P G D Spry1, H M Hussin, J M Sparrow. 1. Bristol Eye Hospital, Lower Maudlin Street, Bristol BS1 2LX, UK. paul.spry@ubht.swest.nhs.uk
Abstract
AIMS: To evaluate performance of frequency doubling technology (FDT) perimetry using the Humphrey Matrix 24-2 thresholding program in a hospital eye service (HES) glaucoma clinic. METHODS: A random sample of individuals referred consecutively to the HES for suspected glaucoma were examined with 24-2 threshold FDT in addition to routine clinical tests. The discriminatory power of FDT and standard automated perimetry (SAP) were assessed using glaucomatous optic nerve head appearance as the reference gold standard. RESULTS: 48 of 62 eligible referred individuals were recruited. Glaucoma prevalence was 31%. Median test duration per eye was 5 minutes 16 seconds for FDT and 5 minutes 9 seconds for SAP. There was no significant difference (p = 0.184) between proportions of individuals with reliable test results (FDT 75%, SAP 63%). Using a clinically appropriate binary criterion for abnormal visual field, sensitivity and specificity levels were 100% and 26% respectively for FDT and 80% and 52% for SAP. Both tests had higher negative than positive predictive values with marginal differences between tests. Criterion free receiver operator characteristic analysis revealed minimal discriminatory power differences. CONCLUSIONS: In a HES glaucoma clinic in which new referrals are evaluated, threshold 24-2 FDT testing with the Humphrey Matrix has performance characteristics similar to SAP. These findings suggest threshold testing using the FDT Matrix and SAP is comparable when the 24-2 test pattern is used.
AIMS: To evaluate performance of frequency doubling technology (FDT) perimetry using the Humphrey Matrix 24-2 thresholding program in a hospital eye service (HES) glaucoma clinic. METHODS: A random sample of individuals referred consecutively to the HES for suspected glaucoma were examined with 24-2 threshold FDT in addition to routine clinical tests. The discriminatory power of FDT and standard automated perimetry (SAP) were assessed using glaucomatous optic nerve head appearance as the reference gold standard. RESULTS: 48 of 62 eligible referred individuals were recruited. Glaucoma prevalence was 31%. Median test duration per eye was 5 minutes 16 seconds for FDT and 5 minutes 9 seconds for SAP. There was no significant difference (p = 0.184) between proportions of individuals with reliable test results (FDT 75%, SAP 63%). Using a clinically appropriate binary criterion for abnormal visual field, sensitivity and specificity levels were 100% and 26% respectively for FDT and 80% and 52% for SAP. Both tests had higher negative than positive predictive values with marginal differences between tests. Criterion free receiver operator characteristic analysis revealed minimal discriminatory power differences. CONCLUSIONS: In a HES glaucoma clinic in which new referrals are evaluated, threshold 24-2 FDT testing with the Humphrey Matrix has performance characteristics similar to SAP. These findings suggest threshold testing using the FDT Matrix and SAP is comparable when the 24-2 test pattern is used.