|Home | About | Journals | Submit | Contact Us | Français|
The aim of the study was to assess the ability of respiratory variations in pulse oximetry plethysmographic waveform amplitude (ΔPOP) to predict the increase of cardiac output (CO) after fluid infusion (fluid responsiveness).
Two authors independently performed a search iterated until November 2009 in MEDLINE, Embase and Cochrane Database using keywords: 'fluid OR preload OR volume responsiveness', 'cardiovascular monitoring', 'fluid challenge', 'functional hemodynamic monitoring', 'dynamic indices OR indexes', 'delta POP', 'pleth index', 'plethysmographic waveform'. Predictive value of ΔPOP was estimated by: difference in mean value of ΔPOP between responders and nonresponders; correlation coefficient between pre-infusion ΔPOP and CO increase after fluids; and sensitivity, specificity and area under the ROC curve (AUC) for ΔPOP to predict a responder state. A meta-analysis was performed using Comprehensive Meta-Analysis v. 2.2. Pooled values of diagnostic odds ratio (DOR), sensitivity, specificity and summary ROC (SROC) curve were calculated using MetaDiSC v. 1.4. Heterogeneity was evaluated with Q and I2 tests.
From the initial dataset of 839 records we identified six articles [1-6] on 135 patients to whom 195 fluid boluses were administered. All patients were in sinus rhythm and adapted to a ventilator. An increase of cardiac index or stroke volume index >10 to 15% defined responders. Mean responder rate was 58.9%. Pooled difference in mean was 10% (95% CI 5.8 to 14.2), correlation coefficient 0.59 (0.41 to 0.73), AUC 0.85 (0.76 to 0.94), sensitivity 81.1% (72.5 to 87.9), specificity 78.6% (68.3 to 86.8), DOR 21.7 (6.3 to 74.4), area under SROC 0.88 (0.80 to 0.96). The best threshold value for identification of responders was 9.5 to 15%. Heterogeneity was significant for all evaluations.
ΔPOP is a non-invasive dynamic parameter able to predict fluid responsiveness with only a moderate level of accuracy. The main limitation of this analysis is heterogeneity between included studies.