Bladder Diverticulum. Axial-plane image shows a urine-filled diverticulum (D) projecting through a defect (arrow) in the bladder wall. B, bladder.

http://www.msdlatinamerica.com/ebooks/FundamentalsofDiagnosticRadiology/sid454050.html

Enlarged Prostate. Midline sagittal US image shows an enlarged prostate (P) protruding into and elevating the base of the bladder (B). The urethral orifice (long arrow) forms a V-shaped depression in the prostate. The seminal vesicles (open arrow) are also seen. The bladder wall is mildly thickened (between short arrows).

http://www.msdlatinamerica.com/ebooks/FundamentalsofDiagnosticRadiology/sid454050.html

Prostate Abscess. Transverse transrectal US reveals an abscess (arrows) in the right side of the prostate gland in a patient with fever, pelvic pain, and pyuria. The abscess contained purulent debris, which was seen on US as floating particulate matter.

http://www.msdlatinamerica.com/ebooks/FundamentalsofDiagnosticRadiology/sid454050.html

Benign Prostatic Hypertrophy. Transrectal axial US view through the midprostate demonstrates excellent differentiation of a normal peripheral zone (pz, solid arrows). The inner gland (IG) demonstrates mild enlargement and heterogeneity that is characteristic of benign prostatic hypertrophy. A small prostatic cyst is evident (open arrow). The hypoechoic fibromuscular zone (FM) is anterior. A, anterior; P, posterior.

http://www.msdlatinamerica.com/ebooks/FundamentalsofDiagnosticRadiology/sid454050.html

Prostate Cancer. Right sagittal transrectal US image shows a distinct hypoechoic nodule (arrows) in the peripheral zone. US-guided transrectal needle biopsy confirmed prostate carcinoma. sv, seminal vesicle.

http://www.msdlatinamerica.com/ebooks/FundamentalsofDiagnosticRadiology/sid454050.html

Normal Seminal Vesicles. Transrectal US image shows the normal hypoechoic convoluted tubular appearance of the seminal vesicle (arrow). B, bladder.

http://www.msdlatinamerica.com/ebooks/FundamentalsofDiagnosticRadiology/sid454050.html

false negative rate of transrectal ultrasound biopsy

Cancer: An Indian Experience
Gupta N ; Ansari M ; Dass S
Indian Journal of Cancer; issue 3; pp: 151-154; Vol: 42; Year: 2005

Summary

BACKGROUND: With the advent of prostate specific antigen the number of patients undergoing prostate biopsy has dramatically increased. The sextant biopsy technique has been conventionally used for the diagnosis of prostate cancer. Recently, concern has arisen that the original sextant method may not include an adequate sample of the prostate, hence it may result in high false negative rates. We conducted a prospective study to determine whether the 5-region prostate biopsy technique significantly increases the chance of prostate cancer detection as compared to the sextant biopsy technique. AIMS: To evaluate the efficacy of TRUS guided sextant and 5-region biopsy techniques in detecting carcinoma prostate in patients with PSA between 4 and 10 ng/ml and normal digital rectal examination.

METHODS AND MATERIAL: Between December 2001 and August 2003 one forty-two men, aged 49-82 years, who presented with LUTS, normal digital rectal examination (DRE) and PSA between 4 and 10 ng/ml underwent TRUS guided sextant prostate biopsy. Serum PSA was reassessed after 3 months in patients whose biopsies were negative for cancer. If PSA was still raised, the patients underwent extensive 5-region biopsy.

RESULTS: Mean patient age was 64 years and median PSA was 6.9 ng/ml. TRUS guided sextant biopsy revealed adenocarcinoma prostate in 34 men (24%). Median Gleason score was 7. Seven men (4.9%) had cellular atypia and 3(2.1%) had prostatic intraepithelial neoplasia (high grade). On repeat PSA estimation after 3 months, 48 patients showed stagnant or rising trend for which they underwent TRUS guided 13-core biopsy. Five (10.4%) patients were detected to have adenocarcinoma on repeat biopsy. Biopsy negative patients are on regular follow up with yearly PSA estimation. Complications included transient mild haematuria in14 patients (9.82%) and haematospermia in 4 (2.8%). Urinary retention developed in one patient and required an indwelling catheter for 4 days.

CONCLUSION: Transrectal ultrasound guided sextant biopsy has shown a false negative rate of approximately 11%. A repeat 5- region (13-core) biopsy strategy can decrease the false negative rate of conventional sextant biopsy in patients with previously negative biopsies but persistently high PSA levels, high grade PIN or cellular atypia.

