Abstract

Objectives: A chest radiograph (CXR) is still the preferred diagnostic method when pneumonia is suspected, although the sensitivity is relatively low. The aim of this study was to compare the diagnostic sensitivity, specificity, and accuracy of ultrasonography (US) for the diagnosis of community-acquired pneumonia (CAP), compared with CXR. Materials and Methods: A principled search was conducted to identify original English articles using PubMed, EMBASE, Web of Science, Scopus, and the Cochrane library, with the end date of October 2020. A combination of keywords, such as "ultrasound" or "ultrasonography," "pneumonia," "sensitivity," and "specificity," was used. Methodologic quality was assessed using Quality Assessment of Diagnostic Accuracy Studies 2 criteria. Statistical analysis was completed on the resulting study data. Results: The search produced 16 eligible articles that reported on 2040 patients. The overall pooled sensitivity for US and CXR, to diagnose pneumonia, was 0.96 and 0.65, respectively. The overall pooled specificity for US and CXR was 0.85 and 0.81, respectively. The overall pooled positive likelihood ratio for US and CXR was 9.74 and 3.67, respectively. The negative likelihood ratio for US and CXR was 0.05 and 0.42, respectively. In addition, summary receiver operative characteristics areas under the curve were 0.98 for US and 0.77 for CXR. Conclusion: This review demonstrated that lung US is a useful technique for the diagnosis of pneumonia. This diagnostic method can be used by emergency physicians with high accuracy, sensitivity, and specificity. Among an elderly population, this diagnostic method may be a better choice than CXR. The rapid performance of lung US may facilitate a quick, cost-effective, and safe diagnosis of this potentially fatal disease.