SECTION I: Clinical Considerations and Technical Specifications
A. Clinical Considerations
1. Timing of MRI Following Prostate Biopsy Hemorrhage
Manifested as hyperintense signal on T1W, may be present in the prostate gland, most commonly the peripheral zone (PZ) and seminal vesicles, following systematic transrectal ultrasound-guided systematic (TRUS) biopsy and may confound mpMRI assessment. When there is evidence of hemorrhage in the PZ on MR images, consideration may be given to postponing the MRI examination until a later date when hemorrhage has resolved. However, this may not always be feasible or necessary, and clinical practice may be modified as determined by individual circumstances and available resources. Furthermore, if the MRI exam is performed following a negative TRUS biopsy, the likelihood of clinically significant prostate cancer at the site of post biopsy hemorrhage without a corresponding suspicious finding on MRI is low. In this situation, a clinically significant cancer, if present, is likely to be in a location other than that with blood products. Thus, the detection of clinically significant cancer is not likely to be substantially compromised by post biopsy hemorrhage, and there may be no need to delay MRI after prostate biopsy if the primary purpose of the exam is to detect and characterize clinically significant cancer in the gland.
However, post biopsy changes, including hemorrhage and inflammation, may adversely affect the interpretation of prostate MRI for staging in some instances. Although these changes may persist for many months, they tend to diminish over time, and an interval of at least 6 weeks or longer between biopsy and MRI should be considered for staging.
2. Patient Preparation
At present, there is no consensus concerning all patient preparation issues. To reduce motion artifact from bowel peristalsis, the use of an antispasmodic agent (e.g. glucagon, scopolamine butylbromide, or sublingual hyoscyamine sulfate) may be beneficial in some patients. However, in many others it is not necessary, and the incremental cost and potential for adverse drug reactions should be taken into consideration.
The presence of stool in the rectum may interfere with placement of an endorectal coil (ERC). If an ERC is not used, the presence of air and/or stool in the rectum may induce artifactual distortion that can compromise DWI quality. Thus, some type of minimal preparation enema administered by the patient in the hours prior to the exam may be beneficial. However, an enema may also promote peristalsis, resulting in increased motion related artifacts in some instances.
The patient should evacuate the rectum, if possible, just prior to the MRI exam.
If an ERC is not used and the rectum contains air on the initial MR images, it may be beneficial to perform the mpMRI exam with the patient in the prone position or to decompress the rectum using suction through a small catheter. Some recommend that patients refrain from ejaculation for three days prior to the MRI exam in order to maintain maximum distention of the seminal vesicles. However, a benefit for assessment of the prostate and seminal vesicles for clinically significant cancer has not been firmly established.
3. Patient Information
The following information should be available to the radiologist at the time of MRI exam performance and interpretation:
• Recent serum prostate-specific antigen (PSA) level and PSA history
• Date and results of prostate biopsy, including number of cores, locations and Gleason scores of positive biopsies (with % core involvement when available).
• Other relevant clinical history, including digital rectal exam (DRE) findings, medications (particularly in the setting of hormones/hormone ablation), prior prostate infections, pelvic surgery, radiation therapy, and family history.
B. Technical Specifications
Prostate MRI acquisition protocols should always be tailored to specific patients, clinical questions, management options, and MRI equipment, but T2W, DWI, and DCE should be included in all exams. Unless the MRI exam is monitored and no findings suspicious for clinically significant prostate cancer are detected, at least one pulse sequence should use a field-of-view (FOV) that permits evaluation of pelvic lymph nodes to the level of the aortic bifurcation. The supervising radiologist should be cognizant that superfluous or inappropriate sequences unnecessarily increase exam time and discomfort, and this could negatively impact patient acceptance and compliance.
The technologist performing the exam and/or supervising radiologist should monitor the scan for quality control. If image quality of a pulse sequence is compromised due to patient motion or other reason, measures should be taken to rectify the problem and the sequence should be repeated.