MRI OF PROSTATE

INTRODUCTION
Prostate adenocarcinoma (PCa) is the second tumor in incidence and mortality among the malignant neoplasms in men, surpassed only by non-melanoma skin cancer in incidence, and by lung cancer in number of deaths(1,2).
The methods employed in the screening for PCa in the population are prostate-specific antigen (PSA) test and digital rectal examination. Transrectal ultrasonographyguided biopsy is the method of choice for histological diagnosis of the disease.
An accurate staging of the disease is of paramount importance for an appropriate therapeutic planning. The nomograms introduced by Partin combine clinical staging based on digital rectal examination, PSA levels and the Gleason score determined by biopsy (the highest the Gleason score, the highest is the tumor aggressiveness), with a relatively high accuracy in local staging of PCa(3).
Several studies have demonstrated that magnetic resonance imaging (MRI), with its high anatomical resolution, adds relevant information to the clinical nomograms in the local staging of PCa. Endorectal coil MRI presents an accuracy of 85% in the prediction of extracapsular extension and up to 97% for seminal vesicles invasion(4,5).
New prospects include the 3 tesla MRI, that may be utilized with endorectal coil (possibly increasing even further the sensitivity and specificity in the local staging of PCa), or without the endorectal coil with consequential reduction in discomfort for the patient(6,7).
The present study is aimed at presenting the most relevant aspects of the evaluation of the prostate by means of MRI, particularly those related to the detection and staging of PCa.

ANATOMY
Zonal anatomy of the prostate
The prostate is an inverted cone-shaped gland with the base adjacent to the vesical floor and the apex postero-inferior to the pubic symphysis. It is divided into four glandular zones: peripheral, transitional, central and periurethral(8). The transitional, central and periurethral zones, hardly differentiated by imaging methods, are jointly denominated central gland. The limit between the peripheral zone and the central gland is called "surgical capsule", and the discontinuous fibromuscular layer covering the gland is the "prostatic capsule" (Figure 1). For a better localization of lesions and biopsy guidance, the prostate is conventionally divided into three regions by imaginary cross sectional lines (base, middle and apex), and in two sides by an imaginary longitudinal median line (left and right), thus configuring the prostatic sextants.

Seminal vesicles
Normal seminal vesicles present a typical ductal pattern and hyperintense signal on MRI T2-weighted sequences.

DESIGN AND OPTIMIZATION OF THE PROTOCOL
General recommendations
A four-hour fasting and rectal preparation is recommended for greater comfort of the patient. Shortly before the examination 1 mL of N-butylescopolamine (20 mg/mL) shall be administered to reduce intestinal peristalsis.
Study of the pelvis
The images shall be obtained from the aortic bifurcation to the lower limit of the pubic symphysis, utilizing the torso or pelvic phased-array coil.
Suggested protocol - Axial T2-weighted fast spin echo (FSE) with fat saturation, and axial T1-weighted gradient echo (GRE), 7 mm slice thickness, 1-2 mm interslice gap, 256 × 192 matrix, number of excitations (NEX) = 3 for T2-weighted sequences, NEX = 1 for T1-weighted sequences, fieldof-view (FOV) of approximately 30 cm.
Study of the prostate
Initially, the patients shall be given an explanation about the procedure. After that, a digital rectal examination is performed to evaluate the prostate size to guide an appropriate positioning of the coil. The coil (which is disposable) is covered by a condom and lubricated with local anesthetic gel (xylocaine). After the coil is in place, it is filled with 40 to 80 ml of air or perfluorocarbon, with the objective of keeping the coil in place and reducing sphincter contraction artifacts. Perfluorocarbon is preferred in examinations with spectroscopy, and its advantages will be further detailed.
The protocol for specific study of the prostate and seminal vesicles after endorectal coil insertion must comprise high-resolution FSE T2-weighted sequences in the axial, coronal and sagittal planes, from the bottom of the seminal vesicles to the prostate apex (Figure 2).
Suggested protocol - The sequences may be performed with 3-4 mm thick slices, 0-1 mm interslice gap, FOV from 12 to 16 cm, 256 × 256 matrix, NEX = 3 or 4 and TE = 120-160.

TUMOR STAGING
Pathological staging
The systems accepted for PCa staging are TNM and Jewett-Whitmore, TNM being the most widely utilized. The TNM classification describes the primary tumor extent (T), presence or absence of locoregional lymph node involvement (N) and presence or absence of metastases (M)(9).
The appropriate preoperative PCa staging presents therapeutic and prognostic implications. Specifically, the detection of extracapsular extension and seminal vesicles invasion are of great importance, as they allow the differentiation between T2 and T3 stages(10).
Local staging - MRI
Tumor identification - PCa is most frequently found in the peripheral zone (65% to 74%). Usually the lesion presents like a poorly defined nodule with hypointense signal intensity on T2-weighted sequences, in contrast with signal hyperintensity in the normal peripheral zone (Figure 3).

