Acute Pulmonary Artery Embolism Detected by Noncontrast Functional Lung Magnetic Resonance Imaging.

A 30-year-old woman presented to the emergency department with acute shortness of breath, dizziness, and leftsided thoracic pain. The patient had been diagnosed with adenocarcinoma of the lung and metastases in central nervous system, liver, and bones 2 years earlier. After initial radiation and chemotherapeutic treatment, the patient showed a mixed response and was subsequently treated with the tyrosinekinase inhibitor crizotinib, as well as further radiation of the progressive and symptomatic cerebral metastases. On admission, the patient was afebrile, displayed unimpaired oxygen saturation in room air, and presented with mild pitting edema of the calves. The ECG was without pathological findings and d-dimers were slightly elevated. Because of her clinical presentation and the increased pretest probability (clinical presentation and cancer diagnosis), a contrastenhanced multidetector computed tomography was conducted to test for pulmonary embolism (PE). Multidetector computed tomography, however, showed no thrombotic mass in the main and segmental pulmonary arteries but a subpleural wedgeshaped consolidation, rated suspicious for beginning ischemic pneumonia, for example, because of subsegmental embolism, which is commonly missed by conventional multidetector computed tomography (Figure 1). The patient consecutively underwent ventilation and perfusion single-photon emission computed tomography (SPECT), which revealed a mismatch consisting of a perfusion defect in segments 8 and 9 of the left lower lobe but unimpaired ventilation corresponding to PE (Figure 2). After anticoagulation was initiated, the patient was referred to functional lung magnetic resonance imaging (MRI) using Fourier decomposition (FD-MRI) to correlate the findings with this new technique. FD allows for separation of the periodic MR signal changes induced by respiration and pulsatile blood flow in the lungs. Thereby, ventilation and perfusion maps can be generated in free breathing and without application of ionizing radiation or contrast agents. FD-MRI was performed using the SElf-gated Non–Contrast-Enhanced FUnctional Lung (SENCEFUL) MRI approach, providing maps in sufficient spatial resolution and reconstruction of complete cardiac/respiratory cycles. The acquisition of a slice with 1-cm thickness in the coronar plane takes ≈2 minutes. Data acquisition of the entire lung amounts to 15 to 20 slices, resulting in an acquisition time of 25 to 40 minutes. Including image postprocessing, the overall examination time amounts to ≈90 minutes, which is comparable with ventilation and perfusion SPECT. SENCEFUL FD-MRI can be implemented in any commercially available MRI system as technical details of the sequence have been published. Limitations of FD-MRI refer to the common MRI contraindications. Evaluation and development of SENCEFUL-MRI is part of a research protocol on functional lung MRI in the reporting department. The MRI ventilation maps of the reported patient displayed a physiological distribution of the ventilation without regional impairment matching the scintigraphy results (Figure 3A and 3B). Mapping of the lung perfusion showed a regional defect in the left lower lobe corresponding to the previous ventilation and perfusion SPECT findings (Figure 3C and 3D). The reported case demonstrates the potential of functional FD-MRI for diagnosis of PE. FD-MRI results matched the findings of SPECT, which is currently the established reference standard in functional lung imaging. Concerning diagnostic imaging of PE, SPECT is commonly used to assess thromboembolic events in subsegmental arteries. Using multidetector computed tomography, these small subsegmental arteries cannot be safely evaluated about PE as the limited spatial resolution is insufficient for differentiation between thrombotic material and intraluminal partial volume effects. In addition to the promising diagnostic potential as illustrated within this initial report, no ionizing radiation is required for FD-MRI data acquisition in contrast to established imaging modalities for PE diagnosis. Therefore, its application in clinical routine is desirable especially for patients requiring particular radiation protection, such as young individuals or pregnant women. As the latter are known to feature increased risks for thromboembolism and should not be exposed to radiation, detection of PE in pregnant women might be a clinical scenario in which