Diagnostic accuracy of point-of-care ultrasound for shock: a systematic review and meta-analysis.

ABSTRACT

BACKGROUND: Circulatory failure is classified into four types of shock (obstructive, cardiogenic, distributive, and hypovolemic) that must be distinguished as each requires a different treatment. Point-of-care ultrasound (POCUS) is widely used in clinical practice for acute conditions, and several diagnostic protocols using POCUS for shock have been developed. This study aimed to evaluate the diagnostic accuracy of POCUS in identifying the etiology of shock. METHODS: We conducted a systematic literature search of MEDLINE, Cochrane Central Register of Controlled Trials, Embase, Web of Science, Clinicaltrial.gov, European Union Clinical Trials Register, WHO International Clinical Trials Registry Platform, and University Hospital Medical Information Network Clinical Trials Registry (UMIN-CTR) until June 15, 2022. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and assessed study quality using the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Meta-analysis was conducted to pool the diagnostic accuracy of POCUS for each type of shock. The study protocol was prospectively registered in UMIN-CTR (UMIN 000048025). RESULTS: Of the 1553 studies identified, 36 studies were full-text reviewed, and 12 studies with 1132 patients were included in the meta-analysis. Pooled sensitivity and specificity were 0.82 [95% confidence interval (CI) 0.68-0.91] and 0.98 [95% CI 0.92-0.99] for obstructive shock, 0.78 [95% CI 0.56-0.91] and 0.96 [95% CI 0.92-0.98] for cardiogenic shock, 0.90 [95% CI 0.84-0.94] and 0.92 [95% CI 0.88-0.95] for hypovolemic shock, and 0.79 [95% CI 0.71-0.85] and 0.96 [95% CI 0.91-0.98] for distributive shock, respectively. The area under the receiver operating characteristic curve for each type of shock was approximately 0.95. The positive likelihood ratios for each type of shock were all greater than 10, especially 40 [95% CI 11-105] for obstructive shock. The negative likelihood ratio for each type of shock was approximately 0.2. CONCLUSIONS: The identification of the etiology for each type of shock using POCUS was characterized by high sensitivity and positive likelihood ratios, especially for obstructive shock.