Applicability of Transthoracic Contrast Echocardiography for the Diagnosis and Treatment of Idiopathic Pulmonary Arteriovenous Malformations

Yujiao Deng, Xin Huang, Guangyi Wang, Jian Cao, Shengshu Wang, Yue Li, Yiru Wang, Jing Ye, Peifang Zhang, Xiaotian Chen, Yukun Luo, Kunlun HeFrontiers in Cardiovascular Medicine 2021 · DOI

Abstract

Objective: To explore the preferred test to screen for pulmonary arteriovenous malformations (PAVMs) and to predict the probability of interventional embolization. Methods: We performed a retrospective observational study evaluating patients with idiopathic PAVMs from 2009 to 2019. After clinical evaluation, a total of 105 patients were studied, including 71 patients with positive digital subtraction pulmonary angiography (DSPA) findings and 34 with negative DSPA findings. Results: The majority of patients with idiopathic PAVMs were female. We found a good κ-coefficient of 0.77 for inter-observer agreement concerning the pulmonary right-to-left shunt (RLS) grade on TTCE. The positive predictive value (PPV) of pulmonary shunt grades 2 and 3 on TTCE for the possibility of embolization were 0.21 and 0.87, respectively. Conclusion: TTCE is the preferred screening test for PAVMs. The pulmonary RLS grade on TTCE not only identifies the likelihood of PAVMs but also predicts the probability for embolization.

Introduction

Pulmonary arteriovenous malformations (PAVMs) are abnormal direct communications between pulmonary arteries and veins, bypassing the capillary bed. This results in a right-to-left shunt (RLS), leading to complications such as hypoxemia, paradoxical embolism, stroke, or brain abscesses. While often associated with Hereditary Hemorrhagic Telangiectasia (HHT), many cases are idiopathic, particularly in Asian populations.

The clinical challenge lies in effective screening. Conventional chest X-rays frequently miss these malformations due to anatomical overlaps, while Digital Subtraction Pulmonary Angiography (DSPA), the diagnostic gold standard, is invasive and involves radiation. There is a critical need for a reliable, non-invasive screening tool that can also predict the necessity of interventional treatment (embolization).

Methods

This retrospective study analyzed 105 patients at the PLA General Hospital (2009–2019) who underwent clinical evaluation for suspected PAVMs. The diagnostic performance of three modalities was compared: Chest Radiography, Transthoracic Contrast Echocardiography (TTCE), and DSPA.

Flowchart of the study population and patient selection process.

TTCE, often called the "bubble test," involves injecting agitated saline into a peripheral vein. The presence and density of microbubbles in the left heart chambers indicate an RLS. Shunts were graded on a 3-point scale based on the number of microbubbles observed: Grade 1 (<30), Grade 2 (30–100), and Grade 3 (>100).

Key Results

The study found that TTCE is a highly reliable and reproducible test, with a κ-coefficient of 0.77 for inter-observer agreement.

Diagnostic and Predictive Accuracy

TTCE significantly outperformed chest radiography, which had a high false-negative rate (78.9%). The predictive power of TTCE increased dramatically with the shunt grade. The Positive Predictive Value (PPV) for identifying a PAVM on DSPA was:

  • Grade 1: 0.14
  • Grade 2: 0.74
  • Grade 3: 0.97
Digital subtraction pulmonary angiography (DSPA) demonstrating a PAVM.

Crucially, the TTCE grade also predicted the need for intervention. A Grade 3 shunt had a PPV of 0.87 for predicting the necessity of embolization. In contrast, Grade 1 shunts rarely required interventional treatment.

Treatment Efficacy

A total of 64 PAVMs were successfully treated via embolization using coils or Amplatzer devices. Following the procedure, patients showed significant physiological improvement, with mean blood oxygen saturation ($SpO_2$) rising from 91.87% to 97.03% without major complications.

Discussion and Conclusion

Idiopathic PAVMs in this cohort were predominantly solitary (74.6%) and located in the lower lobes. This differs from HHT-related cases, which are often multiple. The study demonstrates that TTCE is not just a binary diagnostic tool but a prognostic one. The intensity of the "bubble" shunt directly correlates with the clinical significance of the malformation.

While the retrospective and single-center nature of the study are limitations, the high PPV of Grade 3 shunts suggests that TTCE should be the primary screening modality. It effectively identifies patients who are most likely to benefit from invasive angiography and subsequent embolization, optimizing patient care and resource allocation.

In conclusion, TTCE is the preferred screening test for PAVMs. The pulmonary RLS grade on TTCE identifies the likelihood of PAVMs and serves as a robust predictor for the probability of requiring interventional embolization.

Summary updated

Abstract

<b>Objective:</b> To explore the preferred test to screen for pulmonary arteriovenous malformations (PAVMs) and to predict the probability of interventional embolization. <b>Methods:</b> We performed a retrospective observational study evaluating patients with idiopathic PAVMs from 2009 to 2019. After clinical evaluation, a total of 105 patients were studied, including 71 patients with positive digital subtraction pulmonary angiography (DSPA) findings and 34 with negative DSPA findings. The following patient data were assessed: blood test, chest radiograph, transthoracic contrast echocardiography (TTCE), and DSPA findings. <b>Results:</b> The majority of patients with idiopathic PAVMs were female (66.2% with positive DSPA findings). We found a good κ-coefficient of 0.77 with strong consistency for inter observer agreement concerning the pulmonary right-to-left shunt (RLS) grade on TTCE, which was superior to conventional chest radiographs. The positive predictive value (PPV) of the radiographic features for PAVMs on DSPA was 0.83 (95% CI 0.64-1.0) and 0.44 for the possibility of embolization (95% CI 0.19-0.70). The PPV of the shunt grade of PAVMs on DSPA was 0.14 (95% CI 0.01-0.29) for grade 1, 0.74 (95% CI 0.60-0.88) for grade 2, and 0.97 (95% CI 0.92-1.0) for grade 3. The PPVs of pulmonary shunt grades 2 and 3 on TTCE for the possibility of embolization for PAVMs were 0.21 (95% CI, 0.05-0.36) and 0.87 (95% CI, 0.79-0.99), respectively. <b>Conclusion:</b> TTCE is the preferred screening test for PAVMs. The pulmonary RLS grade on TTCE not only identifies the likelihood of PAVMs but also predicts the probability for embolization.
BibTeX
@inproceedings{deng2021applicability,
  author = {Yujiao Deng and Xin Huang and Guangyi Wang and Jian Cao and Shengshu Wang and Yue Li and Yiru Wang and Jing Ye and Peifang Zhang and Xiaotian Chen and Yukun Luo and Kunlun He},
  title = {{Applicability of Transthoracic Contrast Echocardiography for the Diagnosis and Treatment of Idiopathic Pulmonary Arteriovenous Malformations}},
  booktitle = {Frontiers in Cardiovascular Medicine},
  year = {2021},
  doi = {10.3389/fcvm.2021.656702},
}
Figure