The main finding of this study is that in a large group of hospitalized and community ambulatory patients, TUS identifies a range of pathologic entities and changes management interventions. There are many studies validating application of the technique for community-acquired and ventilator-associated pneumonia, pneumothorax, and acute respiratory failure. However, in this study TUS was applied in diverse patient populations, including patients with underlying malignancy, pregnancy and trauma/surgical patients [21,22,23].

Lung ultrasound represents an emerging and useful imaging technique in the diagnosis and management of pulmonary and pleural diseases. Thanks to its simplicity and non-invasive nature, it can be useful as a bedside tool for establishing diagnosis, monitoring and guiding management strategies in various clinical setting [7]. It can definitely serve as a basic application in critically ill patients, as it demonstrates sensitivity and specificity ranging from 90% to 100% (using CT as the “gold standard”) in the assessment of different respiratory disorders encountered in the ICU [2, 20, 24]. We have previously shown that lung ultrasound has better diagnostic performance than CXR for common pathologic entities such as consolidation, interstitial syndrome, pneumothorax and pleural effusion [24]. In addition, it has a significant impact on clinical decision making in critically ill patients [25]. However, outside the critical care setting data are not so powerful. The limitations associated with lung ultrasound need to be mentioned, irrespective of the fact that they do not outweigh its benefits. First, LU is a highly operator-dependent imaging modality, which requires training. Subcutaneous emphysema and thoracic dressings, conditions common in the critical care setting, can prevent image acquisition and make the examination impossible from a technical point of view. Moreover, ultrasound can underestimate lesions that involve deep lung layers [1].

Although TUS can guide diagnosis in a variety of chest diseases in the emergency department [26] and affect therapeutic management [27], its use is not widespread among clinicians. This is despite the existing literature supporting its usefulness: studies have shown that lung ultrasound can be used in the diagnosis and follow up of community-acquired pneumonia [28,29,30], while TUS has been found to be a significant diagnostic tool in evaluating and managing patients with pleural effusions [31]. It has also been shown to be a valuable tool in the assessment and monitoring of lung congestion in cardiogenic pulmonary edema [32,33,34] or in guiding diuretic treatment in patients with heart failure [35]. Recently, TUS has been used in the evaluation of connective tissue disease-associated interstitial lung disease [36, 37]. The assessment of diaphragm thickness, thickening fraction, and excursion in both ambulatory and mechanically ventilated patients could be crucial to identifying diseases responsible for diaphragm dysfunction [38]. Serial diaphragmatic assessment by ultrasound may lead to earlier non-invasive ventilation initiation in patients with amyotrophic lateral sclerosis [39]. During the COVID-19 pandemic, lung ultrasound was recognized as an invaluable tool in the diagnosis and monitoring of the disease in the lung, greatly influencing the clinical-decision making process in affected patients [40, 41]. Obviously, the higher sensitivity of lung ultrasound compared to chest radiography is an advantage, though it can also prompt diagnoses of conditions that erroneously guide unnecessary treatment modalities. Integrated clinical and imaging skills are needed to achieve appropriate patient management in this setting [42].

In this study, the vast majority of TUS examinations were performed due to underlying malignancy. Primary or secondary lung cancer can present with pleural effusion, pulmonary edema, post-obstructive pneumonia or ascites, which can be assessed and managed successfully by the clinician with point-of-care ultrasound [43]. In a cross-sectional study of 53 patients with confirmed bronchogenic carcinoma, thoracic ultrasound had a significant complementary role to computed tomography in the diagnosis and staging of this type of cancer, especially if peripherally located [18]. In a recent meta-analysis, although TUS was not useful as a ruling-out test for malignant pleural effusion, the identification of pleural nodularity could motivate further investigation if there is a strong suspicion it is present [44]. In this study, a large number of patients with malignancy were evaluated and TUS was used to guide invasive procedures. Primary or secondary malignant disease was associated with significant bilateral asymmetrical pleural effusions. The main finding of this study was the paradoxical movement of the diaphragm due to large malignant effusion. Indeed, almost in 58% (168/293) of examinations, diaphragmatic inversion or paradoxical movement of the hemi-diaphragm was detected during inspiration, and full recovery was observed after pleural fluid evacuation. Although diaphragmatic paradoxical movement has already been reported [13, 45], this study is the first to describe this reversible phenomenon in a considerable number of malignant effusions.

Additionally, we performed 42 TUS studies in the context of pregnancy (pregnancy, post-partum, ovarian hyperstimulation syndrome), thus avoiding exposure to radiation in this vulnerable group of patients. TUS patterns have already been described in healthy parturients [46], in women during labor [47] as well as in pregnant women with preeclampsia [48]. Lung ultrasound has been proved to be a valuable tool in identifying excess lung water in severe preeclamptic patients [49] and predicting interstitial syndrome and hemodynamic profile in this group [50].

The present study is the first to describe the ultrasonographic findings and clinical course in a diverse group of ambulatory patients presenting to the lung ultrasound department for a variety of reasons, e.g., due to an underlying clinical condition, or an unexplained clinical picture or imaging tests. The strengths of this study are the large number of participants, the application of TUS as a single diagnostic method in most cases, and the use of TUS as a decision making and monitoring tool. Limitations include the retrospective design of the study, single-center conduction, the inevitable existence of (limited) missing data, and the fact TUS findings were not confirmed by computed tomography, which is regarded as the gold standard technique. However, a previous study from the same center has proved the high diagnostic performance of TUS and its applicability as alternative to CT.