Preoperative Multiparametric Prostate Magnetic Resonance Imaging Structured Report Informs Risk for Positive Apical Surgical Margins During Radical Prostatectomy : Journal of Computer Assisted Tomography

Secondary Logo

Journal Logo

Abdominopelvic Imaging: Genitourinary

Preoperative Multiparametric Prostate Magnetic Resonance Imaging Structured Report Informs Risk for Positive Apical Surgical Margins During Radical Prostatectomy

Costa, Daniel N. MD; Meng, Xiaosong MD, PhD; Tverye, Aaron BS; Bagrodia, Aditya MD; Recchimuzzi, Debora Z. MD; Xi, Yin PhD; Arraj, Patrick BS; Shah, Rajal B. MD; Subramanian, Naveen MD; de Leon, Alberto Diaz MD; Roehrborn, Claus G. MD; Rofsky, Neil M. MD; Chen, Heng BS§; Pedrosa, Ivan MD, PhD∗,†

Author Information
Journal of Computer Assisted Tomography 47(1):p 38-44, 1/2 2023. | DOI: 10.1097/RCT.0000000000001377
  • Free


Prostate cancer (PCa) is the most common noncutaneous cancer and the second most common cause of death from cancer among men.1 In patients with localized PCa, radical prostatectomy (RP) remains a first-line treatment option.2

The presence of tumor along the resection surgical margin (positive surgical margins [PSMs]) is an adverse surgical outcome3 associated with increased risk of recurrence4 and need for further treatment.5 One of the concerns after RP is the recovery of urinary continence. Therefore, efforts are made to avoid unnecessarily radical surgery and preserve as much membranous urethral length as possible, which in turn may affect the degree of resection near the apex, hence explaining the higher risk of margin positivity in apical lesions. The goal of surgery, therefore, is to completely resect the prostate and its primary tumor (ie, negative surgical margins) while decreasing morbidity by preserving the integrity of adjacent structures (eg, neurovascular bundles, bladder neck, and membranous urethra).

Tumor location is one of several factors predictive of PSM.6 Excision of low apical prostate lesions can be particularly challenging based on location near the expected site of the vesicourethral anastomosis, as well as specific anatomical characteristics that make dissection more difficult,7 including the broad extent of the dorsal venous complex, proximity to the neurovascular bundles, the pear-shaped appearance and tapering/narrowing of the prostate toward the apex, and an incompletely developed pseudocapsule.8 Indeed, the prostatic apex is the most frequent location of PSM after surgery representing up to 29% to 39% of all PSM cases.9 Surgical attempts at relatively wider excision margins risk postoperative urinary or sexual dysfunction, whereas a more conservative approach may jeopardize complete removal of tumor. Multiparametric magnetic resonance imaging (mpMRI) has been extensively used for PCa local staging.10,11 More recently, the utility of mpMRI to guide the use of nerve-sparing (NS) technique was reported although without specific assessment of the apical margin.12 Effective interdisciplinary communication of imaging findings is vital for patient care, as referring physicians depend on the contained information for the decision making and subsequent treatment. Traditional radiology reports contain nonstructured free text in narrative language, which can hamper the information transfer and diminish the clarity of the report.13 We hypothesize that mpMRI may also able to identify patients at increased risk for apical PSM (aPSM) compared with other apical and nonapical lesions, and this information can be conveyed in a structured report to urologists to facilitate patient counseling and treatment decisions.

Data regarding the ability of mpMRI to prospectively identify men at risk for aPSM using a structured report are lacking. Thus, the purpose of this study is to determine and compare the rate of aPSM in men with versus without prospectively flagged at risk prostate lesions during clinical mpMRI interpretation using whole-mount histopathology as the reference standard.


Study Design and Eligibility Criteria

This institutional review board–approved and Health Insurance Portability and Accountability Act–compliant, single-center, retrospective case-control study of prospectively collected data included all treatment-naive men with abnormal 3T mpMRI (PI-RADS v2 score ≥3) between January 2016 and December 2018 who underwent subsequent RP. Patients with more than 6 months elapsed between MRI and RP, and incomplete or nondiagnostic MRI (eg, severe motion) were excluded (Fig. 1). The requirement for informed consent was waived. No overlap existed between subjects enrolled in this study and prior publications.

Eligibility criteria. n, number of men.

