Assessment of fluid responsiveness using cardiac power index in the prone position: a pilot study

Abstract

Background Fluid therapy guidelines tend to be evaluated using indicators representative of preload, and they have yet to be assessed from the standpoint of contractility. We hypothesized that the cardiac power index (CPI) might affect fluid responsiveness in settings where postural change triggers hemodynamic changes. Methods Twenty-eight patients who underwent lumbar spine surgery in the prone position on a jack’s table were enrolled in this study. Hemodynamic variables were measured before and after the position change and after the colloid bolus (5 mg/kg) in the prone position. Fluid responsiveness was defined as an increase in the stroke volume index (SVI) ≥ 10%. Results After a position change, the responder's CPI (mean ± sd) significantly decreased after a change to the prone position (0.45 ± 0.04 W/m2 vs. 0.34 ± 0.02 W/m2, p = 0.024). Cardiac index (CI, mean ± sd) decreased from 2.39 ± 0.15 l/min/m2 to 1.95 ± 0.16 l/min/m2 (p = 0.045). CPI was 0.50 ± 0.04 W/m2 in the non-responders and 0.34 ± 0.02 W/m2 in the responders (p = 0.002) in the prone position. After fluid loading, in the responder, CI increased from 1.95 ± 0.16l/min/m2 to 2.22 ± 0.16l/min/m2 (p = 0.025), and SVI increased from 27.62 ± 2.2 ml/m2 to 34.5 ± 2.06 ml/m2 (p = 0.001). CPI increased from 0.34 ± 0.02 W/m2 to 0.4 ± 0.03 W/m2 (p = 0.045) in the responders. However, CI decreased from 2.5 ± 0.18l/min/m2 to 2.29 ± 0.18 l/min/m2 (p = 0.009), and SVI (mean ± sd) decreased from 31.05 ± 1.78 ml/m2 to 28.55 ± 1.63 ml/m2 (p = 0.004) in the non-responders. CPI decreased from 0.5 ± 0.04W/m2 to 0.42 ± 0.03W/m2 (p = 0.015). The prone's CPI could predict fluid responsiveness under the receiver operating characteristic curve of 0.78 (95% CI, 0.60–0.95; p = 0.025). An optimal cut-off value of CPI in the prone was < 0.43 W/m2 (100% sensitivity and 65% specificity). Conclusion CPI may support and guide fluid management in the prone position. However, more data are required to confirm these findings.