Echocardiographic characteristics of pulmonary artery involvement in Takayasu arteritis

Background Up to 50% patients with Takayasu arteritis have pulmonary artery involvement. Hence, the early identification of pulmonary artery involvement to facilitate prompt treatment is required. Methods This retrospective study was performed in patients diagnosed with Takayasu arteritis between January 2009 and January 2016. Pulmonary artery involvement was confirmed with computed tomographic pulmonary angiography. Images from transthoracic echocardiography in three windows (suprasternal right pulmonary artery long‐axis view, parasternal aortic short‐axis view, and subxiphoid view) were documented and analyzed. Results A total of 27 patients had Takayasu arteritis and pulmonary artery involvement. Characteristic changes identified by echocardiography included luminal medium‐to‐high echogenic signals, stenosis, and occlusion, as well as intimal thickening. Left pulmonary artery involvement was revealed in the parasternal aortic short‐axis view. Right pulmonary artery involvement was best observed in the suprasternal right pulmonary artery long‐axis view, with complementary views from the parasternal aortic short‐axis and subxiphoid angles. Pulmonary trunk involvement was not observed in all three windows. Conclusions Transthoracic echocardiography could be a useful noninvasive test to detect pulmonary artery involvement in patients with Takayasu arteritis.

studies have shown that transthoracic ultrasound could provide detailed vascular resolutions and mural involvements, which was useful in the evaluation of Takayasu arteritis. [10][11][12][13] However, these studies only reported the evaluations for carotid, brachiocephalic, and subclavian vascular changes. Furthermore, no previous study has investigated the role of ultrasound in the diagnosis of pulmonary artery involvement in Takayasu arteritis.
In this study, we investigated the features of transthoracic echocardiography on pulmonary artery involvement in patients with Takayasu arteritis. We have reported our results and discussed the clinical applications in this study.

| Study design and patient selection
This study is a retrospective study performed in an academic teaching hospital. Study protocol was reviewed and approved by the hospital ethics committee. All patients were treated at our hospital from January 2009 to January 2016.
Patient selection criteria were as follows: (1) diagnosis of Takayasu arteritis based on the American College of Rheumatology Criteria 14 and (2) diagnosis of pulmonary artery involvement based on computed tomographic pulmonary angiography (CTPA). Exclusion criteria were as follows: (1) pulmonary involvement only limited to lobules, subsegmental artery, but not the right and left pulmonary artery and (2) incomplete echocardiography records on pulmonary arteries.

| Echocardiography
Transthoracic echocardiography was performed routinely as clinical diagnostic tests for patients with Takayasu arteritis by sonographers in our hospital. All sonographers received standard trainings and had the certificate to perform echocardiography. The equipment used was Philips iE33 ultrasound with a S5-1 transducer (Philips, Amsterdam, The Netherlands) and GE Vivid I ultrasound with a 1.7-3.4 MHz Transducer (General Electric, Fairfield, CT, USA). The main pulmonary artery and its main branches were evaluated in three windows: parasternal aortic short-axis view, subxiphoid view, and suprasternal right pulmonary artery long-axis view. 15 Artery inner diameter and intima-media thickness were documented. Pulmonary artery intimal thickening was identified when low-moderate or highmoderate echogenic areas were observed parallel and internal to the hyperechogenic vascular wall images. Color Doppler flow imaging was used to evaluate blood filling status in the pulmonary artery.

| Statistical analysis
Numeric data were presented as mean±standard distribution.
Categorical data were presented as percentage.

| RESULTS
A total of 27 patients were identified. All of them were diagnosed with Takayasu arteritis with pulmonary artery involvement. Their baseline characteristics including pulmonary and systemic artery involvements are listed in Table 1. Pulmonary artery involvement includes luminal stenosis, occlusions, tortuosity, dilation, or filling defects, and these were obtained through CTPA examinations.

| Echocardiographic identification of pulmonary artery involvements
In these 27 patients with pulmonary artery involvements shown in CTPA, 25 of them (92.6%) had positive echocardiographic findings.
None of echocardiography windows examined in this study could identify pulmonary trunk involvement. Left pulmonary artery involvement was visualized in the parasternal aortic short-axis view. Right pulmonary artery, middle segment of right pulmonary artery involvement, and right upper lobe artery involvement were better examined in the suprasternal right pulmonary artery long-axis view ( Table 2). An example of the normal anatomy of the right pulmonary artery with its branches by suprasternal right pulmonary artery long-axis window is shown in Figure 1.  T A B L E 2 Identification of pulmonary artery involvements by echocardiography