Magnetic Resonance Imaging Acquisition and Interpretation

All MRI studies were performed in a 3 Tesla MRI scanner (Philips Healthcare, Best, the Netherlands) with an endorectal and a phased-array surface coil, including T2-weighted, diffusion-weighted, and dynamic contrast-enhanced images. The mpMRI acquisition protocol has been previously published.14 During routine clinical interpretation, each reader (1 of 9 board-certified radiologists, with more than 5 years of experience interpreting multiparametric MRIs of the prostate, each reading an average of more than 100 prostate MRIs annually) independently and prospectively assigned a lesion-specific PI-RADS score,15 size, location and, perceived likelihood of extraprostatic extension using a 1- to 5-point Likert scale.16 Only the apical lesions abutting the apical most aspect of the prostate (ie, no normal-appearing prostate noted distal to the MRI-visible lesion) and/or encircling the distal most prostatic urethra (Fig. 2) were prospectively flagged using standardized language available as a “pick list” option in the “Impression” section of a structured report in PowerScribe360 (Nuance, Burlington, Mass). The statement reads as follows: “This lesion extends to the apical most aspect of the prostate near the expected location of the vesicourethral anastomosis.” The intended purpose of this statement was communicated to urologists via electronic messages and during a urology department case conference. Radiologists were aware of clinical data available at the time of interpretation (eg, prostate-specific antigen [PSA], previous biopsy results). Indications for mpMRI included men considered at clinical risk for PCa (eg, elevated or rising PSA, abnormal digital rectal examination) with or without previous negative biopsy, before entering or during active surveillance prior to confirmatory biopsy, or preoperative staging of newly diagnosed cancer.

Example of lesion identified as at risk for PSMs based on its location in the apical most aspect of the prostate. A 67-year-old biopsy-naïve man with PSA of 7.6 ng/mL referred for prebiopsy multiparametric MRI of the prostate that revealed a 10-mm PI-RADS 4 lesion (arrows) encircling the distal most prostatic urethra (arrowhead) in the anterior and lateral apex shown by the axial T2-weighted (A), b = 2,000 diffusion-weighted (B) and early postcontrast dynamic contrast-enhanced (C) images. Coronal (D) T2-weighted image helps further delineate the apical most location of the lesion and its broad contact with the glandular margin (dashed line) in this region. Targeted biopsy revealed GG 2 adenocarcinoma. Whole-mount RP histopathological analysis confirmed a pT2 GG 2 adenocarcinoma centered at the right anterior apex with a positive margin measuring 6 mm in circumferential length.

Surgical Technique

Radical prostatectomies were performed by 1 of 7 urologists using a robotic-assisted laparoscopic approach. Each urologist has performed at least 100 prostatectomies as an attending physician, with a range between 100 and 2500. Radical prostatectomy with pelvic lymphadenectomy was performed through a variety of approaches, including the anterior approach, posterior approach, and Retzius-sparing approach per surgeon preference. Nerve-sparing decisions were made based on tumor characteristics and patient functional status. Apical dissection included separating the prostate from the rectum posteriorly, separating neurovascular bundles laterally, dividing and ligating the dorsal venous complex, and transecting the urethra at the prostatic urethral junction. For this study, surgeries performed using an NS approach on the same side of the MRI-visible lesion were considered NS surgeries. Surgeries in patients with lesions at the midline or that did not preserve the ipsilateral neurovascular bundle were considered non-NS surgeries. Although the variable nerve preservation encompasses a continuum,17 retrospective review of operative reports did not allow for a standardized discrimination of the subtle differences in the degree of nerve preservation; therefore, this variable was treated as a binary (NS or non-NS).

Standard of Reference

Histopathological evaluation of whole-mount RP specimens served as the standard of reference. Fresh specimens were transferred to the pathology department after surgery. A dedicated genitourinary pathologist evaluated all specimens following the recommended procedures by the International Society of Urological Pathology.18 For each prostate, direct imaging-pathology anatomical registration was accomplished by using whole-mount processing. Before sectioning, the right and left side of the prostate were inked using a predetermined color system. The surgical specimen was cut axially from the base to the mid gland using a cutting plane that matches the MRI plane (Fig. 3). The prostate apex was amputated and cut in a coned manner with vertically oriented slices to obtain maximum surface area. Each cancer focus was assigned a grade group (GG),19 and the largest focus that typically exhibit the highest GG and/or stage for each patient was defined as the index lesion; in men with more than one lesion with the same GG, the largest tumor focus was defined as the index lesion.20 Positive surgical margin was defined as the presence of tumor at the inked margin of resection. Positive surgical margin was further characterized as focal versus nonfocal (established)21 and whenever possible was given a linear tumor extent measurement. Apical surgical margin status and overall surgical margin status (ie, anywhere in the prostate) on whole-mount histopathology at a patient level were tabulated as binary variables (negative, positive).

