1 Patient preparation and positioning
1.a Patient preparation
Patient preparation consists of two parts:
i.
patient preparation before the patient arrives at the MRI department
ii.
patient preparation after the patient arrives at the MRI department
i.
Before the patient arrives at the MRI department:
- Routine medication should not be stopped.
- The patient should not eat or drink any solid food for at least 4-6 hours in an attempt to reduce motion artifacts from bowel peristalsis.
- Non-sparkling water is permissible.
ii.
After the patient arrives at the MRI department:
- Make sure the patient completed the safety questionnaire both for MRI safety as well as antiperistaltic agent (where applicable) and no contraindications were noted.
- Explain the procedure to the patient.
- Offer appropriate gowns to the patient.
In the event that an endorectal coil (ERC) is to be used,
the patient should be instructed to remove underwear as well.
- Ask the patient to remove anything containing metal (e.g piercings,
metal dentures,
personal effects e.t.c).
- Have an intravenous cannula placed (18G or 20G IV cannula is preferred).
- Offer earplugs and headphones to the patient.
- A full bladder should be avoided because it may cause discomfort to the patient (fig.
3).
Mid-filled bladder is preferred.
- If possible,
the patient should evacuate the rectum just prior to the examination in order to eliminate the presence of air.
The presence of air produces susceptibility artifacts and distortions,
which mainly affect the DWI acquisition (because it is based on the EPI technique).
Furthermore,
many papers suggest the administration of enema preparation some hours before the mpMRI examination.
However,
an enema may also produce bowel peristalsis,
resulting in increased motion-related artifacts.
- An antiperistaltic agent (e.g. hyoscine butylbromide) can be used to further reduce the motion artifacts from bowel peristalsis.
Antiperistaltic agents should be administered before the T2-w sequences,
because these acquisitions are very susceptible to motion (fig.
4).
However,
incremental cost and potential for adverse drug reactions should be taken into consideration.
Fig. 3: Sagittal T2-w image of the prostate gland. Figure shows why a full bladder should be avoided. A full bladder may cause discomfort to the patient, resulting in motion artifacts and blurry images. Mid-filled bladder is preferred.
Fig. 4: Sagittal T2-w images of the prostate gland with and without the use of antiperistaltic agents (hyoscine butylbromide). Antiperistaltic agents can be used to reduce motion artifacts from bowel peristalsis, offering "clearer" images. However, in many patients it is not necessary and the incremental cost and potential for adverse drug reactions should be taken into consideration.
1.b Patient and coil positioning
The patient needs to be as comfortable as possible to reduce any motion artifact that may arise due to patient discomfort.
Coil positioning needs to be as accurate as possible in order to enable maximum usage of the available signal.
The aforementioned is achievable through the following steps:
- The patient should be placed in the supine position.
- Prone position is an alternative when air is presented on localizer scans in the rectum,
and on claustrophobic patients.
- Feet first position is preferred in order to reduce claustrophobic reactions.
- Cushion the head.
- Coil positioning is very important for a successful examination.
When a surface coil is used,
the prostate gland should be on the center of the coil (fig.
5).
When an endorectal coil is used,
ERC is inserted a short distance into the rectum and an inflatable balloon helps to maintain the appropriate positioning,
which should provide the maximum SNR from the prostate gland.
- Center the laser beam localizer over the prostate (the prostate lies almost directly behind the pubic symphysis).
Fig. 5: Coil positioning is very important for a successful examination. Improper coil positioning results in decreased SNR that affects the overall image quality. When a surface coil is used, the prostate gland should be on the center of the coil as shown by this figure.
2 Imaging protocol and slice positioning
2.a Prostate mpMRI protocol
Figure 6 shows an example of a completed mpMRI protocol of the prostate gland.
Fig. 6: Multiparametric MRI protocol of the prostate gland. The MRI protocol consists of orthogonal (sagittal, coronal, axial), high resolution two-dimensional (2D) T2-weighted fast or turbo spin echo (FSE or TSE) acquisitions, low, high and very high b-value diffusion-weighted imaging (DWI) with corresponding apparent diffusion coefficient (ADC) map, pre- and post-contrast fat-suppressed three-dimensional (3D) T1-weighted gradient echo (GRE) acquisitions and T1-w Perfusion imaging (DCE PWI). A large FOV T1- or T2-weighted acquisition through the entire pelvis is suggested in order to investigate nodal and/or skeletal metastases. Finally, 3D T2-weighted acquisition with ≤1.5 mm slice thickness may be used as an adjunct to multiplanar 2D T2-weighted FSE acquisitions. However, soft-tissue contrast and in-plane resolution of 3D acquisitions are inferior to 2D acquisitions.
2.b Slice positioning
- Sagittal: Slices should be parallel to the prostatic urethra in the coronal and axial planes. Cover the entire prostate gland and seminal vesicles.
