By Samuel Nadler, MD, PhD

Critical Care, Pulmonary Medicine, The Polyclinic Madison Center, Seattle; Clinical Instructor, University of Washington, Seattle

Dr. Nadler reports no financial relationships relevant to this field of study.

SYNOPSIS: Lung ultrasound may be useful in predicting which patients with acute respiratory distress syndrome would favorably respond to prone positioning.

SOURCE: Prat G, Guinard S, Bizien N, et al. Can lung ultrasonography predict prone positioning response in acute respiratory distress syndrome patients? J Crit Care 2016;32:36-41.

Despite increasing adoption of lung protective ventilation, patients with acute respiratory distress syndrome (ARDS) still experience high morbidity and mortality. Many strategies have been studied to improve the care of these patients. The PROSEVA study demonstrated that the early application of prone positioning decreased 28-day and 90-day mortality in patients with severe ARDS.1 Other strategies include neuromuscular blockade, inhaled pulmonary vasodilators, and alternative ventilation strategies. Which patients could benefit from each strategy may be difficult to determine.

This study tested whether lung ultrasound (L-US) could predict which patients would benefit from prone positioning. Prospectively, 19 patients admitted to a medical ICU with ARDS in whom the attending physician had decided to apply prone positioning were assessed with L-US in 12 different regions. Three ICU physicians with expertise in L-US evaluated each of the 12 segments and classified them as normal (N), with more than two B-lines (B1), with severe loss of aeration (B2), or lung consolidation (C). Researchers placed patients in the prone position and quantified the response. A positive response was defined by an improvement in the PaO2/FiO2 ratio of > 20 mmHg at two hours and 14 hours after prone positioning. All patients received lung protective ventilation according to a standardized protocol.

Of the 19 patients enrolled in the study, 12 and 13 met the criteria for positive response at two and 14 hours, respectively. In comparing patients who demonstrated favorable responses to prone positioning to those who did not, there were no significant differences in age, tidal volumes, plateau pressure, lung compliance, lung aeration scores, rates of immunocompromise, or ICU mortality. However, the presence of a normal L-US pattern in the bilateral anterobasal segments produced a positive predictive value of 100% (95% confidence interval [CI], 59-100%) for response to prone positioning and negative predictive value of 58% (95% CI, 28-85%) at two hours. Similar values were observed at 14 hours.


This study is intriguing in that it proposes a method through which a bedside assessment might predict which patients might benefit from prone positioning. Before adopting this protocol, consider several limitations. First, this was not a prospective study of L-US to determine which patients should undergo prone positioning. Rather, all patients were placed in the prone position, and there was no difference in the mortality of patients identified as responders vs. non-responders. A prospective trial that decided whether proning should occur based on L-US findings would be needed to determine if this protocol is truly beneficial. Second, 12 ultrasound zones were evaluated, but only the anterobasal segments correlated with beneficial outcomes. Without a correction for multiple comparisons, it may be simple chance that one of the 12 segments correlated with improvement. Third, the assessments were performed by physicians with expertise in L-US. Thus, the generalizability of this protocol remains a question.

Ultimately, the decision whether to place a patient in prone position relates to the risks vs. benefits of this intervention. To this end, the benefits of prone positioning for ARDS are becoming clear. PROSEVA demonstrated a clear mortality benefit of placing all patients with severe ARDS in the prone position early in the course of the disease.1 Conversely, the risks of prone positioning such as facial edema, skin breakdown, accidental extubation, and catheter displacement can be mitigated by experience with proning. In the Prat et al study, five of the 12 patients who did not experience a favorable L-US pattern in the anterobasal segments still responded favorably to prone positioning at two hours, and six of the 12 patients at 14 hours. Therefore, it becomes difficult to propose deferring prone positioning based on consolidation in the anterobasal segments of the lung as determined by lung ultrasound.

Overall, consider prone positioning in all patients with severe ARDS. Should other factors exist that increase the risk of prone positioning (e.g., trauma, abdominal incisions, obesity, difficult airway), L-US may be useful in determining which patients undergo this intervention. A prospective study to test L-US in determining which patients are placed in prone position would shed light on this question.


  1. Guerin C, Reigner J, Richard JC, et al. Prone positioning in severe acute respiratory distress syndrome. N Engl J Med 2013;368:2159-2168.