Background
Acute respiratory distress syndrome (ARDS) is a clinical syndrome, which is a type of acute diffuse, inflammatory lung injury, leading to increased pulmonary vascular permeability, increased lung weight, and loss of aerated lung tissue [
1]. ARDS has a high incidence rate and caries mortality of nearly 50% for patients with severe ARDS [
2]. At present, there are limited methods of improving the accuracy of ARDS diagnosis in patients [
3‐
5]. Elderly patients generally have a higher number of comorbidities and poorer homeostatic capability. The diagnosis of ARDS in this patient population is even more challenging and needs improvement [
2]. Therefore, early diagnosis and intervention are very important to improve the prognosis for elderly patients with ARDS. The diagnosis of ARDS at our institution is largely based on change noted on lung ultrasonography (LUS). The data collection of thoracic X-ray, especially bedside X-ray examination, can be affected by various factors. Hence we found it to not be valuable in the diagnosis of ARDS [
6,
7]. Even with advantages over chest X-ray, thoracic CT has limited usage among critically ill patients, due to its high risk in patient transport, high cost, and risk of radiation exposure. Therefore, there are various underlying difficulties in early diagnosis of ARDS, especially imaging diagnosis.
LUS has drawn increasing attention among clinical physicians recently [
8‐
12]. The guideline of international evidence-based recommendations for point-of-care lung ultrasound [
13] aimed to improve and standardize the clinical application and scientific research of lung ultrasound. However, there has yet to be a good comparison between point-of-care ultrasound (PoCUS) and the imaging gold standard of CT diagnosis of ARDS. Persistent questions include: what ultrasound signs are best for the diagnosis diagnose ARDS? What are the sensitivity and specificity of PoCUS in diagnosing ARDS? What is the effect of the underlying diseases of the patients in diagnosis, especially those in lungs? In order to answer these questions, we performed a prospective observational study using dynamic ultrasound monitoring and other clinical indicators on 51 elderly patients with suspected ARDS, in order to investigate the diagnostic value of cardiopulmonary ultrasound in this patient group.
Discussions
Chinese society is aging [
19],and resulting in an increasing number of critically ill elderly patients. The mortality and morbidity of those patients may be much higher than younger patients. This may be attributed to a lower reserve capacity in most important organs and systems functions, which will reduce ability to deal with physical stress and the presence of acute or chronic comorbidities. Therefore, early diagnosis maybe have a marked impact on interventions and outcomes of elderly patients with ARDS. However, the Berlin diagnostic criteria of ARDS widely used in clinical practice, is not very clear about the evaluation standard of pulmonary imaging, especially thoracic X-ray, which may lead to poor reliability of ARDS diagnosis [
8]. The development of clinical application and research of PoCUS provides a novel way of ARDS diagnosis in imagine. Cardiopulmonary ultrasound can help evaluate the cardiopulmonary morphology and function, but more exploration is needed to investigate the relation between cardiopulmonary ultrasound and chest CT scan in chest imagines for diagnosis of ARDS. If ultrasound proved sensitive and specific in early ARDS diagnosis, it may become part of a novel diagnostic imaging standard of ARDS diagnosis.
This study was a single-center, prospective observational study. Patients were divided into ARDS group (
n = 33) and control group (
n = 18), based on the results of chest CT scan on day 3. There were no significant differences between the two groups in gender, age, underlying disease, APACHEII score, SOFA score, PaO
2/FiO
2 ratio and PEEP levels day 1. It’s reported that NT-proBNP is an important biomarker of heart failure [
20]. Previous studies suggested that patients with diagnosis of ARDS often had right ventricular dysfunction [
21,
22], which would damage cardiopulmonary function. This study showed that, the levels of NT-proBNP on day 1 were increased in both groups, while the levels in control group were statistically significantly higher than those in the ARDS group (
p = 0.046). Respiratory failure of patients without ARDS was mainly caused by cardiogenic factors, whereas for patients with ARDS, other factors, such as hypoxia, can also affect the cardiac function. There are obvious limitations to using NT-proBNP alone to differentiate ARDS from acute or chronic heart failure.
