Meta-Analysis of Ventilated versus Spontaneously Breathing Patients in Predicting Fluid Responsiveness by Inferior Vena Cava Variation

Purpose: Respiratory variation in inferior vena cava (ΔIVC) has been extensively studied in predicting fluid responsiveness, but the results are conflicting. We performed a systemic review and meta-analysis of studies aiming at investigating the diagnostic accuracy of ΔIVC in predicting fluid responsiveness. Methods: MEDLINE, EMBASE, Cochrane Database and Web of Science were screened for relevant original and review articles from inception to July 2016. The meta-analysis determined the pooled sensitivity, specificity, diagnostic odds ratio (DOR) and area under the ROC curve (AUROC). In addition, subgroup analyses were performed in mechanically ventilated patients and spontaneously breathing patients. Results: A total of 20 studies involving 635 patients were included. Cutoff values of ΔIVC varied from 12% to 42%, the pooled sensitivity and specificity was 0.68 (0.62 - 0.75) and 0.80 (0.75 - 0.85), respectively. The DOR was 14.2 (6.0 - 33.6) and the AUROC was 0.86 (0.78 - 0.93). Subgroup analysis showed better diagnostic performance in patients on mechanical ventilation than in spontaneously breathing patients with higher sensitivity (0.75 vs. 0.56), specificity (0.82 vs. 0.78), DOR (22.9 vs. 7.9) and AUROC (0.90 vs. 0.80). The best threshold of ΔIVC in patients on mechanical ventilation was IVC distensibility index (ΔIVC ≥ 17% ± 4%), compared to IVC collapsibility index (ΔcIVC ≥ 33% ± 12%) in spontaneously breathing patients. Conclusion: ΔIVC is not an accurate predictor of fluid responsiveness in patients with acute circulatory failure. In patients on mechanical ventilation, the predicting ability of ΔIVC was moderate with acceptable sensitivity and specificity; in spontaneously breathing patients, the specificity remains acceptable but its sensitivity is poor.