topeeng.gif (8383 bytes)

[ Scientific Activity - Actividad Científica ] [ Brief Communications - Temas Libres ]

The Pulmonary Artery Index: A new echo criterion that should be considered in suspected pulmonary artery embolism.

Sorrell, Vincent Leigh; Simpson, Jerry; Carraway, Deena; Sumner, Andrew David.

East Carolina University School of Medicine. Section of Cardiology
Greenville, USA


Introduction. Pulmonary artery thromboembolism (PTE) is a deadly disease requiring a high index of suspicion in addition to sophisticated imaging data for confirmation. We have recently identified a number of patients with isolated, dilated right pulmonary artery branch and performed this study to investigate this finding.
Objective. Determine if the pulmonary artery size is beneficial in predicting the presence of a PTE.
Methods. We retrospectively reviewed the echo findings in 17 patients with confirmed PTE. In addition to the more commonly identified echocardiographic variables, we calculated the pulmonary artery index (PAI) as the ratio of the main pulmonary artery (MPA) to the right pulmonary artery (RPA). PAI = MPA / RPA.
Results. The following results are listed in order of decreasing incidence: Abnormal septal motion (88%); PAI < 1.3 (86%); TR velocity > 3 M/s (83%); Dilated IVC (70%); TR grade > moderate (67%); Dilated RV (57%); Abnormal RV function (57%); LV cavity obliteration (41%).
We also examined a cohort of normal volunteers and compared them to patients: MPA 22.1 mm vs 26.4 mm; RPA 13.3 vs. 23.2; PAI 1.14 vs 1.66. (p value < 0.03 for all comparisons).
Discussion. In this pilot study, a PAI < 1.3 was one of the most common echo variables identified. The clinical importance of this variable as well as the effect of other causes has not been clarified in this retrospective study. Since echocardiography is becoming more common during the evaluation of these patients, and the prognosis is worse when RV dysfunction is seen, further investigations should proceed in this area. We strongly believe an "echo score" should be developed utilizing numerous echo variables, and the PAI may be included within this score.
Conclusion. We conclude that the PAI is an additional echo variable that should be considered during the evaluation of suspected PTE. Importantly, the PAI was abnormal in 5 patients with a normal appearing RV.



Pulmonary artery thromboembolism (PTE) is a deadly disease that requires a high index of suspicion in addition to sophisticated imaging tools for adequate diagnosis and subsequent treatment. The most commonly utilized noninvasive tools in the US for this purpose are the ventilation-perfusion nuclear scan (V-Q Scan) and the high-resolution, computed tomographic scan (HR-CT scan). Each of these has the disadvantage of requiring an intravenous (IV) line placement and administration of an imaging agent. Furthermore, neither is a portable tool, requiring the patient to be transported to another imaging area and preventing any bedside diagnosis.

Echocardiography (echo) offers a number of technical advantages to these aforementioned tools, but is variably utilized for the assessment of suspected PTE. Echo is widely available, even in under-developed regions where these other tools are not available. The cost of echo is usually less than V-Q scan or HR-CT scan. The ultrasound system is portable allowing bedside imaging. The images are on-line and allow immediate information to be analyzed without the need for IV placement.

Despite these technical advantages, there are serious limitations of echo for the diagnosis of suspected PTE. Most importantly, no definite diagnostic criteria exist. This project was initiated to develop a better understanding of the incidence of previously documented echo variables that are identified in patients with PTE and to further clarify the utility of a new echo variable: the pulmonary artery index (PAI). It was hoped that this type of investigation would continue to raise interest and continuing research in this important area. These authors further believe that similar studies are needed and may culminate in the eventual development of an "echo score" that would utilize multiple echo variables to predict the likelihood of an individual’s risk of PTE and possibly even risk stratify patient outcomes. Already, the current recommended guidelines for the management of PTE utilize the echo finding of a dilated right ventricle (RV), even in the absence of hemodynamic compromise, as a criterion for the administration of thrombolytic therapy.



This study was performed to investigate the relationship between the ratio of the size of the main pulmonary artery (MPA) to the proximal right pulmonary artery (RPA) in patients with high probability VQ scans (hence: 87% likelihood of having a PTE). This MPA / RPA ratio is termed the PAI (figure 1). This index was chosen after an incidental case in which the echo was performed for a confirmed PTE and we noticed an isolated, dilated, RPA. We hypothesized that the RPA may dilate from acute cor pulmonale, but that the total extent of dilatation may vary from the main to branch arteries. The possibility that the right or left pulmonary arteries (LPA) may dilate variably dependent upon the location of the PTE, was also a consideration, but could not be adequately investigated due to the consistently poorer visualization of the LPA.

fig1.jpg (98313 bytes)

