Ability of adjusted pulse pressure variation in predicting fluid responsiveness in spontaneously breathing patients with sepsis / 中华急诊医学杂志

ABSTRACT

Objective:To investigate the effect of tidal volume (Vt DI) on pulse pressure variation (ΔPP DI) during deep inspiration maneuvers in spontaneously breathing patients with sepsis and to test if adjusting ΔPP DI by Vt DI can further improve its ability in predicting fluid responsiveness (FR). Methods:Spontaneously breathing, nonintubated sepsis or septic shock patients who were admitted to the Intensive Care Unit of the Characteristic Medical Center of Chinese People's Armed Police Force and Nanjing Gaochun People's Hospital were prospectively enrolled from October 2017 to October 2019. Volume expansion (VE) was performed by infusing 500 mL saline over 20 min. Prior to VE, measurements including pulse pressure variation and tidal volume were obtained during quiet spontaneous breathing (ΔPP TB and Vt TB, respectively) and during the deep inspiration maneuver (ΔPP DI and Vt DI, respectively). Patients were classified as responders if stroke volume (SV) increased ≥ 15% after VE, otherwise non-responders. Multiple linear regression analysis was conducted to investigate the correlation of ΔPP DI with Vt DI and VE-induced percentage changes in SV (ΔSV). Receiver operating characteristic (ROC) curve analysis and the gray zone approach were used to assess the ability of each index to predict FR. Changes in gray zone limits according to the cost ratio (R = cost[false positive (FP)]/cost[false negative (FN)]) were also evaluated. Results:Of the included 31 patients, 17 were responders. There was no significant difference in ΔPP TB between fluid responders and non-responders ( P>0.05), whereas ΔPP DI was significantly higher in responders than in non-responders [(19.1±7.4)% vs (11.2±4.5)%; P=0.001]. The area under the ROC curve (AUC) of ΔPP DI predicted FR was 0.832, sensitivity of 76.47% and specificity of 71.43%, which was significantly higher than ΔPP TB (AUC=0.580, sensitivity of 47.06% and specificity of 71.43%; P<0.05). Multiple linear regression analysis showed that both Vt DI and ΔSV were independently associated with ΔPP DI ( P<0.01), the AUC of ΔPP DI adjusted by Vt DI was signigicantly higher than that of ΔPP DI alone ( P=0.03). Among the ΔPP TB, ΔPP DI and ΔPP DI/Vt DI, ΔPP DI/Vt DI had the narrowest gray zone (12.7-14.5) for the normal fluid policy (R=1), which only included 19% of the patients. When applying "restrictive" fluid management (R=2), the gray zone for ΔPP DI/Vt DI was 12.8-14.5 and included only 2 patients (6.5%). Conclusions:In spontaneously breathing, nonintubated patients with sepsis or septic shock, the ΔPP value obtained during the deep inspiration maneuver predicts FR with moderate accuracy. Given the close correlation between Vt DI and ΔPP DI, ΔPP DI adjusted by Vt DI performs better than ΔPP DI alone in predicting FR.