagnostic performance of IOP). Further studies, are however required, both cross-sectional and longitudinal, to warrant the use of such tests in clinical management of glaucoma. Furthermore, the mechanisms underlying visual impairment as detected by the PAVS test are not necessarily the same as the ones implicated in glaucoma, and the impact of other eye diseases on the PAVS results need also to be investigated. Visual Function Tests

---

• ← Previous
• Next →

---