Whole-mount sectioning scheme. Whole prostate gland is received intact (left), right and left sides are inked using a predetermined color system, and then specimen is cut axially from the base to mid gland. The prostate apex is amputated and cut in a coned manner with vertically oriented slices at the apex and extreme base to obtain maximum surface area (diagram in the middle, actual specimen in the right).

Study End Points and Statistical Analysis

The primary end point of the study was the rate of aPSM. Men with PSM only in the apical region were tabulated as aPSM. Men with PSM in the apex or anywhere else in the prostate were tabulated as overall PSM (oPSM). Logistic regression was used to compare the rate of aPSM in apical lesions in the 2 groups (flagged vs nonflagged mpMRI reports for at risk of aPSM). A second comparison between the rate of aPSM in flagged apical lesions and the rate of oPSM in nonflagged lesions (ie, regardless of location) was performed. Biochemical recurrence (BCR), defined as a PSA greater than 0.2 ng/mL after RP on 2 measurements, was correlated to surgical margin status. A propensity score covariate adjustment22,23 was used to correct for potential selection bias according to age, PSA, PSA density, Gleason GG, pT stage, index lesion size, and PI-RADS score version 2. The estimate was further adjusted by including surgeon and surgical technique (NS vs non-NS approach) as covariates. Odds ratios (ORs) were reported as point estimates and 95% confidence intervals (CIs). To assess the potential impact of surgeon's experience on rate of PSM, surgeons were divided into high (at least 50 prostatectomies performed during the study period) and low volume surgeons. A significance level of 0.05 was used for statistical testing. All analyses were performed using SAS 9.4 (SAS Institute, Inc, Cary, NC).


A total of 428 men were included (Fig. 1). Table 1 provides detailed patient, index lesion and surgical data for men with (“flagged”) and without (“nonflagged”) lesions perceived as at risk for aPSM, and Table 2 presents characteristics of men with and without PSM. Flagged lesions were larger (mean size of 20 mm vs 16 mm in the nonflagged group, P = 0.0002) and had higher PI-RADS score (66% were PI-RADS 5 lesions vs 48% of the nonflagged cases, P = 0.01). More locally advanced tumors were noted in the flagged cohort (60% [55/91] were pathology-proven T3 tumors vs 47% [157/337] in the nonflagged cohort, P = 0.04; Table 1). An ipsilateral NS approach was used in 54% (232/428) of all patients, including at least partial preservation also in 54% (49/91) of the patients with flagged lesions.

TABLE 1 - Clinical, Imaging and Histopathological Characteristics of Men Without (Nonflagged) Versus With (Flagged) Lesions Interpreted on Multiparametric MRI as at Risk for Positive Surgical Margins
Nonflagged Flagged All P*
n 79% (337/428) 21% (91/428) 100% (428/428) NA
Age,† y 63.7 ± 7.3 64.3 ± 7.6 63.8 ± 7.3 0.6
PSA,† ng/mL 9.0 ± 9.6 12.3 ± 3.4 9.7 ± 17.9 0.2
Prostate volume,† mL 43.3 ± 21.1 44.5 ± 23.1 43.6 ± 21.5 0.8
PSA density,† ng/mL/mL 0.2 ± 0.3 0.3 ± 0.3 0.2 ± 0.3 0.7
Index lesion size,†‡ mm 16 ± 9 20.1 ± 11.4 16.9 ± 9.7 <0.01
Apical location No 61% (204/337) NA 48% (204/428) NA
Yes 40%(133/337) 100% (91/91) 52% (224/428)
PI-RADS v2 score 3 5%(16/337) 1% (1/91) 4% (17/428) 0.01
4 47% (158/337) 33% (30/91) 44% (188/428)
5 48% (163/337) 66% (60/91) 52% (222/428)
NS approach No 46% (154/337) 46% (42/91) 46% (196/428) 1.0
Yes 54% (183/337) 54% (49/91) 54% (232/428)
Grade group 1 5% (17/337) 3% (3/91) 5% (20/428) 0.6
2 54% (182/337) 48% (44/91) 53% (226/428)
3 23% (76/337) 23% (21/91) 23% (97/428)
4 6% (20/337) 7% (6/91) 6% (26/428)
5 13% (42/337) 19% (17/91) 14% (59/428)
pT stage 2 53% (180/337) 40% (36/91) 51% (216/428) 0.04
3a 33% (110/337) 39% (35/91) 34% (145/428)
3b 14% (47/337) 22% (20/91) 16% (67/428)
Surgical margin status 69% (232/337) 44% (40/91) 64% (272/428) <0.01
aPSM 31% (41/133) 53% (48/91) 21% (89/428)
oPSM 31% (105/337) 56% (51/91) 36% (156/428)
*P values calculated using Fisher exact test and Wilcoxon rank sum test.
†Mean ± standard deviation; other data shown represent number of patients.
‡Magnetic resonance imaging–visible index lesion.
NA indicates not applicable.

