Abstract
Background
Prediction of fluid responsiveness (FR) in critically ill patients is challenging. Passive leg raising (PLR) has been proven to adequately predict FR. PLR consists of a thoracic (T-PLR) and a limb (L-PLR) movement. Since a complete PLR (C-PLR) is not always feasible, this study focused on investigating the predictive value of partial PLRs on FR.
Methods
A prospective, observational study was performed in 40 post-cardiac surgery patients. C-PLR was performed, followed by a T-PLR, L-PLR and fluid challenge (FC). Invasive cardiac index (CI) measurements were continuously recorded during all maneuvers. FR was defined as a CI-increase ≥ 15% after FC, thereby identifying responders (R) and non-responders (NR). The predictive value of the PLR-elements was assessed with receiver operating characteristic (ROC) curves. Changes over time were analyzed with generalized linear model (GLM) analyses. Intraclass correlation coefficient (ICC) was used to assess absolute agreement between PLR and FC.
Results
Forty patients were included (35 R / 5 NR). AUC was similar for all PLRs (C-PLR = 0.84; T-PLR = 0.86, L-PLR = 0.86). ICCs between FC versus the three PLRs were 0.81 (0.63–0.90), 0.78 (0.59–0.88), and 0.71 (0.46–0.85), respectively. Median CI-increase during C-PLR was 27.8% (21–48%) in responders vs. 10.7% (7.5–12.6%) in non-responders (p = 0.012). After FC, median CI-increase was 30.0% (22.2–42.9%) in responders vs. 7.4% (6.3–12.4%) in non-responders (p = 0.002).
Conclusion
Partial PLRs have similar predictive values compared to a C-PLR. This could improve the prediction of FR in specific patient categories where C-PLR is restricted.