|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Radiology (Drs. Khalil, Farres, Tassart, Bigot, and Carette) and Intensive Care (Dr. Mangiapan), Tenon Hospital, Paris, France.
Correspondence to: Antoine Khalil, MD, Hôpital Tenon, Service de radiology, 4 rue de la chine, 75020 Paris/FRANCE; e-mail: antoine.khalil{at}tnn.ap-hop-paris.fr
| Abstract |
|---|
|
|
|---|
Patients and methods: Five consecutive patients with PAVMs underwent HCT, CEMRA, and pulmonary artery digital subtraction angiography (PADSA). CEMRA was performed during the pulmonary arterial phase of an IV bolus of gadolinium. PADSA was performed during the embolization procedure. All images were examined for PAVMs. The site and size of aneurysms were specified, as well as the diameter of the vascular pedicles.
Results: Thirty PAVMs were detected by CEMRA and 38 by HCT. All 20 PAVMs at least 5 mm in diameter and 10 of the 18 PAVMs < 5 mm in diameter identified on HCT were also identified by CEMRA. Whatever the site, all PAVMs with a feeding artery diameter of at least 3 mm (ie, PAVMs with clinical consequences) were detected by CEMRA. No false-positive results were obtained with CEMRA. CEMRA therefore had a sensitivity of 78% and a specificity of 100%.
Conclusions: CEMRA, a nonionizing and noninvasive procedure, has high sensitivity and specificity for the diagnosis of clinically relevant PAVMs.
Key Words: helical CT magnetic resonance angiography pulmonary arteriovenous malformations
| Introduction |
|---|
|
|
|---|
MRI has been used for the diagnosis and evaluation of PAVMs > 10 mm in diameter.4 5 6 Now, it is also possible to diagnose large PAVMs by using enhanced breath-hold three-dimensional magnetic resonance (MR) angiography.7
The aim of this study was to determine if this diagnostic tool was capable of identifying all clinically significant PAVMs.
| Materials and Methods |
|---|
|
|
|---|
|
CEMRA was performed with a 1.5-T Siemens Magnetom Vision system (Siemens; Erlangen, Germany) equipped with 25 mT/m, 600-µs rise time gradients, using a phased-array body coil. Patients were supine with their arms above their heads. A breath-hold sagittal three-dimensional fast imaging with steady-state precession sequence was acquired with the following parameters: repetition time/echo time/flip angle, 5/2/40; 10- to 11-cm slab, with 30 to 44 partitions (partition thickness, 2.3 to 3.5 mm); 350 to 400-mm rectangular field of view, and 128 x 256 matrix, resulting in a 20- to 30-s breath hold. Each lung was explored separately in the sagittal plane. The second lung was explored about 5 min after the first. Contrast was administered as a bolus via a 20-gauge IV catheter introduced into an antecubital vein, using an automated injector (Spectris; Medrad; Pittsburgh, PA) at a rate of 2.5 mL/s. Fifteen milliliters of gadolinium was administered, followed by 30 mL of saline solution, independently of the patients weight. To ensure that peak pulmonary arterial enhancement coincided with the acquisition of the central lines of the K space (region of maximal contrast), a test dose of 1 mL of gadolinium was administered, followed by 30 mL of saline solution (power injector) to determine exactly when peak enhancement occurred. The test slice was placed at the level of the pulmonary artery bifurcation. Analysis was based on MIP reconstruction and 3-mm-thick three-dimensional multiplanar reformation in the axial, sagittal, and coronal planes. MIP images of the anterior view (180° angle view) of each lung were available at 24 different projection angles. The good image quality of the CEMRA was assessed as a good visualization of the segmental pulmonary artery without motion artifact or low contrast enhancement.
Diagnosis of PAVMs by CT and MR angiography was based on visualization of an anatomic configuration of a vascular lesion (enlarged arteries feeding a nodule or a serpiginous mass with an enlarged vein). The following parameters were analyzed for each PAVM:
CEMRA and unenhanced HCT findings were reviewed separately and classified by two radiologists (A.K. and M.F.C.)
