Abstract
Evaluating fluid responsiveness with dynamic parameters is recommended for fluid management. However, in hemodynamically stable patients who are breathing spontaneously, accurately measuring stroke volume variation via echocardiography and passive leg raising is challenging due to subtle SV changes. This study aimed to identify normal SV changes in healthy volunteers and evaluate the precision of hemodynamic parameters in screening mild hypovolemia in patients. This prospective, repeated-measures, cross-sectional study screened 269 subjects via echocardiography. Initially, 45 healthy volunteers underwent a fluid challenge test, the outcomes of which served as criteria to screen 215 ICU patients. Among these patients, 53 underwent additional fluid challenge testing. Hemodynamic parameters, including medians of maximum velocity time integrals (VTImaxs), peak velocity of VTImax (PV), internal jugular vein diameters (IJVD), and area (IJVA) were repeatedly measured first at a 60° upper body elevation (UBE), second in a supine position, third at UBE, fourth in a supine position, and lastly in a supine position after fluid loading. The hemodynamic responses to the position changes were compared between 83 fluid non-responders and 15 fluid responders. Fluid responsiveness was defined as fluid-induced medians' change of VTImaxs (fluid-induced median VTImax change) ≥ 10%. None of the healthy volunteers showed the mean value of repeatedly measured medians of VTImaxs ≥ 7%, following either UBE position (UBE-induced median VTImax change) or fluid loading (fluid-induced median VTImax change). UBE-induced median VTImax and PV changes were significantly correlated with fluid responsiveness (p < 0.001, AUC 0.959; p < 0.001, AUC 0.804). The significant correlations were demonstrated via multivariable analysis using binary logistic regression (p = 0.001, OR 90.1) and the correlation coefficient (R(2) = 0.793) using linear regression analysis. UBE-induced median VTImax changes (≥ 11.8% and 7.98%) predicted fluid-induced median VTImax changes ≥ 10% and 7% (AUC 0.959 and 0.939). The collapsibility and variation of IJVD and IJVA showed no significant correlation. An increase in the mean value of medians of repeatedly measured VTImaxs transitioning from an UBE to a supine position, effectively screened mild hypovolemia and demonstrated a significant correlation with fluid responsiveness in spontaneously breathing patients maintaining hemodynamic stability.