New Approach Improves MR Detection of Pulmonary Embolism

James Brice

March 22, 2012

March 22, 2012 — A new approach to magnetic resonance angiography (MRA) for detecting life-threatening pulmonary emboli has taken a bite out of the high technical failure rate. This high failure rate discouraged physicians from applying it to patients who are considered poor candidates for the gold standard of computed tomography scanning (CT) to diagnose the condition, according to a study published in the April issue of Radiology. Bobby Kalb, MD, an associate professor of radiology at the University of Arizona Medical Center in Tucson is first author.

With help from radiology colleagues at the University of Arizona and biostatisticians at Emory University in Atlanta, Georgia, Diego R. Martin, MD, has shown that the combined use of 3 standard vascular MR pulse sequences and a respiratory gating technique may be reasonably accurate for detecting pulmonary embolism (PE). It is also less likely to produce the kind of nondiagnostic exams that dominated the findings of a major clinical trial 3 years ago.

The 2010 Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) III trial, sponsored by the National Institutes of Health, found that the combination of contrast-enhanced MRA and venography was technically inadequate in 1 of every 4 exams. When they were readable for detecting PE among 371 adult patients at 7 facilities, the sensitivity of the 2-sequence exams was 78%, and the specificity was 99%.

Large pulmonary embolus (arrows) at the bifurcation of the right pulmonary artery in a 67 year old female. (A) CTPA, (B) bolus-triggered 3D MRPA, axial reformations, (C) recirculation phase 3D GRE and (D) free-induction steady state precession with ECG gating and respiratory triggering (TFISP-T).
Source: Diego R. Martin, MD, PhD, FRCPC/The Cosden Professor and Chair of Radiology/University of Arizona/University Medical Center

Dr. Martin, chairman of radiology at the University of Arizona Medical Center, demonstrated with data from 22 patients who had confirmed PE that an alternative MR protocol cuts the technical failure rate to 14%. The sensitivity of images generated by 3-pulse sequences was 84%.

The protocol included a standard bolus-triggered, contrast-enhanced MR angiographic sequence; an enhanced, low-flip angle 3-dimensional gradient-echo (3D GRE) sequence; and a nonenhanced, respiratory-triggered, free induction with steady-state precession pulse.

Researchers have been eager to develop effective ways to apply radiation-free MRA to diagnose PE since contrast-enhanced CT angiography (CTA) became the gold standard for diagnosis in the mid-2000s.

High-Radiation CTA

Multidetector CTA produces extremely high-quality angiographic images while exposing patients to high levels of ionizing radiation. CTA for PE involves an effective dose of 10 mSv to radiosensitive parts of the body, and about 20 mSv to glandular breast tissue. One study, cited in Dr. Martin's trial, estimated that a 40-year-old woman has a 1 in 20 chance of developing a radiation-induced cancer from pulmonary CTA. The risk doubles for a 20-year-old woman ( Arch Intern Med. 2009;169:2078-2086).

Because of CT's safety implications, imaging practitioners have seen a need for MR imaging to diagnose PE for pregnant women and young adults who have compromised renal function that places them at high risk for adverse reactions from infused iodinated CT contrast media, Dr. Martin told Medscape Medical News. Many young adults undergo CT for PE unnecessarily, he noted, as only 5 or every 100 who undergo CT to detect PE are actually diagnosed with the potentially lethal condition.

Improved Grayscale Presentation

The Arizona group's 3-sequence MR protocol attempts to improve on previous approaches by adding a broader range of grayscale values to make emboli stand out from the vessel walls and other background anatomy, Dr. Martin noted.

The difference can be seen by comparing emboli detected with standard MR pulmonary angiogram and 3D GRE. The angiogram generates a binary type of contrast, with structures depicted as either very bright or dark. 3D GRE creates a grayscale for an array of signal intensities that make PE appear more conspicuous.

The protocol addresses other sources of technical failure through its use of a navigator respiratory gating technique originally developed for MR cardiac applications. It eliminates motion artifacts for patients who cannot hold their breath for more than a few seconds. The use of non–contrast free precession imaging addresses the problem of correctly timing image acquisition with the arrival of infused gadolinium contract media in the targeted field of view.

"The beauty of MRI is that there is more than one way to do it," Dr. Martin said. "The objective here was to look at a multiple sequence approach."

Examined individually, the 3 sequences were from 55% to 73% sensitive to emboli, but in combination, they were 84% sensitive for patients previously diagnosed with PE. At least 1 of 3 sequences produced technically acceptable results for all patients in the trial.

"This is certainly a step forward," Pamela Woodard, MD, professor of radiology at Washington University, St. Louis, Missouri, said to Medscape Medical News. She designed the MR imaging protocol for the PIOPED III trial and serves on the American College of Radiology Appropriateness Criteria panel for PE imaging.

Dr. Martin believes the findings from his group's protocol set the stage for another prospective, multicenter trial similar to PIOPED III to establish its clinical efficacy.

Dr. Woodard sees opportunities to first fine-tune the procedure through single-center trials to optimize the 3-pulse sequence and to determine whether gadofoveset trisodium (Ablavar, Lantheus Medical Imaging, Inc), the first intravascular MR contrast agent approved by the US Food and Drug Administration, will improve on the performance of conventional gadolinium-based contrast media for the procedure.

Dr. Martin has disclosed no relevant financial relationships. Dr. Woodard is principal investigator for numerous US Food and Drug Administration phase 1 and 2 trials, including research partially funded by Astellas Pharma and Siemens Medical Systems.

Radiology. 2012;263:271-278. Full text


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.