Superior vena cava collapsibility index as a predictor of fluid responsiveness: a systematic review with meta-analysis

Abstract

Background: The superior vena cava collapsibility index (SVC-CI) is a potential marker of fluid responsiveness (FR) in mechanically ventilated patients. Few studies reporting its diagnostic performance are currently available. Methods: A systematic search, using the PRISMA approach, was performed using the Medline and EMBASE databases. Prospective studies evaluating the SVC-CI as a marker of FR in ventilated adult patients were included. A bivariate random-effect model was utilised to generate the summary receiver operating characteristic (SROC) curve. The area under the ROC curve (AUC), the sensitivity and specificity of the curve operating point were calculated. Results: We included eight studies with a total of 857 patients, in whom SVC-CI was evaluated a total of 1083 times prior to the volume expansion trial. In 609 (56.23%) trial cases FR was present. The SROC curve demonstrated that the test's operating point has a sensitivity and specificity of 80.8% (95% CI: 66.3-90%) and 81.4% (95% CI: 76.4-85.5%), respectively. The model's AUC was equal to 0.848 (95% CI: 0.824-0.863) with P < 0.001. No significant inter-study heterogeneity was found (I2 = 0%). A subgroup analysis revealed a significantly lower sensitivity of SVC-CI in patients with higher levels of positive end-expiratory pressure (PEEP) (> 5 cm H2O) (χ2= 7.753, df = 2, P = 0.0207). The study setting and type of intervention for volume expansion did not significantly change the performance of the test. Conclusions: SVC-CI is a reliable predictor of FR for mechanically ventilated patients in intensive care units and operating rooms. A PEEP level exceeding 5 cm H2O may impair the sensitivity of the test.