Kelly McHugh, PGY-1
The Article: Jones et al. “Feasibility and Safety of Substituting Lung Ultrasonography for Chest Radiography When Diagnosing Pneumonia in Children.” CHEST 2016; 150(1):131-138
The Idea: CXR is the diagnostic imaging of choice for pediatric pneumonia (PNA). Compared to CXR, lung ultrasound (LUS) limits radiation and may be more accessible in low income settings, where PNA continues to cause a high burden of mortality in pediatrics. Many studies have shown that (LUS) has a high accuracy for diagnosing PNA in the pediatric population. However, no studies have looked at outcomes when LUS is substituted for CXR. This study assesses the safety and feasibility of substituting LUS for CXR.
The Study: This study was a randomized controlled trial conducted between August 2012-July 2013.
Inclusion Criteria: Patients age birth to 21 years old presenting to a pediatric ED with symptoms concerning PNA.
Exclusion Criteria: Patients with previously performed CXR or who were hemodynamically unstable.
191 children were randomized to either the investigational group or the control group. In the investigational group, all patients received LUS, and CXR only if the clinician felt it was necessary or the admitting provider, PCP, or parent requested a CXR. In the control group, patients underwent CXR followed by LUS. Sonographers were pediatric EM attendings and fellows who underwent 1 hour of training in LUS. They were further categorized as novice if they had <25 scans or experienced if they had > 25 scans. A LUS was considered positive for PNA if there was consolidation with air bronchograms. Small subpleural consolidations without air bronchograms and focal B-lines were considered viral in etiology.
Primary Endpoints: Percent reduction in CXRs performed in the investigational group.
Secondary Endpoints: Rates of missed PNA, antibiotic use, hospital admission, and unscheduled health care visits, as well as ED length of stay.
The Results: There were 103 patients randomized to the investigational arm and 83 to the control arm. There was a 38.8% reduction in CXRs performed in the investigational arm. For experienced and novice sonographers there was a 60% and 30% reduction in CXRs respectively. When CXRs were performed at the request of admitting provider, PCP, or guardian after what was considered a conclusive LUS, there was a 67% potential maximum reduction in CXRs. In respect to secondary outcomes, there were no differences in rates of PNA diagnosed in the two groups. There were zero cases of missed PNA in both arms. There were no statistically significant differences in rates of hospital admission, antibiotic use, or unscheduled health care visits. There was a significant decrease in ED length of stay in those who received LUS only in the investigational group compared to the control group.
The Takeaway: LUS appears to be a safe alternative to CXR in the diagnostic evaluation of PNA in pediatrics. When LUS is performed first, the rates of CXRs can be significantly decreased. This is important as LUS avoids radiation, may be more cost effective than CXR, and may be used in resource limited settings where CXR is not available. Importantly there were no cases of missed PNA. One can argue the significance of these findings, given the minimal risk of radiation associated with CXR. Although these findings may not change practice for this reason, I think LUS may be useful in the right clinical setting; LUS could be used by the provider with ample experience to rule out a PNA, as an adjunct to CXR when findings are equivocal, or even as a quick screening tool prior to CXR.