TABLE 2 - Clinical, Imaging, and Histopathological Characteristics of Patients With Negative Versus Positive Surgical Margins
Surgical Margins
Negative Positive (aPSM) Positive (oPSM) All P*
n 63.6% (272/428) 35.7% (153/428) 36.4% (156/428) 100% (428/428) NA
Age,† y 63.8 ± 7.5 63.9 ± 7.1 63.9 ± 7 63.8 ± 7.3 0.7
PSA,† ng/mL 8.0 ± 6.3 12.7 ± 28.6 12.6 ± 28.3 9.7 ± 17.9 <0.01
Prostate volume,† mL 43.9 ± 22.4 43.2 ± 20.1 43 ± 20 43.6 ± 21.5 0.7
PSA density,† ng/mL/mL 0.2 ± 0.2 0.3 ± 0.5 0.3 ± 0.5 0.2 ± 0.3 <0.01
Index lesion size,†‡ mm 14.4 ± 6.5 21.2 ± 12.7 21.1 ± 12.6 16.9 ± 9.7 <0.01
Apical location No 51.5% (140/272) 41.8% (64/153) 41% (64/156) 47.7% (204/428) 0.048
Yes 48.5% (132/272) 58.2% (89/153) 59% (92/156) 52.3% (224/428)
PI-RADS version 2 score 3 4.8% (13/272) 2.6% (4/153) 2.6% (4/156) 4% (17/428) <0.01
4 51.5% (140/272) 30.1% (46/153) 30.8% (48/156) 43.9% (188/428)
5 43.8% (119/272) 67.3% (103/153) 66.7% (104/156) 51.2% (219/428)
NS approach No 43.8% (119/272) 49% (75/153) 49.4% (77/156) 46.7% (200/428) 0.3
Yes 56.3% (153/272) 51% (78/153) 50.6% (79/156) 54.2% (232/428)
Grade group 1 7% (19/272) 0.7% (1/153) 0.6% (1/156) 4.7% (20/428) <0.01
2 57% (155/272) 45.1% (69/153) 45.5% (71/156) 52.8% (226/428)
3 20.6% (56/272) 26.1% (40/153) 26.3% (41/156) 22.7% (97/428)
4 5.9% (16/272) 6.5% (10/153) 6.4% (10/156) 6.1% (26/428)
5 9.6% (26/272) 21.6% (33/153) 21.2% (33/156) 13.8% (59/428)
pT stage 2 59.6% (162/272) 35.3% (54/153) 34.6% (54/156) 50.5% (216/428) <0.01
3a 32% (87/272) 36.6% (56/153) 37.2% (58/156) 33.9% (145/428)
3b 8.5% (23/272) 28.1% (43/153) 28.2% (44/156) 15.7% (67/428)
Cohort Flagged 14.7% (40/272) 31.4% (48/153) 32.7% (51/156) 21.3% (91/428) <0.01
Nonflagged 85.3% (232/272) 68.6% (105/153) 67.3% (105/156) 78.7% (337/428)
*P values calculated using Fisher exact test and Wilcoxon rank sum test.
†Mean ± standard deviation; other data shown represent number of patients.
‡Magnetic resonance imaging–visible index lesion.
NA indicates not applicable.

