Comparative accuracy of non-invasive imaging versus right heart catheterization for the diagnosis of pulmonary hypertension: A systematic review and meta-analysis.

Abstract Background Right heart catheterization (RHC) is the gold-standard in the diagnosis of pulmonary hypertension (PH) but at the cost of procedure-related complications. We sought to determine the comparative accuracy of RHC versus non-invasive imaging techniques such as computed tomography (CT), magnetic resonance imaging (MRI), and transthoracic echocardiography (TTE). Methods Pulmonary hypertension was defined as a mean pulmonary artery pressure (mPAP) of>20 mmHg. Multiple databases were queried for relevant articles. Raw data were pooled using a bivariate model to calculate the measures of diagnostic accuracy and to estimate Hierarchical Summary Receiver Operating Characteristic (HSROC) on Stata 13. Results A total of 51 studies with a total patient population of 3947 were selected. The pooled sensitivity and specificity of MRI for diagnosing PH was 0.92(95% confidence interval (CI) 0.88–0.96) and 0.86 (95% CI, 0.77–0.95), respectively. The net sensitivities for CT scan and TTE were 0.79 (95% CI 0.72–0.89) and 0.85 (95% CI 0.83–0.91), respectively. The overall specificity was 0.82 (0.76–0.92) for the CT scan and 0.71 (95% CI 0.61–0.84) for TTE. The diagnostic odds ratio (DOR) for MRI was 124 (95% CI 36–433) compared to 30 (95% CI 11–78) and 24 (95% 11–38) for CT scan and TTE, respectively. Chi-squared (x2) test showed moderate heterogeneity on the test for equality of sensitivities and specificities. Conclusions MRI has the highest sensitivity and specificity compared to CT and TTE. MRI can potentially serve as a surrogate technique to RHC for the diagnosis of PH.