As angiography was performed exclusively for therapeutic purposes; an optimal amount of contrast medium was injected during the procedure to allow treatment of the greatest number of PAVMs per session. Thus, only hyperselective angiograms of feeding arteries were obtained, and not generalized pre-embolization angiograms. The final control lateral view after embolization of all the PAVMs in one lung was obtained by injection into the main pulmonary artery on that scale to demonstrate both complete closure of the treated arteries and the absence of any additional pedicle feeding the malformations. All but one of the PAVMs visualized on HCT were occluded during one treatment session. In the last case, the postvaso-occlusion control of a giant PAVM showed numerous small PAVMs. These small PAVMs were not visualized on either HCT or CEMRA. However, they were visible on HCT performed after vaso-occluson of the giant PAVM.
| Results |
|---|
|
|
|---|
|
|
In patient 4, CT showed one PAVM in the right lower lobe and one nodule, with no dilated feeding artery or draining vein in the middle lobe. No PAVMs were visualized either by CEMRA or by pulmonary artery digital substraction angiography of the middle lobe during percutaneous transcatheter embolization of the PAVM in the right lower lobe (Fig 2 ). CEMRA showed increased signal intensity on the arterial phase, pointing to a hypervascular tumor such as a carcinoid tumor. After surgical resection, the diagnosis of a carcinoid tumor was confirmed.
|
| Discussion |
|---|
|
|
|---|
In the adult population, indications for treatment comprise three broad categories: intolerance of exercise, prevention of neurologic complications, and prevention of hemoptysis and hemothorax.8 The primary indication for PAVM occlusion is to prevent neurologic complications. Serious neurologic events including transient ischemic attack, stroke, and brain abscess occur in 30 to 40% of patients with PAVMs that have feeding arteries at least 3 mm in diameter.8 White et al8 believe that all PAVMs with arteries > 3 mm in diameter should be occluded to potentially limit serious neurologic complications. PAVMs with arteries < 3 mm in diameter should be monitored and treated only when they expand.
Among the available examinations, pulmonary angiography is considered the standard diagnostic tool for PAVMs. Recently, unenhanced three-dimensional HCT was reported to be a reliable, noninvasive diagnostic tool for PAVMs,2 but it is a method that involves significant radiation exposure, especially for young women.3
A few authors have reported the use of MRI for the diagnosis of large PAVMs.4 5 However, MR techniques have low sensitivity for the diagnosis of all clinically significant PAVMS. A new MR angiography technique has recently been developed, called breath-hold gadolinium-enhanced MR angiography.10 11 The examination time is significantly shortened because ECG gating is not necessary and the breath holding eliminates breathing-induced motion artifacts. However, it also requires the use of a phased-array body coil. This MR angiography technique has accurately identified stenotic and occluded abdominal vessels10 and pulmonary emboli.12 Only one report of breath-hold gadolinium-enhanced MR angiography of a solitary large PAVM has been published in the literature, but the diameter of the aneurysm was > 5 cm.7
Our results indicate that breath-hold enhanced three-dimensional MR
angiography is an accurate method for the detection of PAVMs > 5 mm
in diameter. All PAVMs that have feeding arteries
3 mm in diameter
were diagnosed by this technique. In this short series, the sensitivity
and specificity of MR angiography were both 100% for PAVMs with
probable clinical consequences. The use of nonferromagnetic coils to
occlude the feeding artery would permit follow-up MRI.
In conclusion, MR angiography is highly useful for displaying small PAVMs (feeding artery < 3 mm in diameter). It is also useful for counting the vessels feeding the aneurysm sac, making it valuable for treatment planning. The need for diagnostic pulmonary angiography would be obviated. Global pulmonary angiography was only used to determine whether vaso-occlusion was fully successful.
|
| Footnotes |
|---|
Received for publication March 26, 1999. Accepted for publication October 18, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y-L Hsu, H-C Wang, and P-C Yang Desbaric air embolism during diving: an unusual complication of Osler-Weber-Rendu disease Br. J. Sports Med., August 1, 2004; 38(4): e6 - e6. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-P. Dinkel and J. Triller Pulmonary Arteriovenous Malformations: Embolotherapy with Superselective Coaxial Catheter Placement and Filling of Venous Sac with Guglielmi Detachable Coils Radiology, June 1, 2002; 223(3): 709 - 714. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |