Rate of false positive answers is the only reliable reliability parameter in computerized perimetry

The Science behind the Tip

In computerized perimetry patient?s reliability is measured by 3 parameters, namely rate of false positive and false negative answers (FPA and FNA)(1) and of fixation losses (FL). A FPA is recorded when the patient responds to a zero-luminance (older standard strategies) or responds during periods when no true responses are expected (SITA). FNA are measured by presentation of bright stimuli at locations were threshold sensitivities have been determined. FL is typically monitored using the blind spot method where visible stimuli are presented at location for the blind spot at random intervals.

In eyes with severe visual field (VF) loss covering the blind spot, the rate of FL is easily underestimated. On the other hand, FL is sometimes overestimated when the blind spot has been erroneously located at the start of the test. High rate of FNA is highly correlated to VF loss(2,3). This is explained by the high intra-test variability in the threshold estimates in areas with reduced sensitivity, and cannot be blamed on the patient?s reliability. Rates of FPA show no association to VF status(4).

In the infancy of computerized perimetry visual fields were flagged as unreliable when rates of false answers were =33%. This limit was arbitrary settled when including healthy subjects in the original Statpac interpretation tool. A cut off at 15%, the ?95th percentile in a healthy population, is a more appropriate limit for FPA, while FNA can be neglected in patients with VF loss.

Contributors: Boel Bengtsson, Malmö - Ann Hoste, Antwerp
Co-editors: John Thygesen and Ann Hoste
Peer reviewers: Roger Hitchings and Anders Heijl

References

1. Fankhauser F, Spahr J, Bebie H. Some aspects of the automation of perimetry. Surv Ophthalmol. 1977;22:121-41.

2. Katz J, Sommer A. Reliability indexes of automated perimetric tests. Arch Ophthalmol. 1988;106:1252-4.

3. Bengtsson B, Heijl A. False-negative responses in glaucoma perimetry: indicators of patients performance or test reliability? Invest Ophthalmol Vis Sci. 2000;41:2201-4.

4. Bengtsson B. Reliability of computerized perimetric threshold tests as assessed by reliability indices and threshold reproducibility in patients with suspect and manifest glaucoma. Acta Ophthalmol Scand. 2000;78:519-22.

Tip Reviewer: Roger Hitchings
Tip Editors: Ann Hoste, John Salmon and John Thygesen