Technical Note: How many readings are required for an acceptable accuracy in pulsatile ocular blood flow assessment?

PURPOSE: Pulsatile ocular blood flow (POBF) assessment aids the diagnosis of ocular diseases with defective ocular haemodynamics, such as glaucoma. Although each successful POBF measure given by the instrument represents five 'repeatable' pulses, there has been no study verifying how repeatable they are. There is also no report on the minimal number of measurements for an acceptable accuracy. METHODS: Forty-three healthy young subjects were recruited and each subject had five consecutive POBF measurements obtained from one randomly selected eye. The coefficient of variation was calculated from the raw data of the five 'repeatable' pulses. The average from five consecutive measurements was considered as the standard for comparison with the first, average of the first two, the first three and the first four measurements. The 95% limits of agreement were determined using the Bland and Altman approach. RESULTS: The coefficient of variation was greater than the manufacturer's claim of within 10%. The mean (+/-S.D.) POBF calculated from five consecutive measures was 732.5 +/- 243.2 microL min(-1). The mean (+/-S.D.) difference between the standard POBF and the first, average of the first two, the first three and the first four measurements was (in microL min(-1)): 12.5 +/- 59.8, 7.8 +/- 42.1, 9.6 +/- 32.5 and 3.7 +/- 19.6 respectively. The corresponding 95% limits of agreement were (in microL min(-1)): +/-117.2, +/-82.6, +/-63.8 and +/-38.4 respectively. CONCLUSIONS: As the five 'repeatable' pulses were not as repeatable as the manufacturer claims, practitioners should not rely on one single POBF measure. An average of three consecutive measurements will be adequate to detect the minimum reported difference in POBF between glaucoma and normal patients.