- Coronal: Slices should be perpendicular to the prostatic urethra in the axial plane and parallel to the line joining the apex and base of the prostate gland in the sagittal plane.
This is typically between 5-25 degrees.
Cover the entire prostate gland and seminal vesicles.
- Axial: Slices should be perpendicular to the prostatic urethra in the coronal plane and perpendicular to the line joining the apex and base of the prostate gland in the sagittal plane.
Cover the entire prostate gland and seminal vesicles.
Slice positioning is shown by figure 7.
Fig. 7: Figure 7 demonstrates the slice positioning for the mpMRI of the prostate gland.
3 Technical Considerations
3.a Field strength
Prostate MRI should be performed on superconducting magnets with a field strength of either 1.5 Tesla or 3 Tesla.
The main advantage of 3T is the increased SNR.
This can be used to facilitate higher image quality through improved spatial and temporal resolution,
and/or reduced acquisition times.
For this reason,
3T systems are preferred.
An additional item to consider is patient safety and artifact generation.
Some medical implants may be incompatible at 3T for safety reasons.
In some cases, safety is not an issue but the implant may generate sufficient artifact to obscure or degrade the image.
In these cases,
MR Imaging should be performed at 1.5T.
3.b Coil selection
Surface/external phased array coils (torso or cardiac) and/or endorectal coil (ERC) can be used for the multiparametric MR Imaging of the prostate gland.
Although many papers suggest the use of ERC at 1.5T scanners,
high quality imaging can be obtained at both 1.5T and 3T without the use of an ERC.
The use of an ERC deforms the shape of the gland,
cannot be used for whole pelvis examinations,
and is usually uncomfortable for patients.
Additionally,
the cost of the ERC,
as well as the supplies and added time involved in the procedure,
make the use of this method less practical.
These are drawbacks that need consideration from the MR staff prior to its utilization.
In conclusion,
the use of an endorectal coil is not necessary,
however an ERC can be beneficial in some specific situations:
- With older 1.5T MRI scanners,
the use of an ERC is considered indispensable for achieving the type of high resolution imaging needed for prostate mpMRI.
- ERC can also be advantageous for larger patients where the SNR in the prostate may be compromised using only surface phased array coils.
- When integrated with surface phased array coils,
ERC increases the SNR in the prostate at any magnetic field strength.
This can be valuable for inherently lower SNR sequences,
such as DWI,
MR Spectroscopy,
and high temporal resolution DCE Perfusion.
3.c Imaging Parameters
i) T2-w imaging
- Planes: sagittal,
coronal,
axial
- Slice thickness and spacing: ≤3 mm with no gap between the slices
- FOV: 16-22 cm,
centered to the prostate gland
- Spatial resolution (pixel size): ≤0.7 mm (phase) and ≤0.4 mm (frequency),
not interpolated.
- Phase-encoding direction: right to left on axial and coronal sequences and anterior to posterior on sagittal sequences.
Utilizing these phase encoding directions will aid in reducing motion artifacts from breathing and flow.
Motion artifacts can degrade the image quality and may obscure pathology (fig.
8).
- Parallel imaging (PI): PI technique can be used to reduce motion and pulsation artifacts.
An acceleration factor of 2 is most common; this strikes a balance between scan time reduction and acceptable SNR.
- Saturation bands: anterior sat bands should be used on the sagittal T2-w sequence to minimize motion artifacts from breathing.
- Signal averages (NEX/NSA/ΝΑQ): high resolution T2-w imaging requires multiple signal averages.
Typically ≥3 signal averages should be used.
- Receiver Bandwidth (rBW): high rBW should be used in order to minimize the echo spacing,
which results in reduced image blurring (fig.
9).
Typically,
an rBW of at least 27.7kHz should be selected.
- Echo Spacing: as low as possible
- TR: ≥4000 msec
- TE: 80-120 msec
- ETL/Turbo factor: ≥16
Fig. 8: Figure 8 shows the effect of phase direction selection on axial T2-w imaging. Phase encoding direction should be right to left (R/L) in order to eliminate motion artifacts from breathing, pulsation and bowel peristalsis. Motion artifacts can degrade the image quality and may obscure pathology.
Fig. 9: Figure 9 shows the effect of receiver bandwidth (rBW) on T2-w imaging. Low rBW results in increased echo spacing and image blurring (upper row). High rBW (at least 27.7kHz) can provide "clear", high quality images (lower row).
ii) Diffusion-weighted imaging (DWI)
- Plane(s): Axial.
Sagittal plane can be used on patients with hip prosthesis in order to minimize image distortions and susceptibility artifacts.
- b-values: Diffusion-weighted acquisition should include low (50-100 s/mm2),
high (800-1000 s/mm2) and very high (≥1400 s/mm2) b-values with corresponding apparent diffusion coefficient (ADC) map.
Non-zero b-value helps to suppress the signal in vessels.
A normal prostate gland presents diffusion restriction and very high b-value images can show "suspicious" areas that preserve high signal intensity (fig.
