Automatic Algorithm for Monitoring Systolic Pressure Variation and Difference in Pulse Pressure

G Pestel, K Fukui, V Hartwich, P. M Schumacher, A Vogt, L. B Hiltebrand, A Kurz, Y Fujita, D Inderbitzin and D. Leibundgut

Anesthesia & Analgesia, 2009, 108(6), 1823-1829. DOI: 10.1213/ane.0b013e3181a2a8bf


BACKGROUND: Difference in pulse pressure (dPP) reliably predicts fluid responsiveness in patients. We have developed a respiratory variation (RV) monitoring device (RV monitor), which continuously records both airway pressure and arterial blood pressure (ABP). We compared the RV monitor measurements with manual dPP measurements.

METHODS: ABP and airway pressure (PAW) from 24 patients were recorded. Data were fed to the RV monitor to calculate dPP and systolic pressure variation in two different ways: (a) considering both ABP and PAW (RV algorithm) and (b) ABP only (RVslim algorithm). Additionally, ABP and PAW were recorded intraoperatively in 10-min intervals for later calculation of dPP by manual assessment. Interobserver variability was determined. Manual dPP assessments were used for comparison with automated measurements. To estimate the importance of the PAW signal, RVslim measurements were compared with RV measurements.

RESULTS: For the 24 patients, 174 measurements (6–10 per patient) were recorded. Six observers assessed dPP manually in the first 8 patients (10-min interval, 53 measurements); no interobserver variability occurred using a computer-assisted method. Bland-Altman analysis showed acceptable bias and limits of agreement of the 2 automated methods compared with the manual method (RV: –0.33% ± 8.72% and RVslim: –1.74% ± 7.97%). The difference between RV measurements and RVslim measurements is small (bias –1.05%, limits of agreement 5.67%).

CONCLUSIONS: Measurements of the automated device are comparable with measurements obtained by human observers, who use a computer-assisted method. The importance of the PAW signal is questionable.

ASCI-ID: 1372-76