A statistically significant higher proportion of aPSMs was noted in flagged (53% [48/91]) compared with nonflagged apical lesions (31%, 41/133; propensity score–adjusted OR, 2.318; 95% CI, 1.571–3.420). A statistically significant higher proportion of aPSMs was also noted in flagged apical lesions (53% [48/91]) compared with nonflagged lesions when all margins (oPSM) in the latter group were taken into consideration (31% [105/337]; propensity score–adjusted OR, 1.978; 95% CI, 1.496–2.616). Only 3% of the flagged patients (3/91) had a PSM in a location other than the apex.

Of the 7 surgeons in the study, 3 had performed an average of 18 cases over 36 months (low volume), while the remaining four performed an average of 93 cases (high volume) over the same time span. A statistically significant higher rate of PSM was observed with lower (49% [28/57]) versus higher (35% [128/371]) volume surgeons (propensity score–adjusted OR, 1.841; 95% CI, 1.234–2.748). The ratio of odds of PSM in flagged versus nonflagged cases, however, was not statistically different between lower (OR, 2.492; 95% CI, 1.177–5.277) and higher (OR, 2.038; 95% CI, 1.507–2.756) volume surgeons (P = 0.6). Other variables associated with higher PSM rates were PSA, PSA density, lesion size, apical location, PI-RADS score, GG, and pT stage. The rate of PSM was not influenced by the use of ipsilateral NS techniques (PSM in 39.3% [77/196] and 34.1% [79/232] of the men when a non-NS and NS approach was chosen, respectively, P = 0.3; Table 2). Finally, BCR was found to be significantly associated with PSM status (propensity score–adjusted OR, 3.1; 95% CI, 1.8–5.3; P < 0.0001). However patients flagged by radiologists as having lesions at risk for aPSM did not have a significant difference in BCR rates as compared with nonflagged patients (P = 0.11).


Men with PSMs are almost twice as likely to receive radiotherapy than those with R0 resection.24 Several factors may influence surgical margin status. As in our study, lesions located in the apex have been shown to be at increased risk for PSM.8,25 We have shown that a subgroup of apical prostate lesions is at even higher risk of PSM (53% for lesions in the distal most prostate and/or encircling the distal most prostatic urethra compared with 31% for other apical lesions, and 31% for nonapical lesions) and can be prospectively flagged by radiologists using standardized language in a structured report. Our results have several potential implications for management decisions.

In a retrospective review of 84 men who underwent prostate MRI, followed by RP, Yao et al25 identified PSM in 21.4% of the men, of which 83.3% were situated in the apex. Consistent with prior reports,9 our histopathological analysis of the whole-mount RP specimens confirmed a higher incidence of PSM in apical lesions. By defining the “very distal” apical prostatic zone as the distal 6-mm apical portion of the prostate starting from the distal most visualized part of the prostate in the craniocaudal plane, Nix et al26 identified very distal apical tumors in 18% of the men (38/210). This is consistent with our data identifying flagged lesions in 21% (91/428) of patients.

Surgical technique and experience are important factors influencing the risk of PSM.27 Specifically, the use of an NS approach has been associated with an increased risk of PSM.28 Interestingly, our study did not find a statistically significant difference in the rate of PSM when a non-NS versus NS approach was used. In keeping with reports from other centers,27,29 we did find higher PSM rates among lower volume surgeons. However, in our cohort, surgeon's experience did not seem to affect the difference in PSM rates of flagged versus nonflagged cases, suggesting that intraoperative adjustments for surgical margins are likely offset in part by disease biology and the limitations to extent of resection while still preserving critical structures.

The potential value of MRI information for surgical planning has been previously emphasized by others.30 Finley et al31 advocated the use of a RP-geared MRI reporting approach, including spatial anatomic information in the form of an axial clock-face tumor location, capsular distance, and distance from apex to base. Similarly, other groups have explored the value of mpMRI in pre-RP planning.31,32 In a prospective assessment of 104 men who underwent preoperative prostate MRI, McClure et al12 demonstrated a change in the NS approach during RP in 27% of the patients. Moreover, Petralia et al33 reported the use of intraoperative frozen-section analysis (IFS) during NS robotic-assisted RP of the area where the presurgical MRI showed contact between the index lesion and the prostate glandular margin and found that IFS had a lower rate of PSM as compared with a control group without IFS. To our knowledge, however, the use of standardized language in structured reports to flag patients at risk for PSM in specific locations of the prostate has not been studied. The perceived gap in communicating this level of concern to referring physicians when reviewing cases during imaging-pathology conferences was the motivation to propose this standardized approach. This technique has several advantages: (1) it is easy to recognize, (2) has unequivocal meaning, (3) its performance is auditable down to the individual level for both the interpreting radiologist and urologist, and (4) provides a means to assess the impact of different surgical approaches.

