Visual estimation of pulse pressure variation is not reliable: a randomized simulation study

Abstract

Pulse pressure variation (PPV) can be monitored several ways, but according to recent survey data it is most often visually estimated ("eyeballed") by practitioners. It is not known how accurate visual estimation of PPV is, or whether eyeballing of PPV in goal-directed fluid therapy studies may limit the ability to blind the control group to PPV value. The goal of this study was to test the accuracy of visual estimation of PPV. Using a simulator program designed by the authors that runs on a PC, 20 residents and 19 attendings were shown five arterial pressure waveforms each with different PPV values (range 1-30 %) moving at one of three sweep speeds (6.25, 12.5, or 25 mm/s) and asked to determine the PPV. There was a weak but significant relationship between true PPV and eyeball PPV (r (2) = 0.22; p < 0.01). The agreement between true PPV and eyeball PPV was 3.3 ± 8.7 %. The mean percent error was 122 %. The rate of correct response group classification was 65 %. Mean percent error was higher the faster the waveform sweep speed (130 % at 25 mm/s vs. 117 % at 6.25 mm/s), and correct responsiveness classification lower (58 % at 25 mm/s vs. 69 % at 6.25 mm/s). The results from this study show that eyeballing the arterial pressure waveform in order to evaluate pulse pressure variation is not accurate.