Critical Care Explorations (Dec 2023)

Diagnostic Accuracy of Cardiac Point-of-Care Ultrasound in a Tertiary Medical Intensive Care Unit

  • Steven Fox, MD,
  • Mahmoud Alwakeel, MD,
  • Xiaofeng Wang, PhD,
  • Siddharth Dugar, MD,
  • Neal Chaisson, MD

DOI
https://doi.org/10.1097/CCE.0000000000001019
Journal volume & issue
Vol. 5, no. 12
p. e1019

Abstract

Read online

Objective:. Critical care echocardiography (CCE) is a useful tool for managing critically ill patients in intensive care. However, concerns exist regarding the accuracy of CCE examinations because of operator dependence. We sought to evaluate the accuracy of CCE examinations compared with cardiology-performed transthoracic echocardiogram (TTE). Design, Setting, and Subjects:. We retrospectively reviewed charts of patients in a medical ICU in a large academic medical center in the United States. We compared CCE examinations performed by a fellow and reviewed by a staff physician between May 5, 2020, and December 31, 2021, to TTE obtained within 24 hours of the CCE examination. Intervention, Measurements, and Main Results:. We developed a standardized process for documentation of all CCE examinations performed in the medical ICU. We assessed agreement (kappa statistic), sensitivity and specificity of CCE examination compared with TTE. Features included left ventricle (LV) systolic function, right ventricle (RV) size, RV systolic function, pericardial effusion, mitral insufficiency, tricuspid insufficiency, and aortic insufficiency. The study analyzed 504 pairs of CCE and TTE examinations. Kappa statistics for detecting LV and RV systolic dysfunction, pericardial effusion, and RV size ranged from 0.60 to 0.74. CCE showed high sensitivity and specificity for detecting LV and RV systolic dysfunction and pericardial effusion, with values ranging from 0.85 to 0.99. The kappa statistic for detecting RV dilation was 0.59, with a sensitivity of 0.71 and a specificity of 0.85. In contrast, CCE examinations were nondiagnostic for mitral, tricuspid, or aortic insufficiency in 60–70% of cases, whereas TTE examinations were nondiagnostic in 20–30% of cases. Kappa statistics for mitral, tricuspid, and aortic insufficiency ranged from 0.32 to 0.42. Conclusions:. CCE is a reliable tool for assessing LV and RV systolic function, pericardial effusion, and RV size. However, CCE may be limited in its ability to detect mitral, tricuspid, or aortic insufficiency.