Utility of Lung Ultrasound for Evaluating Acute Chest Syndrome in Pediatric Patients with Sickle Cell Disease

Background: Acute chest syndrome (ACS) is a common complication in patients with sickle cell disease (SCD) and is a leading cause of morbidity and mortality. Physical assessment alone is not sensitive for diagnosing ACS and therefore Chest x-ray (CXR) is recommended because of the difficulty of diagnosing ACS on clinical grounds alone. Children with SCD are repeatedly exposed to diagnostic radiation for the evaluation of ACS. Focused chest ultrasound (US) has been used to evaluate for lung consolidation. If lung US can identify patients with ACS this application could potentially limit radiation exposure in patients with SCD at risk for ACS. We evaluated the utility of physician performed US as compared to CXR to identify patients with SCD who have ACS.

Methods: This is a prospective observational study that took place from November 2014-July 2015 in 2 urban pediatric emergency departments (EDs). The study population consisted of a convenience sample of patients with SCD from birth to 18 years of age at risk for ACS and who received a CXR for suspected ACS. Medical students and clinicians with training in lung sonography consented patients and performed a focused study to evaluate for lung consolidation. A blinded expert in point-of-care US reviewed for quality assurance and agreement. Sensitivity, specificity, and likelihood ratios were calculated for test performance characteristics of ultrasound using CXR as a reference standard. Inter-observer agreement (κ) between enrolling sonologists and reviewer was also calculated.

Results: 85 patients were enrolled for a total of 98 cases. Median age was 7 years (IQR 2-13 years) and 53% of patients were male. The prevalence of ACS by CXR was 14%. Lung US was able to detect consolidation with a sensitivity of 86% (95% CI, 56-97%), specificity of 95% (95% CI, 87%-98%), positive likelihood ratio (LR) 18 (95% CI, 7-48) and negative LR 0.2 (95% CI, 0.04-0.5). The agreement between enrolling novice sonologists' interpretation and blinded reviewer's interpretation was very good with a Cohen κ of 0.86 (95% CI, 0.7-1).

Conclusions: Focused lung US was able to identify ACS with high specificity. There was very good agreement between novice and expert sonologist interpretation. Lung US may decrease the need for CXR in patients at risk for ACS. Further studies are needed to see how this test performs within current clinical practice guidelines.