To the Editor.
—Heijl et al1 have clearly shown the usefulness of empiric total deviation probability maps (TDPMs) and pattern deviation probability maps (PDPMs) in visual field interpretation. However, since the PDPM allows the clinician to "focus on localized field loss as opposed to diffuse loss of sensitivity," all too often I have seen clinicians relying on the PDPM to identify visual field loss, while disregarding the TDPM and the reliability criteria (<20% fixation losses, <33% false-positive errors, and <33% false-negative errors). Heijl and colleagues presented an example of significant, diffuse depressions in the TDPM but normal-appearing PDPM. The converse situation of a normal TDPM but falsely abnormal PDPM can occur as shown below.In situations with high fixation losses or high false-positive errors, the patient's measured threshold values (Figure, A) at several test points may be higher than the age-corrected normal threshold values, as reflected by the positive
Johnson LN. Visual Field Interpretation With Empiric Probability Maps. Arch Ophthalmol. 1989;107(10):1423. doi:10.1001/archopht.1989.01070020497016