Our study has limitations. First, a selection bias is possible given the retrospective, nonrandomized design. The data, however, were prospectively generated, and a validated tool to adjust for potential differences across subcohorts—propensity score weighting—was used to correct for confounding variables. The criteria used for flagging certain lesions as at risk for PSM are somewhat subjective. The 9 board-certified radiologists interpreting the MRIs in this study participate in monthly quality assurance prostate conferences where a consensus approach is reached for these kinds of interpretation. However, we did not evaluate interreader agreement and the reproducibility of our results in other academic and private settings needs to be assessed. Furthermore, assessment of interreader agreement in the flagging of apical lesions was not performed herein. In addition, higher complexity cases expected in our tertiary care, academic center may skew the predictive values, potentially further affecting generalizability of our findings. Moreover, although surgeons were communicated of the intended purpose of this tool, we did not assess whether surgeons used the imaging information to modify the surgical approach. While the standard language for the risk of aPSM was introduced in the mpMRI structured report in 2016, urologists were not formally instructed on its value given the lack of supporting data at the time of implementation. Finally, a case-by-case imaging-pathology correlation to identify potential false-positive MRI findings was not performed.


Standardized language in the structured reports for mpMRI of the prostate helps the preoperative identification of patients at risk for aPSM. Follow-up studies addressing the role of this information in facilitating appropriate patient counseling, optimizing treatment decisions in a prospective setting (eg, selective use of modified surgical technique, intraoperative frozen-section margin assessment,33 use of radiation therapy as preferred treatment modality) and its potential for reducing PSMs compared with men who did not undergo preoperative MRI should be encouraged.