1. Magnetic Field Strength
The fundamental advantage of 3T compared with 1.5T lies in an increased signal-to-noise ratio (SNR), which theoretically increases linearly with the static magnetic field. This may be exploited to increase spatial resolution, temporal resolution, or both. Depending on the pulse sequence and specifics of implementation, power deposition, artifacts related to susceptibility, and signal heterogeneity could increase at 3T, and techniques that mitigate these concerns may result in some increase in imaging time and/or decrease in SNR. However, current state-of-the-art 3T MRI scanners can successfully address these issues, and most members of the PI-RADS Steering Committee agree that the advantages of 3T substantially outweigh these concerns.
There are many other factors that affect image quality besides magnetic field strength, and both 1.5T and 3.0T may provide adequate and reliable diagnostic exams when acquisition parameters are optimized and appropriate contemporary technology is employed. Although prostate MRI at both 1.5 T and 3T has been well established, most members of the PI-RADS Steering Committee prefer, use, and recommend 3T for prostate MRI. 1.5T should be considered when a patient has an implanted device that has been determined to be MR conditional at 1.5T but not at 3T. 1.5T may also be preferred when patients are safe to undergo MRI at 3T but the location of an implanted device may result in artifact that could compromise image quality (e.g. bilateral hip prosthesis).
The recommendations in this document focus only on 3T and 1.5T MRI scanners since they have been the ones used for clinical validation of mpMCI. Prostate mpMRI at lower magnetic field strengths (<1.5T) is not recommended without adequate peer reviewed clinical validation.
2. Endorectal Coil (ERC)
When integrated with external (surface) phased array coils, endorectal coils (ERCs) increase SNR in the prostate at any magnetic field strength. This may be particularly valuable for high spatial resolution imaging used in cancer staging and for inherently lower SNR sequences, such as DWI and high temporal resolution DCE.
ERCs are also advantageous for larger patients where the SNR in the prostate may be compromised using only external phased array RF coils. However, use of an ERC may increase the cost and time of the examination, deform the gland, and introduce artifacts. In addition, it may be uncomfortable for patients and increase their reluctance to undergo MRI.
With some 1.5T MRI systems, especially older ones, use of an ERC is considered indispensable for achieving the type of high resolution diagnostic quality imaging needed for staging prostate cancer. At 3T without use of an ERC, image quality can be comparable with that obtained at 1.5 T with an ERC, although direct comparison of both strategies for cancer detection and/or staging is lacking. Importantly, there are many technical factors other than the use of an ERC that influence SNR (e.g. receiver bandwidth, coil design, efficiency of the RF chain), and some contemporary 1.5T scanners that employ a relatively high number of external phased array coil elements and RF channels (e.g. 16 or more) may be capable of achieving adequate SNR in many patients without an ERC.
Although credible satisfactory results have been obtained at both 1.5T and 3T without the use of an ERC, most published data were obtained with an ERC, and there is some evidence that ERC MRI may improve performance compared to non-ERC coil MRI at both 1.5T and 3T.
Taking these factors into consideration as well as the variability of MRI equipment available in clinical use, the PI-RADS Steering Committee recommends that supervising radiologists strive to optimize imaging protocols in order to obtain the best and most consistent image quality possible with the MRI scanner used. However, cost, availability, patient preference, and other considerations cannot be ignored.
If air is used to inflate the ERC balloon, it may introduce local magnetic field inhomogeneity, resulting in distortion on DWI, especially at 3T. The extent to which artifacts interfere with MRI interpretation will vary depending on specific pulse sequence implementations, but they can be diminished using correct positioning of the ERC and distention of the balloon with liquids (e.g. liquid perflurocarbon or barium suspension) that will not result in susceptibility artifacts . When liquid is used for balloon distention, all air should be carefully removed from the ERC balloon prior to placement. Solid rigid reusable ERCs that avoid the need for inflatable balloons and decrease gland distortion have been developed.
3. Computer-Aided Evaluation (CAE)
Technology Computer-aided evaluation (CAE) technology using specialized software or a dedicated workstation is not required for prostate mpMRI interpretation. However, CAE may improve workflow (display, analysis, interpretation, reporting, and communication), provide quantitative pharmacodynamic data, and enhance lesion detection and discrimination performance for some radiologists, especially those with less experience interpreting mpMRI exams. CAE can also facilitate integration of MRI data with some forms of MR targeted biopsy systems.