the number of patients will be included is reach

• The purpose of this study is to evaluate the diagnostic accuracy of thoracic magnetic resonance imaging with gadolinium-enhanced, unenhanced and perfusion sequences in patients with clinically suspected acute pulmonary embolism
• Thoracic CT angiography (CTA) will serve as reference standard
• Result of MRI will not interfere with patients' management
• Untreated patients with negative CTA will have 3-month follow-up to verify they were free of thrombose-embolic disease

Background In patients with clinically suspected pulmonary embolism (PE)°with a contraindication to thoracic CT angiography, there is a need for an alternative diagnostic procedure. MRI has not been fully evaluated in this field; moreover, recent technological advances make it necessary to re-evaluate its performance for PE diagnosis.

Design Prospective monocentric study

• Patients with clinically suspected acute pulmonary embolism will undergo thoracic magnetic resonance imaging if inclusion criteria are fulfilled.
• A non-inclusion register will be establish for patients fulfilling the inclusion criteria , not included because MRI was not available (off-hours presentation , another patients already included in the protocol on the same day )

Estimated enrolment : 280 (based on a 25% prevalence of PE in our institution and an expected 80% sensitivity of MRI) Study start date: June 2007 Estimated study completion date: 17 months later (40 to 50 presentations for PE suspicion each month, 20 inclusions expected per month)

Magnetic Resonance imaging: performed on a 1.5 Tesla unit with 3 different sequences

• Unenhanced Steady State Free Precession (SSFP) sequences
• Perfusion imaging following Gadolinium injection at a rate of 5 c/s and a dose of 0.1mmol/kg
• Magnetic resonance angiography following an injection of 0.1mmol/kg of gadolinium at 3ml/s MRI studies will be interpreted secondarily by 2 independent radiologists, blinded to CTA results and clinical probability.

Two different readings will be performed, one global reading and one reading of each sequence separately, displayed in a random order (access base)

Objectives

• to evaluate MRI performance for PE diagnosis globally
• to evaluate the diagnostic value for each sequence (especially the negative predictive value of a normal perfusion sequence)
• to evaluate inter-observer agreement

MRI and CTA have to be performed within 24 hours

Inclusion Criteria:

• Clinically suspected acute pulmonary embolism
• 18 years old or more
• D-dimers more than 500 or high clinical probability of pulmonary embolism according to the revised Geneva score
• Thoracic CT angiography performed within 24 hours
• Informed consent

Exclusion Criteria:

• - Contraindication to MRI (pace maker, claustrophobia, any implanted ferromagnetic foreign body)
• Obese patients too large to fit in MRI unit
• Allergy to gadolinium-containing contrast agent or to iodinated contrast media
• Renal insufficiency (clearance less than 30 ml/mn)
• Anticoagulation at a curative dose started more than 48 hours prior to MRI
• Life expectancy less than 3 months
• Pregnancy

patients with acute pulmonary embolism suspicion who fulfil the inclusion criteria