BACKGROUNDHypotension due to the induction of sedation with dexmedetomidine infusion may be harmful in critically ill patients. Changes in pulse pressure induced by the passive leg raising test (PLR-ΔPP) as marker of fluid responsiveness, assessed prior to sedation, may predict hemodynamic changes. The present study was to investigate the power of the PLR test in critically ill patients in predicting hypotension induced by the induction of dexmedetomidine sedation.

METHODSFluid responsiveness was estimated by a passive leg raising (PLR) test before dexmedetomidine sedation. Patients were assigned to either the "Nonresponders" or "Responders" group according to their hemodynamic responses to the PLR test ("Nonresponders", PLR-ΔPP < 10.3%; "Responders", PLR-ΔPP ≥ 10.3%). Sedation was performed with a dexmedetomidine infusion (0.5 µg/kg over a 10-minute loading period, then 0.2 - 0.7 µg×kg(-1)×h(-1)) and titrated to maintain the target Richmond agitation sedation scale (RASS) score in the range of -2 to -1 and the bispectral index value in the range of 60 to 75. Radial artery pulse pressure, heart rate (HR), and central venous pressure (CVP) were measured at each phase of the study procedure. Hemodynamic fluctuations during the use of dexmedetomidine sedation were recorded and compared between the two groups.

RESULTSFifty patients had a median (25% - 75% interquartile range) of 71 (61 - 78) years old were studied. At baseline, 39 of the 50 patients were "Nonresponders" and 11 were "Responders". Following dexmedetomidine sedation, patients classified as "Responders" had a significantly greater systolic blood pressure decrease during the induction of dexmedetomidine sedation than the "Nonresponders" ((-26.3 ± 6.8)% vs. -11.8 ± 8.5)%, P < 0.001). In addition, the "Responders" group required significantly more fluid boluses (8 vs. 3; P < 0.001) and vasopressors (2 vs. 0; P < 0.05) than the "Nonresponders" group to restore blood pressure. Finally, PLR-ΔPP was positively correlated with changes in systolic blood pressure (PLR-ΔSBP) (r(2) = 0.576; P < 0.001) and significantly correlated with dexmedetomidine infusion-induced changes in SBP (r(2) = 0.202; P < 0.05). AUC for PLR-ΔPP was 0.84 (95%CI 0.71 - 0.93). PLR-ΔPP predicted hypotension with a sensitivity of 73% and a specificity of 92%.

CONCLUSIONSThe fluid responsiveness assessment pre-sedation was found to predict blood pressure fluctuation during the induction of dexmedetomidine sedation. The PLR test conducted prior to sedation may be a useful tool to identify patients with a high risk of hemodynamic events and may be used to indicate the need for prophylactic treatment.