LUS-ARDS Score May Accurately Diagnose Acute Respiratory Distress Syndrome

Using the LUS-ARDS score to diagnose patients with ARDS had comparable accuracy with diagnoses performed by expert clinicians using chest x-rays.

The lung ultrasound for acute respiratory distress syndrome (LUS-ARDS) score provides good diagnostic accuracy for ARDS, according to a study in the American Journal of Respiratory and Critical Care Medicine.

Investigators sought to develop and validate a data-driven LUS score for diagnosing ARDS. Accuracy of the LUS-based diagnoses were evaluated through comparison with the diagnoses of expert clinicians.

The prospective, observational study was performed at 2 academic hospitals in the Netherlands. Participants, recruited from March 2019 to February 2021, were patients in the intensive care unit (ICU) who had begun mechanical ventilation within 48 hours of study enrollment. Investigators conducted LUS examinations and obtained data on patient characteristics, ventilation, and gas-exchange parameters, both upon study enrollment and 24 hours later.

All participants were independently assessed for meeting the Berlin criteria for ARDS by 3 expert clinicians with extensive experience in diagnosing and treating ARDS. The participants were classified into the following categories: (1) certain no ARDS, (2) certain ARDS, and (3) uncertain ARDS. Patients with uncertain ARDS were subsequently discussed in a consensus meeting, where they were reclassified as either “likely ARDS” or “likely no ARDS.” Notably, the expert panel did not evaluate patients’ LUS exams.

The logistic regression model for diagnosing ARDS via LUS ultimately included 3 variables: the left LUS aeration score (range, 0-18); the right LUS aeration score (range, 0-18); and the number of antero-lateral lung regions with an abnormal pleural line (range, 0-8). After weighting these variables, investigators came up with an LUS scoring system with a range of 0 to 91.

The use of the LUS-ARDS score as a screening and diagnostic tool may improve the currently high number of underdiagnoses of ARDS in clinical practice and increase the use of appropriate treatment in these patients.

A total of 453 patients were analyzed. Of this group, 324 patients were in the derivation cohort (certain no ARDS, n=175; likely no ARDS, n=49; likely ARDS, n=47; and certain ARDS, n=53); and 129 patients were in the validation cohort (certain no ARDS, n=46; likely no ARDS, n=25; likely ARDS, n=14; and certain ARDS, n=44).

In the derivation cohort, the LUS-ARDS score had high discriminative performance for diagnosing ARDS in patients who were not categorized as having uncertain ARDS diagnosis, with an area under the receiver operating characteristics curve (AUROCC) of 0.90 (95% CI, 0.85-0.95). The discriminative ability of the LUS-ARDS score decreased to an AUROCC of 0.83 (95% CI, 0.77-0.88) when applied to all patients in the derivation cohort, including those with an uncertain ARDS diagnosis.

In the validation cohort, the LUS-ARDS score had good discrimination performance in diagnosing ARDS in participants who did not have an uncertain ARDS diagnosis (AUROCC: 0.85; 95% CI, 0.77- 0.93) as well as in all participants (AUROCC: 0.80; 95% CI, 0.72-0.87).

In the derivation and validation cohorts, 229 patients had an available computed tomography (CT) scan, which is considered the gold standard imaging modality. Among these patients, the AUROCC of the LUS-ARDS score was 0.84 (95% CI, 0.79-0.89). Of 65 patients from the validation cohort with a CT, the AUROCC was 0.82 (95% CI, 0.72-0.93).

In those patients categorized with uncertain ARDS by expert clinicians and then reclassified in a panel consensus meeting, a greater LUS-ARDS score was associated with an increased diagnosis of ARDS in the panel consensus meeting (P <.001). In this group of patients, an LUS-ARDS score of less than 8 led to a consensus diagnosis of ARDS in 10 of 41 (24%) patients, and an LUS-ARDS score greater than 27 was associated with a consensus diagnosis for 18 of 20 (90%) patients.

Study limitations include: lack of inclusion of some previously published LUS signs for ARDS in the scoring model; failure to obtain an index LUS for approximately 10% of participants; a low proportion of with nonpulmonary ARDS; and potential inter-operator variability in the performance of LUS exams.

Study authors concluded that “The use of the LUS-ARDS score as a screening and diagnostic tool may improve the currently high number of underdiagnoses of ARDS in clinical practice and increase the use of appropriate treatment in these patients.”


Smit MR, Hagens LA, Heijnen NFL, et al. Lung ultrasound prediction model for acute respiratory distress syndrome: a multicenter prospective observational study. Am J Respir Crit Care Med. Published online February 15, 2023. doi:10.1164/rccm.202210-1882OC