Detection of extracapsular extension - The detection of extracapsular extension can be inferred from both specific and non-specific signs at MRI(11) .
Specific signs of extracapsular extension (Figure 4): solid tissue in the periprostatic fat; irregular bulging of the prostatic capsule; obliteration of the rectoprostatic angle.
Non-specific signs of extracapsular extension: regular capsular thickening; capsular discontinuity; regular bulging of the prostatic capsule.
Invasion of seminal vesicles - Invasion of seminal vesicles may occur by direct extension of the tumor localized in the base of the prostate gland, through the ejaculatory duct, or by hematogeneous spreading (Figure 5).
Diagnostic findings of seminal vesicle invasion are: hypointense signal on T2-weighted sequences and loss of the typical ductal pattern(12).
Regional staging
It is also important to evaluate the sequences of the pelvis and of the prostate in the search for regional lymphadenopathies (> 1 cm in the smallest diameter), invasion of periprostatic structures and bone lesions (Figure 6).

SPECTROSCOPY
General principles
Proton spectroscopy allows a noninvasive evaluation of anatomic and biological characteristics of the tumor, emphasizing the detection, localization and staging of PCa. Spectroscopy utilizes a powerful magnetic field and radiofrequency waves to obtain metabolic data, based on the relative concentration of endogenous prostatic metabolites. It is always performed in association with endorectal coil MRI, increasing total examination time by 10 to 17 minutes.
Voxels of 0.3 cm3 are programmed in the whole prostate extent, utilizing T2-weighted images as a reference, which allows the joint analysis of anatomical and metabolic alterations. The metabolites analyzed are: choline, creatine, citrate and, more recently, polyamine. The analysis is performed by means of amplitude × frequency graphs, the frequencies being specific for each metabolite. The areas under the peaks correlate with the concentration of each metabolite in the prostatic tissue(13,14) (Figure 7).

Citrate - peak: 2.6 ppm. Metabolite synthesized by a healthy prostate, most remarkably in the peripheral zone. Hyper-plastic nodules may present citrate levels as high as those observed in the peripheral zone. In the presence of cancer, citrate levels are reduced or undetectable.
Choline - peak: 3.22 ppm. Cell metabolism marker. Increased choline levels are observed in malignant tissues, because of high phospholipid metabolism of the cell membrane.
Creatine - peak: 3.03 ppm. Energy reserve with constant concentration. Because of the proximity between creatine and choline peaks, they may be inseparable on the spectral line; therefore, the (choline + creatine)/citrate ratio is utilized in the spectral analysis. (Choline + creatine)/citrate ratios > 0.5 are suggestive of malignancy (the higher the ratio, the higher is the possibility of cancer).
Polyamine - peak: 3.1 ppm. This peak resolution is better observed when the endorectal coil is filled with perfluorocarbon. It is reduced in suspicious areas, being identified as an acute decreasing curve between the choline and creatine peaks.
Perfluorocarbon, a fluorine and carbon compound without hydrogen atoms, can be utilized to inflate the endorectal coil, reducing the magnetic susceptibility difference between the air and the prostatic tissue. It improves the characterization of peaks of low spectral signals in the peripheral zone, as well as the peak resolution of each metabolite(15).
Jung et al.(16) described a score system in an attempt to standardize and systematize the prostatic metabolic evaluation. This five-score system utilizes progressively more suspicious spectral patterns, with good accuracy (74.2% to 85.0%) and excellent interobserver agreement, using a score > 4 for tumor detection. Additional criteria, such as polyamine peak inversion, should be considered.
Role of proton spectroscopy
To increase the MRI specificity - Lesions with hypointense signal on T2-weighted sequences may be observed in other non-tumor conditions such as hemorrhage and chronic prostatitis, among others. The addition of spectroscopy to endorectal coil MRI increases specificity and reduces interobserver variability(17).
To improve local staging (number of tumor voxels × risk for extracapsular extension) - Studies demonstrate that the higher the number of tumor voxels at spectroscopy, the higher is the risk for extra-capsular extension. Patients with small tumors at MRI spectroscopy (0-1 tumor voxel/slice) presented a risk of 6% for extracapsular extension, while those with extensive tumors (> 4 tumor voxels/slice) presented a risk of 80%(18) (Figure 8).

To predict tumor aggressiveness - MRI spectroscopy has the potential to noninvasively infer tumor aggressiveness. Tumor changes vary from a subtle increase in choline levels and moderate citrate levels in low grade tumors, to a sharp rise in choline levels and reduction/absence of citrate in high grade tumors(19).
To predict tumor volume - Three-dimensional spectroscopy in combination with MRI increases global accuracy of tumor volume estimates. The prediction of tumor volume by MRI spectroscopy in lesions > 0.5 cm3 obtained a significant correlation with histopathologic tumor volume(20).

