Accuracy of transthoracic lung ultrasound for diagnosing pulmonary embolism: An updated systematic review and meta-analysis

Abstract

Background

Computed tomography pulmonary angiography (CTPA) simplifies the diagnosis of pulmonary embolism (PE) but is not suitable for all patients. Transthoracic lung ultrasound (LUS) is a potential alternative; this meta-analysis evaluates its accuracy for diagnosing PE.

Methods

We systematically searched PubMed, Embase and Cochrane Library from the inception of each database up to April 2024 according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy Studies guidelines. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool, and a bivariate random effects model was used to pool sensitivity and specificity.

Results

A total of 18 studies with 2158 patients were analyzed. Lung ultrasound showed a sensitivity of 0.80 (95 %, confidence interval (CI): 0.71–0.86; I² = 85.2 %) and specificity of 0.87 (95 %, CI: 0.81–0.92; I² = 87.3 %). The diagnostic score was 3.27 (95 %, CI: 2.75–3.78; I² = 61.9 %), and the diagnostic odds ratio was 26 (95 %, CI: 16–44; I² = 100.0 %). The pooled positive likelihood ratio was 6.2 (95 %, CI: 4.2–9.1; I² = 79.2 %), and the negative likelihood ratio was 0.24 (95 %, CI: 0.16–0.34; I² = 83.7 %). The summary area under the curve was 0.91 (95 %, CI: 0.88–0.93). Significant heterogeneity was observed, which may impact the generalisability of the results, and no publication bias was detected.

Conclusion

Transthoracic LUS shows potential as an alternative to CTPA for PE diagnosis, but further research is needed to improve its accuracy and establish standardised diagnostic criteria. The observed heterogeneity highlights the need for a cautious interpretation of the results.