1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018;68:7–30.
2. Sanda MG, Cadeddu JA, Kirkby E, et al. Clinically localized prostate cancer: AUA/ASTRO/SUO guideline. Part I: risk stratification, shared decision making, and care options. J Urol. 2018;199:683–690.
3. Wieder JA, Soloway MS. Incidence, etiology, location, prevention and treatment of positive surgical margins after radical prostatectomy for prostate cancer. J Urol. 1998;160:299–315.
4. Pfitzenmaier J, Pahernik S, Tremmel T, et al. Positive surgical margins after radical prostatectomy: do they have an impact on biochemical or clinical progression?BJU Int. 2008;102:1413–1418.
5. Hackman G, Taari K, Tammela TL, et al. Randomised trial of adjuvant radiotherapy following radical prostatectomy versus radical prostatectomy alone in prostate cancer patients with positive margins or extracapsular extension. Eur Urol. 2019;76:586–595.
6. Meeks JJ, Eastham JA. Radical prostatectomy: positive surgical margins matter. Urol Oncol. 2013;31:974–979.
7. Lee H, Kim K, Hwang SI, et al. Impact of prostatic apical shape and protrusion on early recovery of continence after robot-assisted radical prostatectomy. Urology. 2014;84:844–849.
8. Stamey TA, Villers AA, McNeal JE, et al. Positive surgical margins at radical prostatectomy: importance of the apical dissection. J Urol. 1990;143:1166–1172; discussion 1172-3.
9. Wadhwa H, Terris MK, Aronson WJ, et al. Long-term oncological outcomes of apical positive surgical margins at radical prostatectomy in the Shared Equal Access Regional Cancer Hospital cohort. Prostate Cancer Prostatic Dis. 2016;19:423–428.
10. Gupta RT, Faridi KF, Singh AA, et al. Comparing 3-T multiparametric MRI and the Partin tables to predict organ-confined prostate cancer after radical prostatectomy. Urol Oncol. 2014;32:1292–1299.
11. Schnall MD, Imai Y, Tomaszewski J, et al. Prostate cancer: local staging with endorectal surface coil MR imaging. Radiology. 1991;178:797–802.
12. McClure TD, Margolis DJ, Reiter RE, et al. Use of MR imaging to determine preservation of the neurovascular bundles at robotic-assisted laparoscopic prostatectomy. Radiology. 2012;262:874–883.
13. Wetterauer C, Winkel DJ, Federer-Gsponer JR, et al. Structured reporting of prostate magnetic resonance imaging has the potential to improve interdisciplinary communication. PLoS One. 2019;14:e0212444.
14. Costa DN, Pedrosa I, Roehrborn C, Rofsky NM. Multiparametric magnetic resonance imaging of the prostate: technical aspects and role in clinical management. Topics in magnetic resonance imaging: TMRI. 2014;23:243–257.
15. Prostate Imaging and Reporting and Data System: Version 2. 2015.
16. Freifeld Y, Diaz de Leon A, Xi Y, et al. Diagnostic performance of prospectively assigned Likert scale scores to determine extraprostatic extension and seminal vesicle invasion with multiparametric MRI of the prostate. AJR Am J Roentgenol. 2019;212:576–581.
17. Schatloff O, Chauhan S, Sivaraman A, et al. Anatomic grading of nerve sparing during robot-assisted radical prostatectomy. Eur Urol. 2012;61:796–802.
18. Samaratunga H, Montironi R, True L, et al. International Society of Urological Pathology (ISUP) Consensus Conference on handling and staging of radical prostatectomy specimens. Working group 1: specimen handling. Mod Pathol. 2011;24:6–15.
19. Epstein JI, Zelefsky MJ, Sjoberg DD, et al. A contemporary prostate cancer grading system: a validated alternative to the Gleason score. Eur Urol. 2016;69:428–435.
20. Ahmed HU. The index lesion and the origin of prostate cancer. N Engl J Med. 2009;361:1704–1706.
21. Magi-Galluzzi C, Evans AJ, Delahunt B, et al. International Society of Urological Pathology (ISUP) consensus conference on handling and staging of radical prostatectomy specimens. Working group 3: extraprostatic extension, lymphovascular invasion and locally advanced disease. Mod Pathol. 2011;24:26–38.
22. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46:399–424.
23. Li F, Zaslavsky AM, Landrum MB. Propensity score weighting with multilevel data. Stat Med. 2013;32:3373–3387.
24. Grossfeld GD, Chang JJ, Broering JM, et al. Impact of positive surgical margins on prostate cancer recurrence and the use of secondary cancer treatment: data from the CaPSURE database. J Urol. 2000;163:1171–1177; quiz 1295.
25. Yao A, Iwamoto H, Masago T, et al. The role of staging MRI in predicting apical margin positivity for robot-assisted laparoscopic radical prostatectomy. Urol Int. 2014;93:182–188.
26. Nix JW, Turkbey B, Hoang A, et al. Very distal apical prostate tumours: identification on multiparametric MRI at 3 Tesla. BJU Int. 2012;110(11 Pt B):E694–E700.
27. Vickers A, Bianco F, Cronin A, et al. The learning curve for surgical margins after open radical prostatectomy: implications for margin status as an oncological end point. J Urol. 2010;183:1360–1365.
28. Preston MA, Breau RH, Lantz AG, et al. The association between nerve sparing and a positive surgical margin during radical prostatectomy. Urol Oncol. 2015;33:18.e1–18.e6.
29. Eastham JA, Kattan MW, Riedel E, et al. Variations among individual surgeons in the rate of positive surgical margins in radical prostatectomy specimens. J Urol. 2003;170(6 Pt 1):2292–2295.
30. McEvoy SH, Raeside MC, Chaim J, et al. Preoperative prostate MRI: a road map for surgery. AJR Am J Roentgenol. 2018;211:383–391.
31. Finley DS, Margolis D, Raman SS, et al. Fine-tuning robot-assisted radical prostatectomy planning with MRI. Urol Oncol. 2013;31:766–775.
32. Secin FP, Serio A, Bianco FJ Jr., et al. Preoperative and intraoperative risk factors for side-specific positive surgical margins in laparoscopic radical prostatectomy for prostate cancer. Eur Urol. 2007;51:764–771.
33. Petralia G, Musi G, Padhani AR, et al. Robot-assisted radical prostatectomy: multiparametric MR imaging–directed intraoperative frozen-section analysis to reduce the rate of positive surgical margins. Radiology. 2015;274:434–444.

prostate cancer; MRI; structured report; prostatectomy; surgical margin

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.