10).
Very high b-value images can be either obtained directly or synthesized by lower b values.
If the SNR is adequate,
the first option (obtained) is the preferred one. However,
the very high b-value images should be excluded from the ADC map calculation; their intrinsic low SNR affects the quality of the parametric map.
Information about perfusion effects can be acquired by obtaining multiple low b-value data,
ranging from 0 to 300 sec/mm2 (bi-exponential/IVIM model).
- Diffusion directions: Diffusion-sensitizing gradients in 3 orthogonal directions should be used (trace or isotropic DWI).
Information about the non-Gaussian movement of water molecules (diffusion kurtosis imaging) can be obtained using ≥15 diffusion directions.
- Fat suppression: Fat saturation is necessary to eliminate chemical shift artifacts.
Spectral or hybrid fat suppression techniques are preferred.
STIR technique is the recommended choice when imaging patients with metallic prosthesis.
- TR: ≥4000 ms
- TE: as low as possible to reduce image distortions and improve SNR
- Slice thickness and spacing: ≤4 mm (ideally ≤3 mm) with no gap between the slices.
- FOV: 16-32 cm (ideally 16-22 cm),
centered to the prostate gland.
FOV may be larger than T2-w sequence in order to gain SNR.
- Spatial resolution (pixel size): ≤2.5 mm in phase and frequency encoding directions,
not interpolated.
- Phase-encoding direction: anterior to posterior in order to reduce image distortions.
Distortions can degrade the image quality and may obscure pathology.
- Parallel imaging: the use of PI technique is necessary to further reduce susceptibility artifacts,
image distortion,
and shot time.
- Sat bands: anterior sat bands can be used to minimize the motion artifacts from breathing.
Sat bands can also minimize wrap-around artifacts on patients with a large body habitus.
- Signal averages (NEX/NSA/NAQ): Multiple signal averages should be used at high and very high b-values in order to gain SNR and maintain an adequate image quality.
- Receiver Bandwidth (rBW): very high rBW should be used in order to minimize the echo spacing,
which results in reduced image distortions and susceptibility artifacts.
Typically,
an rBW of at least 62.5kHz should be selected.
Fig. 10: Figure 10 illustrates the effect of very high b-value DW Imaging on the prostate gland. Very high b-value images can only show the “suspicious” areas that preserve high signal intensity, while the signal on normal prostate is nulled.
iii) Dynamic contrast-enhanced perfusion-weighted imaging (DCE PWI)
- Sequence: Two- or three-dimensional T1-weighted gradient echo (2D or 3D T1-w GRE) pulse sequence can be used.
However,
3D T1-w GRE technique is preferred.
- Plane(s):Axial.
Sagittal plane can be used on patients with hip prosthesis in order to minimize image distortions and susceptibility artifacts.
In addition,
when searching for post-prostatectomy recurrence,
imaging in sagittal plane may be more effective than imaging in axial plane because anatomical landmarks have moved and contrast uptake is often linear in front of the urethra at the apex.
- Slice thickness and spacing: ≤3 mm with no gap between the slices
- FOV: 16-22 cm,
centered to the prostate gland
- Spatial resolution: ≤2 mm in phase and frequency encoding directions,
not interpolated.
- Temporal resolution: ≤15 sec.
A temporal resolution of ≤6 sec is suggested when quantitative assessment is required.
- Parallel imaging: the use of parallel imaging technique is necessary to reduce the total acquisition time and to increase the temporal resolution.
- Fat suppression: Fat suppression (or subtraction) is recommended when qualitative (visual) assessment is required.
Fat suppression is not necessary when quantitative assessment is needed.
- Flip Angle (FA): 15-20°
- TR/TE: minimum
- Total scan time: ≥2 min (ideally 4-5 min)
- Flow rate: 2-3 ml/sec
- Signal averages (NEX/NSA/NAQ): 1 or Partial Fourier in order to increase the temporal resolution of PWI acquisition.
- Receiver Bandwidth (rBW): very high rBW should be used in order to minimize the echo spacing,
which results in reduced image distortions and susceptibility artifacts.
Typically,
an rBW of at least 62.5kHz should be selected.
- Sat bands: Sat bands should not be used in an effort to decrease the temporal resolution and the total acquisition time.
- Data analysis: DCE images can be evaluated qualitatively,
semi-quantitatively and/or quantitatively.
Quantitative analysis provides information about ktrans,
plasma flow, and extravascular/extracellular volume,
assisting in the differentiation and accurate characterization of pathology.
Figure 11 shows an example of qualitative (visual) assessment,
which is the most used DCE PWI data analysis worldwide.
Fig. 11: Figure 11 is an example of qualitative (visual) assessment. This illustrates a lesion in the transition zone with focal, early enhancement (positive DCE). In addition, note that the “signal intensity – time” graph shows that the signal intensity decreases after its highest point after its initial rise (type III enhancement kinetic curve/washout pattern).