|Home | About | Journals | Submit | Contact Us | Français|
It is suggested that transesophageal echocardiography (TEE), by detecting thromboemboli in the proximal parts of the pulmonary arteries, is useful in the diagnosis of pulmonary embolism. However, the data on visualization of the pulmonary arteries are limited. The extent of the pulmonary arteries that can be precisely visualized during biplane TEE was assessed in 51 consecutive patients (23 female, 28 male, aged 56.6±12.5 years) without structural heart disease. The main pulmonary artery and the right pulmonary artery were detected in 96.1% and 94.1% of patients, respectively. Although the proximal part of the left pulmonary artery was found in only 47.0% of patients, its distal part was visualized in 92.2%. During TEE, proximal parts of the lobar arteries on both sides were visualized in 88.2% of patients. Thus, the central pulmonary arteries including proximal parts of the lobar branches can be precisely visualized by biplane TEE in the majority of patients. Only the proximal part of the left pulmonary artery is difficult to assess.
Transesophageal echocardiography (TEE) has proved to be very useful in the diagnosis of various heart diseases and pathologies of the thoracic aorta. There is increasing evidence that TEE can confirm pulmonary embolism by direct visualization of thromboemboli located in the proximal pulmonary arteries, while transthoracic echocardiography only rarely detects pulmonary artery thrombi (1–3). However, it should be remembered that transthoracic echocardiography offers precise noninvasive evaluation of right ventricular overload, which may be a very helpful indirect sign of pulmonary embolism. Although TEE is used in patients with suspected pulmonary embolism, some controversies exist regarding the limitations of this technique due to the topography of the mediastinal structures. This applies especially to the left pulmonary artery (LPA), reported by many authors to be impossible to visualize with TEE (3–5). We feel that in a patient suspected of having pulmonary embolism, TEE should be performed with the knowledge of which parts of the proximal pulmonary arterial bed should be interrogated, in what way and with what chances of receiving interpretable images. Therefore, we systematically evaluated the feasibility of imaging the proximal pulmonary arteries during biplane TEE.
Visualization of the pulmonary arteries was assessed in 51 consecutive patients, 23 female and 28 male with a mean age of 56.6±12.5 years, who were undergoing routine TEE. The study group comprised 16 patients referred to the echocardiography laboratory with suspected infective endocarditis, 14 with suspected cardiogenic sources with peripheral embolism, 11 with suspected aortic dissection as an indication for TEE and 10 who were undergoing TEE before elective cardioversion of chronic nonvalvular atrial fibrillation. In all of them transthoracic echocardiography and TEE excluded any abnormalities of the heart or of the thoracic aorta.
TEE was performed with a Sonos 2000 ultrasound system (Hewlett Packard, USA) equipped with a biplane 5.0 MHz probe. An intravenous catheter was inserted into the right antecubital vein. Patients were investigated after at least a 6 h fast. They were placed on their left side. After local superficial anesthesia of the pharynx with 2% lignocaine spray and in some cases intravenous injection of 2 to 5 mg diazepam, the transesophageal probe was placed into the middle portion of the esophagus in the standard way.
The pulmonary arteries are presented schematically in Figure 1. The examination of the pulmonary arteries was started by visualizing the main pulmonary artery (MPA) in the horizontal plane. Then, after a 1 to 2 cm withdrawal of the probe, the proximal part of the right pulmonary artery (RPA) was detected. A subsequent clockwise rotation of the transducer allowed the visualization of the distal part of the RPA in the long axis view. Transverse scanning of the RPA was performed in the vertical plane. With the advancement of the probe, the area of branching to the lobar vessels was visualized. Moreover, imaging in the vertical plane showed the initial parts of lobar arteries: the intermediate trunk and the right upper lobe pulmonary artery.
After assessment of the RPA, the LPA was interrogated. This was started by imaging the MPA in the horizontal plane and then by counterclockwise rotation of the probe to follow the LPA. However, in the majority of patients, attempts to follow this vessel were unsuccessful. Only a short portion of the LPA could be detected, and the continuity of the artery was usually lost, presumably because it was shielded by the left main bronchus. In such cases, visualization of the LPA was attempted by rotating the probe approximately 120° counterclockwise from the position, permitting clear imaging of the RPA to the region close to the descending aorta. This manoeuvre usually allowed the distal part of the LPA to be located. The identity of this vessel was then confirmed by its position with respect to the descending aorta, by the massive intraluminal appearance of microbubbles following the intravenous injection of agitated saline and, most important, by its characteristic branching giving rise to the left upper lobe artery. Two-dimensional imaging was always completed by colour flow Doppler examination.
To be reported as echocardiographically visualized, segments of the pulmonary arteries had to fulfil the following criteria. The vascular walls of arteries should be unequivocally detected with their distinct borders. The lumen of the vessel should be clearly visualized. Colour flow Doppler should allow blood flow recording. Partially visualized vessels, segments without reliable image quality and undetected segments were reported as not imaged.
The local ethics committee approved the study protocol. All patients gave their informed consent.
The χ2 test was used to assess the frequency of differences between the evaluated parameters.
Table 1 summarizes the results of the visualization of the pulmonary arteries during TEE.