Suprasternal right pulmonary artery long-axis view Parasternal aortic short-axis view Subxiphoid view
Pulmonary trunk (2) 0, 0% 0, 0% 0, 0% F I G U R E 1 Normal anatomy of the right pulmonary artery with its branches by suprasternal right pulmonary artery long-axis window (movie clip S1). A. Suprasternal right pulmonary artery long-axis view showed the right pulmonary artery trunk, RPSA, and RPIA. B. Angiography showed the right pulmonary artery and its branches.   Histopathological examination could reveal the replacement of normal vessel lumen by loose fibrosis tissue and is presented as "vesselin-vessel" pathologic. 3 Patients with this type of lesion have increased risk for site thrombosis. The third type was the gradual narrowing of the distal vessel to an occluded end or short residue lumen (Figures 8   and 9). This typical vessel lesion was described as "rat-tail stenosis" in pulmonary angiography and CTPA studies, which indicates significant distal vessel stenosis or occlusion.
Previous studies reported that luminal occlusion was the most common finding in pulmonary artery involvement examined by CTPA. 3 In the present study, we more frequently observed luminal medium-to-high echogenic signals. This might be due to the fact that CT has relatively poor resolution to small vessels, whereas echocardiography has higher resolution compared to CTPA (10 times than CTPA) and is more sensitive to detect soft tissue and luminal changes. 9 This might imply that echocardiography could reveal a more detailed arterial wall and luminal lesion compared to computed tomographic study. In addition, in patients with luminal lesions and artery stenosis or occlusion, CTPA can show its filling defect by contrast agent, but has difficulty in providing direct evidence of intimal thickening and distinguishing the thickened intima from the extravascular soft tissue (Figures 2-7). This filling defect could be mistaken for other diseases such as pulmonary embolism or pulmonary tumor. 19 Echocardiography could clearly identify lesions in lumen and three layers of vessel (intima, media, adventitia) and distinguish them from adjacent soft tissues and luminal thrombosis. It might provide evidence for early pulmonary artery involvement in patients with Takayasu arteritis and help in the diagnosis and differential diagnosis of Takayasu arteritis.
Despite its high resolutions, echocardiography examination also has its limitations. The biggest limitation is its low penetration through bone or gas. We were not able to identify pulmonary trunk lesions. This might be due to the location of the pulmonary trunk (behind the sternum) and influences from the air in the lung.
In addition, the direction of the pulmonary trunk is parallel with the echocardiographic beam in the parasternal aortic short-axis window, was not able to be visualized due to its deep anatomy location. The right pulmonary artery, as well as its branch involvement, was best observed in the suprasternal right pulmonary artery long-axis view.
In this view, the transducer was close to the pulmonary artery trunk and the beams were perpendicular to the right pulmonary artery.
Both right pulmonary artery and its branches could be visualized very well in this window. The initial segment of right pulmonary artery lesions could also be revealed in the parasternal aortic shortaxis view and subxiphoid view. In patients with pulmonary disease, such as emphysema, subxiphoid view could visualize right pulmonary artery lesions.
The limitations of the present study were as follows: the retrospective nature of the study, single central study, and the small number of patients. We only selected patients with abnormal CTPA findings in the pulmonary artery. Therefore, we were not able to report false positive results of echocardiography. Individual performance variations by sonographers could also affect our results. Future studies with large sample size are required to confirm our findings in the current study.

| CONCLUSIONS
Echocardiography is a noninvasive test and could provide evidence of characteristic changes in intimal thickening, luminal lesions, and luminal stenosis in pulmonary arteries in patients with Takayasu arteritis. Left pulmonary artery involvement was revealed in the parasternal aortic shortaxis view. Right pulmonary artery involvement was best observed in the suprasternal right pulmonary artery long-axis view, with complementary views from the parasternal aortic short-axis and subxiphoid angles.

SUPPORTING INFORMATION
Additional Supporting Information may be found online in the supporting information tab for this article.
Movie clip S1. Normal anatomy of the right pulmonary artery with its branches by suprasternal right pulmonary artery long-axis window.