Pathophysiology Versus Etiology Using Lung Ultrasound: Clinical Correlation Required

Point-of-care lung ultrasound (POC-LUS) is a non-ionizing, bedside imaging modality capable of characterizing lung morphology and pathophysiology (1). Emergency and critical care practitioners increasingly use POC-LUS as a diagnostic tool for patients with acute respiratory failure (ARF) and to guide subsequent management (2). Lung ultrasound is different from other diagnostic ultrasound applications; interpretations rely upon both direct visualization of structures as well as patterns arising from the unique interface of the technology and the anatomy. Since air reflects ultrasound waves, the lung is a poor acoustic medium for imaging. Therefore, the interpretation of POC-LUS relies on the presence and absence of imaging artifacts from the pleural surface and direct visualization of consolidated or fluid filled lung parenchyma (1). Despite these limitations, adult POC-LUS protocols are highly sensitive and specific for differentiating between common etiologies of ARF (3). Additionally, clinicians use lung ultrasound scoring systems to characterize severity of acute respiratory distress syndrome (ARDS) and assess lung aeration over time (4). In pediatric intensive care units, widespread uptake of POC-LUS for clinical diagnostics and management guidance lags behind our adult counterparts.

In this edition of Pediatric Critical Care Medicine, DeSanti et al (5) prospectively evaluated the diagnostic sensitivity and specificity of POC-LUS for common etiologies of pediatric ARF. Experienced critical care practitioners obtained protocolized POC-LUS images in the anterior, mid-axillary, and posterior thorax bilaterally. In critically ill patients, the presence of mechanical ventilation, sedation or neuromuscular blockade may limit the assessment of posterior lung fields. Of the 87 patients in this study, 86 were able to have complete evaluations, with only one patient not having posterior thorax evaluation. Only 5 patients were mechanically ventilated, thus this study and its methodology may be biased towards a “less sick” population, though the study also provides encouragement that clinicians can perform a complete POC-LUS protocol in many of our children within the intensive care unit.

Clinicians blinded to a patient’s history of present illness, past medical history, and current clinical condition reviewed ultrasound images to determine the ultrasound-based etiology of ARF. The study team compared this ultrasound diagnosis with a clinical diagnosis of pneumonia, status asthmaticus, or bronchiolitis/ viral pneumonitis retrospectively determined by an independent clinician review. POC-LUS agreed with clinical diagnosis of ARF in 56% of children (49/87) and demonstrated moderate sensitivity and specificity for diagnosing bronchiolitis (sensitivity 44%, specificity 74%), pneumonia (sensitivity 76%, specificity 67%), and status asthmaticus (sensitivity 60%, specificity 88%) (5).

A key difference from previous POC-LUS diagnostic accuracy studies and a strength of this study was the use of a clinician blinded to the patient’s clinical history to determine the ultrasound diagnosis. Previous adult and pediatric studies that have shown sensitivity and specificity for diagnosis of ARF etiology in excess of 80% (3, 6) utilized clinical history in addition to ultrasound imaging. This approach mimics what occurs in clinical practice, but does not test the ability of POC-LUS to autonomously discriminate etiologic diagnosis. The discrepancy between the present study’s findings and previous works is striking. Should we conclude that POC-LUS is not actually useful in the work-up of pediatric ARF?

Both ultrasonographic findings and etiologic clinical diagnosis of pediatric ARF frequently overlap within and between themselves. Bronchiolitis, pneumonia and ARDS can all demonstrate sub-pleural consolidations. Pleural effusions can be seen in pneumonia and ARDS. Pneumonia and bronchiolitis can both result in ARDS. With so much overlap, how can we expect ultrasound alone to sort out these complex diagnostic challenges? The highly sensitive and now frequently employed adult BLUE protocol (3) differentiates etiologies with much more heterogenous pathophysiology compared to the pediatric etiologies assessed by DeSanti et al. This does not mean pediatric POC-LUS has no diagnostic utility. The recently published European Society of Pediatric and Neonatal Intensive Care (ESPNIC) point of care ultrasound guidelines (7) support the use of POC-LUS to assist in detection of pneumonia with higher diagnostic accuracy than chest radiograph (8). In children with hemodynamic instability, pneumothorax can quickly be ruled out by the presence of lung sliding (9). POC-LUS can differentiate between simple and complex collections and guide antibiotic selection and need for pleural fluid drainage in pediatric patients with bacterial pneumonia and effusion (10).

This study also utilized experienced practitioners to perform and interpret POC-LUS imaging, thereby limiting the generalizability of these findings. Skill in image acquisition and interpretation across any diagnostic ultrasound application requires both defining the scope of the application as well as sufficient training and oversight. Expert derived competence standards in POC-LUS have been developed (11) and studies performed by trainees within supervised training programs highly correlate with expert practitioners in as few as 25 ultrasounds (12).

Instead of asking lung ultrasound to tell us etiologically “what is this,” should we instead be asking physiologically “what is going on?” The ease and safety of initial and repeat imaging in POC-LUS allows for assessment of lung morphologic characterization, pathophysiologic assessment, and longitudinal response to ongoing care. Lung aeration can be quantified by the previously mentioned lung ultrasound scoring system (4). Increasing regional and global scores correspond to loss of lung aeration and these findings have been correlated to density of lung tissue on chest CT (13). How can an intensivist incorporate POC-LUS in clinical care? Clinicians can use ultrasound to evaluate lung aeration over time by identifying changes in areas of lung consolidation in response to protocolized titration of mechanical support and performance of recruitment maneuvers (14). It is possible through the use of POC-LUS that we can monitor lung aeration response to changes in ventilator positive end-expiratory pressure, the application or discontinuation of non-invasive positive pressure ventilation, and the response to prone ventilation. The change from an A-line to confluent B-lines pattern during fluid resuscitation for a child in shock heralds the development of pulmonary edema (15) and may be used to help guide fluid resuscitation for children in shock.

Just like the chest radiograph before it, POC-LUS is not the holy grail to diagnose pneumonia, and clinical correlation is still required. In contrast to adult POC-LUS imaging algorithms, POC-LUS images alone have only moderate sensitivity and specificity for etiologic determination of ARF in pediatrics. We should not take this as a discouraging finding; deciphering the underlying cause of ARF is not always easy, and even gold standards of diagnosis are controversial (16). Instead, we should take this as an opportunity to rethink how to best use POC-LUS in pediatric critical care. Framing POC-LUS as a window to understanding physiology, and not necessarily etiology, impacts how we might translate findings to clinical care. Lung ultrasound, as with other diagnostic ultrasound applications in pediatric critical care medicine, will no doubt grow in use. Understanding its role in assessing children with respiratory distress, changes in oxygenation or ventilation, or other cardiopulmonary perturbations undoubtedly requires further study. But knowing the questions we should ask, and aligning them with what ultrasound can answer, is an important first step.