The objective of this study was to determine the accuracy of point-of-care lung ultrasound compared to the gold standard of chest X-ray in identifying an infiltrate suggestive of acute chest syndrome in patients with sickle cell disease.

Acute chest syndrome is the leading cause of mortality in patients with sickle cell disease, and therefore this is a question with significant clinical relevance. As a result, these patients experience multiple radiological exposures starting at very young ages. If ultrasound is comparable to the current gold standard of chest radiography, patients with sickle cell disease may experience significantly less radiation exposure over their lifetimes.

This prospective observational study was conducted at 2 freestanding urban pediatric centers between November 2015 and July 2017. The study population was a convenience sample based on the availability of ultrasonographers. Exclusion factors included hemodynamic instability, chest radiography at an outside institution, and unavailability of a study ultrasonographer. Inclusion factors included documented sickle cell disease, aged 0-21 years old, and chest radiography within the current admission. Symptoms concerning for acute chest syndrome included chest pain, fever, vomiting, and respiratory symptoms such as cough. Ultrasonography was performed by sonographers who were blinded to physical examination and chest radiographic findings, ideally before chest radiography was performed.

Study sonographers included one expert blinded reviewer with 6 years of experience and specialized training in point-of-care lung ultrasonography, as well as five novice ultrasound trainees. The 5 novice sonographers underwent a 1-hour lecture on lung ultrasonography, a 1-hour practical hands-on imaging session, and 5 point-of-care lung ultrasonographic examinations before enrolling patients in the study. Consolidated lung findings by ultrasound were characterized by a hypoechoic region with irregular borders containing hyperechoic air bronchograms, requiring measurement greater than 1 cm to be determined positive. This is because consolidations less than 1 cm are often not identified on chest radiography. The expert sonographer was blinded to any previous interpretation and reviewed all studies that were performed.

Secondary outcomes included patient or guardian satisfaction, tolerability of the examination, and inter-operator reliability between trainees and the expert blinded reviewer. Satisfaction was assessed through a 10-question survey, using a Likert scale of 1 to 5 (1 being the least satisfied and 5 being the most satisfied). Satisfaction questions included queries about the entire ED visit as well as specific questions concerning the experience with point-of-care lung ultrasonography, including measures such as “pain and comfort during the scan, efficiency of the scan, whether the patient and family would like a point-of-care lung ultrasonographic examination at subsequent visits, how the scan affected interactions with healthcare providers, and the total patient experience in the ED.” The gold standard was the pediatric radiologist’s interpretation of the chest radiography.

Based on the afore-mentioned inclusion and exclusion criteria, 220 patients were eligible for participation. Of these, 21 patients declined enrollment, ultrasonography was incomplete for 7 patients, and 1 patient withdrew in the middle of examination due to discomfort. Ultimately, 191 patients were included in the data analysis.

1. This is a prospective observational study, which is not randomized or double-blinded. This study design is intrinsically at risk for confounding factors and potential selection bias. Furthermore, observational studies are only able to support correlation, never causation.

2. Included only six study ultrasonographers

3. Only 191 patients were enrolled out of possible 220. The small study size can skew the results and subsequent conclusions by encapsulating only a limited group, which is not generalizable to the general population.

4. Included only two centers

5. Convenience sample based on availability of ultrasonographers

6. Included only pediatric patients, so not generalizable to adults

Enrolled patients had a median age of 8 years (interquartile range 3 to 13 years), with 41% being female. Chest radiography identified an infiltrate in 32 patients (17%). 31 of 32 patients (97%) who had positive chest radiographic results and 67 of 159 (42%) who had negative chest radiographic results were admitted, making up ninety-eight admitted patients in total (51% of the study population). Two required critical care.

Forty of the total 191 patients in the study (21%) were found to have lung consolidation greater than 1 cm. Novice sonographers were able to detect acute chest syndrome with an accuracy of 89%, while the expert sonographer had higher accuracy at 92%. Interobserver agreement between novice and reviewing sonographer was moderate (κ=0.67 [95% CI 0.54 to 0.80]). Notably, ultrasound had a negative predictive value of 92-100% among trainees and 93-99% for the expert. Positive predictive value was also found to be greater than that of clinical symptoms such as cough, abnormal respiratory rate, shortness of breath, and chest pain.

In terms of secondary outcomes, questionnaires were at least partially completed by a total of 141 families (73%). Ninety percent of individuals (124/138) responded that they had little or no pain; however, 1 patient withdrew from the study because of discomfort. 93% of patients reported being satisfied with their ED care, with only 4% of enrolled patients stating that they would not like an ultrasonographic examination during subsequent visits.

The results of this study suggest that ultrasonography may be useful as a screening tool with relatively good accuracy for the presence of consolidation in young patients with sickle cell disease and concern for acute chest syndrome, thereby potentially limiting radiation exposure. This is supported by the high negative predictive value shown by both trainees and an advanced clinician.

• POCUS lung ultrasonography has high negative predictive value for acute chest syndrome in young patients with sickle cell disease

• Screening with POCUS lung ultrasonography could potentially reduce the necessity for radiation exposure

• Lung ultrasonography can be learned by novice sonographers, suggesting that expanded education may have clinical utility

• Recruitment of patients based on convenience sample

• Imperfect gold standard (chest radiography itself has shown poor inter-operator reliability and lower sensitivity compared to ultrasound. CT chest would be more accurate, but has ethical issues with significantly more radiation exposure)

• No assessment of improved trainee sonography with more examinations

• Difficulty differentiating acute versus chronic lung changes by ultrasonography. In previous lung ultrasonography studies, patients with chronic cardiorespiratory changes have been excluded, as pulmonary scarring looks similar to consolidation under ultrasound. This is not a feasible exclusion criteria when studying patients with sickle cell disease, who by definition have chronic lung changes. This was demonstrated by four false positives within this study due to chronic pulmonary scarring being incorrectly identified as new consolidations.

• Potential loss of blinding (unable to blind from visible tachypnea or respiratory distress)

This study makes me more likely to perform a screening point-of-care lung ultrasound searching for a consolidation of greater than 1 cm prior to ordering a chest X ray in young patients concerning acute chest syndrome, a life threatening complication in sickle cell disease.