Figure 1



48 high-probability VQ scans were performed at our institution in 1998. 17 of these patients also had a transthoracic (TTE) or transesophageal (TEE) echo performed during that hospitalization. These 17 studies were retrospectively analyzed for this investigation. Not all data was available on every echo study. The following echo variables were analyzed: RV size (semi-quantitated as normal or enlarged, based upon the relative LV size); global RV function (visually estimated as normal, mildly or severely hypokinetic); inferior vena cava (IVC) size in expiration and inspiration (mm); LV global function (visually estimated as hyperdynamic, normal, or reduced); interventricular septal (IVS) motion (normal or paradoxical / ventricular interdependence); tricuspid regurgitation (TR) (visually estimated as mild, moderate, or severe using color flow Doppler); TR velocity (continuous wave Doppler envelope – meters / second [M/s]); MPA size (measured within 1-2 cm of the bifurcation into the RPA and LPA [mm]); RPA size (measured within 1-2 cm of the bifurcation [mm]); and the PAI (the ratio of the MPA / RPA). All measurements were performed by a level 3 trained echocardiographer (VLS). Data was not included when the image quality was not adequate for interpretation. It became apparent that the size of the proximal LPA could not be adequately measured in the majority of patients due to technical constraints. Therefore, only the RPA and MPA were statistically analyzed.

In addition to the 17 patients evaluated, 7 normal controls (echo sonographers and medical students without known cardiovascular diseases and normal echo studies) were imaged and their PAI was obtained.

An example of the PAI measurements of 2 patients (surface and TEE) is shown in Figure 2. Controls are also shown for a direct comparison. (note: The control TEE is an example only and was not included in the analysis).

fig2.jpg (103306 bytes)

Figure 2



Table 1 illustrates the incidence of the commonly obtained echo variables and the new PAI in order of frequency. The most commonly identified echo abnormality in this patient cohort was the paradoxical IVS motion - or what many have termed "ventricular interdependence". This was noted in 88%. The next most frequent occurrence was the finding of a PAI < 1.3 (86%). This ratio was used since it provided the greatest distinction from the patient population and the control population. Other echo variables that were commonly present were a dilated and plethoric IVC, an increased TR grade and velocity, RV dysfunction, and an underfilled, hyperdynamic LV. The other echo variables were less frequently observed.

table1.jpg (57788 bytes)

The PAI could be measured retrospectively in 86 % of patients (14 / 17) and prospectively in 100 % (7/7) of controls. Table 2 illustrates the direct comparison of these "pulmonary artery" parameters. The difference in the average PAI in the patient population (1.15) and in the control population (1.64) was statistically significant (p-value < 0.02).

table2.jpg (32487 bytes)



The major limitation of this study is the retrospective design and related potential biases. To minimize error, all data was read blindly and measurements were not analyzed until all had been completed. Also, there is the possibility that some patients with a high probability VQ scan did not have a PTE or that the PTE was not acute and the echo findings may actually relate to chronic pulmonary changes.

Despite these limitations, we believe this study is important. Requests for the performance of echo are becoming more frequent during the routine evaluation of patients suspected of having a PTE. Except for the direct visualization of a thrombus within the pulmonary artery, no single echo variable is 100 % accurate for the diagnosis of PTE. However, the other commonly utilized noninvasive, diagnostic tools for this purpose, are also not absolute. Certainly, echo provides incremental information that is useful during the evaluation of this patient cohort. Already, the echo finding of RV dysfunction is an indicator of worse patient outcomes and has led to the recommendation of thrombolytic treatment in these circumstances. The prognostic outcome of the other echo variables has not been reported.

Numerous echo variables have been reported in the literature as occurring in the setting of PTE. Similar to all echo diagnoses, it is important to assess multiple variables when considering whether the echo study is consistent with the presence of a PTE. Being retrospective, this investigation should be thought of as a pilot study that has further generated an interesting hypothesis. Do the pulmonary arteries differentially dilate in the setting of a PTE? Certainly, this limited study suggests that the ratio of the MPA to the RPA may indeed be an additional differentiating variable and should be included during the routine echo assessment of the patient with a suspected PTE.



A PAI < 1.3 was identified in 86 % of patients with a high-probability VQ scan. The high incidence of this variable makes it one of the more frequently identified parameters in this patient cohort. A prospective study will need to be performed prior to any conclusions of the validity of the pulmonary artery index. With the increasing awareness of the utility of echo by the medical-pulmonary specialties, the number of requests for echo studies in this patient population are likely to increase. These authors hope that a heightened interest by the echo community will result in additional investigations surrounding the echo findings during PTE. Determining the prospective incidence of the numerous echo parameters in acute PTE should eventually result in the development of an echo score that has both diagnostic and prognostic capabilities.


Questions, contributions and commentaries to the Authors: send an e-mail message (up to 15 lines, without attachments) to , written either in English, Spanish, or Portuguese.




gador2.gif (2307 bytes)     klchlogo4.gif (6328 bytes)