The MPA was visualized in almost all of the patients investigated (96.1%) (Figure 2). Both the proximal and the distal parts of the RPA were assessed precisely in 94.1% of subjects. In 57%, a characteristic perimural, immobile intra-luminal echo was found in the middle portion of the RPA close to the esophagus (Figure 3). This artefact was approximately 2 cm long and was especially visible in the horizontal plane. Transesophageal examination of the RPA in the vertical plane and colour flow Doppler imaging verified it as a side lobe probably originating from the bronchus.
The proximal LPA was less frequently visualized than the proximal RPA (47.0% versus 94.1%, respectively; P<0.01) (Figure 4). However, the distal LPA was echocardiographically detected in 92.2% of all patients (Figure 5). Furthermore, the initial part of the left upper lobe artery was found in 88.2% of investigated patients. The initial parts of the lobar arteries were also assessed on the right side. Thus, examination of the distal RPA in the transverse section of the vessel (vertical plane of the transducer) detected the proximal extensions of the lobar branches (the intermediate trunk and the right upper lobe artery) in 88.2% of subjects (Figure 6). Evaluation of the distal RPA only in the horizontal plane did not allow visualization of the intermediate trunk and right upper lobe pulmonary artery.
Although the pulmonary arteries are accessible for TEE, the feasibility of such examination has not been assessed. However, it was reported that RPA and the proximal LPA can be detected by TEE (4), in contrast to the distal LPA, which has rarely been described (6). Also, analysis of numerous reports of the detection of thromboemboli in pulmonary arteries by TEE indirectly indicates the extent of the pulmonary arteries that can be visualized. The majority of the reported thromboemboli were found in the RPA (7–14), while left-sided intrapulmonary masses were reported much less frequently (2,15,16). Authors underlined the difficulties in assessing the LPA. Moreover, it was pointed out that the lack of reliable visualization of the LPA during TEE is one of the major limitations of this method in the diagnosis of pulmonary embolism (2,3,5,14,17). The present study confirmed that the MPA and RPA can be precisely evaluated in almost all cases. However, the MPA and the proximal RPA were shielded by the main bronchi in 4% and 6% of patients, respectively. The proximal LPA was visualized in only a minority of subjects and significantly less frequently than the proximal RPA (P<0.01). However, a large counterclockwise rotation of the transducer in the direction of the descending aorta after initially scanning the left atrial appendage in the horizontal plane found the distal LPA in 88.2% of patients. A high rate of visualization of the left upper lobe artery not only increases the extent of the arteries that can be examined but also helps to verify the visualized structures. A round reverberation artefact of the descending aorta was commonly reported (18), which may be mistaken for the distal LPA. Massive intraluminal microbubbles following intravenous injection of agitated saline into a peripheral artery helps to verify the pulmonary arteries.
Precise visualization of the distal parts of both pulmonary arteries and the initial parts of their lobar branches may be especially important in the diagnosis of pulmonary embolism. Our observations indicate that thromboemboli are most frequently lodged in the regions of abrupt narrowing of the branching areas of both pulmonary arteries. In our experience, TEE evaluation of both pulmonary arteries reached 80% sensitivity in patients with suspected pulmonary embolism (19). Had we limited transesophageal examination to the MPA and the RPA, the sensitivity would have decreased to 67.5% (19), similar to the rate reported by other authors (3,5,14).
Interestingly, in 57% of patients, an immobile, perimural artefact was detected in the RPA. This intraluminal artefact may be misinterpreted by less experienced echocardiographers as perimural thrombus. This structure was especially visible in the horizontal plane. In our opinion, scanning in the vertical plane and colour flow mapping allow verification of this structure. The use of intravenous contrast may also help in verifying the artefact. The appearance of microbubbles in the whole lumen of the RPA following intravenous injection of agitated saline is proof of blood flow in the non-narrowed vessel. Therefore, it would be useful to precede TEE examination for suspected pulmonary embolism with systematic attempts to assess the pulmonary arterial tree in patients referred for TEE for other reasons.
This study was performed with a biplane transducer. Presumably, omniplane TEE would further improve visualization of the pulmonary artery, especially distal parts. However, the impression of authors who are using omniplane probes is that the additional benefit is not very significant, and seems unlikely to change the main results and conclusions of the present study.
Reliable assessment of pulmonary arteries during TEE may allow for definitive confirmation of pulmonary embolism. However, it should be remembered that topography limits TEE visualization to the proximal parts of the pulmonary vascular bed. This indicates that distally located thrombi cannot be detected. Thus, TEE may potentially serve as a way to confirm pulmonary embolism in selected patients with suspected central pulmonary embolism.
The present study was performed in patients free from pulmonary embolism, and the results may not be directly applicable to the population with suspected pulmonary embolism. However, dilation of the proximal pulmonary arteries, almost always found in patients with hemodynamically significant acute or chronic pulmonary embolism, may facilitate visualization of pulmonary arteries during TEE.
The central pulmonary arteries including the proximal lobar branches on both sides can be precisely visualized by biplane TEE in the majority of patients. Despite the echocardiographic discontinuity of the LPA, its distal part should be actively sought because of the high probability that it can be visualized.
Only the proximal LPA, which is shielded by the left main bronchus, is difficult to assess. A perimural artefact, often present in the RPA, may be potentially misinterpreted as thrombus.