Our study showed that the results of LUS examinations were comparable with those of ARDS diagnosis using CT scan. Consistent with the results of the preliminary animal experiment in the research group, the Kappa value was 0.82 on day 3,indicating early ARDS diagnosis using LUS was of similar clinical value as a CT scan. In our study, the results of CT scan in 33 patients with ARDS on day 3 showed different degrees of pulmonary consolidation, especially obvious in the gravity-dependent areas, such as lateral chest and back. This is a characteristic finding of ARDS on chest CT scan. The relation between ultrasound signs and ARDS diagnostic criteria through chest CT scan was analyzed according to the ARDS diagnosis standard using ultrasound suggested in this study. The sensitivity, specificity and AUROC of LUS on day 1, day 2 and day 3 were high and increased between day 1 and day 3. Our results indicated that diagnostic imaging of ARDS could be partially based on LUS signs. The results on day 3 were of higher diagnostic value, since the third day may be the peak of the inflammation.
Pulmonary ultrasound yields considerable advantage in the diagnosis and management of various pleural cavity and pulmonary diseases [
23‐
26]. Change in pulmonary ventilation area can be determined by LUS before the reduction of PaO
2/FiO
2 [
27]. Our previous animal experiment showed that LUS could semi-quantify pulmonary edema [
28,
29]. Inevitably, the diagnosis of ARDS must be differentiated from acute left heart failure, and cardiac ultrasound is an important means of comprehensive evaluation of cardiac function. Previous studies showed that cardiopulmonary ultrasound had obvious advantages in investigating the etiology of acute respiratory failure [
30,
31]. Therefore, we believe that a cardiopulmonary ultrasound approach will have broad application prospects in ARDS diagnosis. The sensitivity, specificity and AUROC of ARDS diagnosis using cardiopulmonary ultrasound on day 1, day 2 and day 3 were 0.879,0.889,0.924;0.939,0.889,0.961;and 0.970,0.833,0.956, respectively.
Our study’s main aim was to investigate the diagnostic value of cardiopulmonary ultrasound in elderly patients with ARDS. According to the guideline of point of care LUS [
13], there are no consistent standards in diagnosing ARDS using LUS findings. Combined with Berlin criteria of ARDS [
1] and our clinical experience, pulmonary edema and consolidation are the main imaging findings in ARDS patients. If there is normal LVEF or even hyperdynamic state in patients, combined with the results of LUS,we basically believe that the patient’s respiratory distress is caused by pulmonary edema caused by ARDS. Our results showed that cardiopulmonary ultrasound was of greater advantage compared with LUS alone in ARDS diagnosis. The reason may be that cardiopulmonary ultrasound can exclude the interference of heart failure in ARDS diagnosis. Both NT-proBNP and PaO
2/FiO
2 are important clinical indicators in pathophysiology. We also evaluated the combination of cardiopulmonary ultrasound and NT-proBNP or PaO
2/FiO
2 in ARDS diagnosis, with sensitivity, specificity and AUROC on day 2:0.938,0.887,0.964;0.939,0.889,0.965, respectively. These results suggest a combination of ultrasound signs and pathophysiology indicators was of greater application value in ARDS diagnosis.
This study investigated the ARDS diagnosis in elderly patients using PoCUS. Lung ultrasound, especially cardiopulmonary ultrasound was of important clinical application value in ARDS diagnosis in elderly patients. A combination of ultrasound signs and pathophysiology indicators was of more application value than ultrasound signs alone. PoCUS is mainly applied not only in critical care, emergency medicine, and trauma surgery, but also in pulmonary and internal medicine, especially in the assessment of cardiopulmonary function. Respiratory failure of patients without ARDS was mainly caused by cardiogenic factors, whereas for patients with ARDS may be promoted in the diagnosis and treatment management in elderly patients with ARDS.
This study had a few of limitations including its observational design and finite study arms. Initial indications are promising for cardiopulmonary ultrasound in the diagnosis of ARDS. However, the effect of underlying disease of elderly patients on its diagnostic value needs to be investigated in larger prospective and also interventional studies to evaluate any effect on outcomes.