PERFUSION STUDY (DYNAMIC CONTRAST-ENHANCED MRI)
Dynamic contrast-enhanced MRI is useful in the detection of areas under suspicion for neoplastic involvement. It is performed with 3D T1-weighted sequences, before and after intravenous gadolinium injection (0.1 mmol of gadopentate dimeglumine/kg of weight), administered by injection pump at least 2.5 mL/s, followed by 15 mL saline solution. Initially, a non-contrast enhanced series is obtained ("mask"), followed by repeated high temporal resolution, contrast-enhanced series (<>Figure 9).

DIFFUSION-WEIGHTED MRI
Diffusion-weighted MRI may be utilized to increase MRI sensitivity and specificity in the evaluation of prostate tumors, considering that neoplasms usually restrict the diffusion of water molecules. Sequences acquisition should be performed with high b values (500 to 1000) and always with analysis of images at the apparent diffusion coefficient map for tumor detection. Recent studies have shown that diffusion adds accuracy to MRI in tumors detection, particularly when combined with spectroscopy(26,27).

CENTRAL GLAND
Approximately 65% to 74% of tumor nodules originate in the prostatic peripheral zone. Although frequently tumors originated in the central gland present extension to the peripheral zone at the time of the diagnosis, this region may hide PCa in more than 25% of the cases, as demonstrated in specimens from radical prostate-ctomy(27). Therefore a significant percentage of all PCa's may not be diagnosed in case the radiologist focuses on the peripheral zone only. Prostate cancer detection in the central gland is difficult, as this region presents a high incidence of benign hyper-plastic nodules, whose signal intensity is heterogeneous, and sometimes similar to the one in cases of cancer.
One may suspect of tumors in the central gland by the presence of a homogeneously hypointense area on T2-weighted images, with poorly defined or spiculate margins, lenticular shape conditioning the poor definition of the surgical capsule, invasion of the urethra or anterior fibromuscular stroma, (choline + creatine)/citrate ratio > 0.7 at spectroscopy, and/or early enhancement peak and prominent clearance at the perfusion study (Figure 10). In contrast, benign prostatic hyperplasic nodules generally present well defined contours, hypointense halo and signal intensity heterogeneity on T2-weighted images(28,29).

TUMOR DETECTION
Patients with high PSA levels and negative biopsy
Approximately 66% of patients with PSA levels > 4 ng/ml present negative prostate biopsies, many of them being submitted to rebiopsy with a higher number of fragments. MRI spectroscopy and/or per-fusion studies have demonstrated good results in the identification of suspicious foci in patients with clinical-laboratory suspicion of PCa and at least one previous negative biopsy, targeting ultrasound-guided rebiopsy with additional samples of those areas (Figure 11). Accuracy between 80% and 90% has been demonstrated for spectroscopy and perfusion in the detection of tumors > 0.5 cm3(30,31).
Postoperative tumor recurrence
MRI may demonstrate post prostatectomy local recurrence and/or residual tumor, which are seen as soft tissue masses with subtle hyperintense signal on T2-weighted images as compared with the adjacent musculature (Figure 12). The main differentiation to be considered is with postoperative fibrosis, which shows marked hypointensity on T2-weighted images. The main locations for tumor recurrence include: retrovesical (40%), perianastomotic (29%) and in retained seminal vesicles (22%)(32).

Post-radiotherapy tumor recurrence
Approximately 25% of patients with prostate cancer are treated with radiotherapy, with the rate of PSA recurrence after 5 years ranging from 15% to 67%, depending on each patient's risk.
After radiotherapy, a diffuse reduction in the signal intensity on T2-weighted sequences is observed, as a consequence of fibrosis and glandular atrophy. The following imaging criteria are adopted in case tumor recurrence is suspected(33):
- MRI criterion: nodular area with marked hypointense signal on T2-weighted sequences.
- Spectroscopy criterion: voxels with a (choline + creatine)/citrate ratio > 0.5; or any voxel with increased choline levels, when there is too much noise limiting the appropriate individualization of the other metabolites.

DIFFERENTIAL DIAGNOSIS
Other conditions with hyposignal intensity in the peripheral zone on T2-weighted images may mimic prostate cancer, such as hemorrhage, prostatitis, benign prostatic hyperplasia, trauma, other tumors such as lymphoma and sarcoma, among other causes.
Post-biopsy hemorrhage is characterized by hypointense signal on T1-weighted images and variable signal intensity on T2-weighted images. Hemorrhagic foci present a significant decrease 21 days after the procedure, so that this minimum interval should be utilized between biopsy and MRI.