Original article
Prostate magnetic resonance imaging: The truth lies in the eye of the beholder

https://doi.org/10.1016/j.urolonc.2017.12.013Get rights and content

Highlights

  • Multiparametric MRI (mpMRI) is increasingly used to aid surgical planning before prostatectomy.

  • A cohort of 233 men undergoing prostatectomy had mpMRI performed and rereviewed with focus on accuracy for interpreting EPE and SVI.

  • For EPE, there was low concordance comparing the initial vs. repeat MRI interpretation (κ = 0.22).

  • Comparison of initial MRI interpretation vs. reread by senior radiologist noted universal improvements in sensitivity, specificity, PPV, NPV, and accuracy.

  • Even at a tertiary referral center, interobserver variability among radiologists regarding local extent of disease on prostate MRI is high.

Abstract

Purpose

To determine the diagnostic accuracy and interobserver variability of radiologic interpretation of magnetic resonance imaging (MRI) performed for surgical planning before prostatectomy.

Patients and methods

The records of 233 men undergoing prostatectomy with presurgical multiparametric 3T surface body coil MRI were reviewed. All initial films were read by a fellowship-trained body radiologist provided with relevant clinical information. A senior radiologist then reread all pelvic MRIs blinded to the initial interpretation with findings from both readings compared to final pathology. Kappa (κ) scores as well as sensitivity, specificity, positive predictive values (PPV), negative predictive value (NPV), and accuracy were determined.

Results

When considering extraprostatic extension (EPE), there was low concordance comparing the initial vs. repeat MRI interpretation (κ = 0.22). Additionally, when the senior radiologist reread his own initial interpretation (n = 93, blinded to initial result), concordance for EPE was greater (κ = 0.36) albeit similarly low. With regard to EPE, a comparison of initial MRI interpretation vs. reread by senior radiologist noted universal improvements in diagnostic characteristics including sensitivity (30.3% vs. 56.1%), specificity (80.2% vs. 88.6%), PPV (37.7% vs. 66.1%), NPV (74.4% vs. 83.6%), and accuracy (66.1% vs. 79.4%). In contrast, seminal vesicle invasion interpretation was more uniform whereby initial MRI interpretation vs. reread yielded similar sensitivity (18.2% vs. 27.3%), specificity (97.2% vs. 93.8%), PPV (40.0% vs. 31.6%), NPV (91.9% vs. 92.5%), and accuracy (89.7% vs. 87.6%).

Conclusions

Even at a tertiary referral center, interobserver variability among radiologists regarding local extent of disease on prostate MRI is high. These observations underscore the importance of uniformity when defining criteria for EPE and seminal vesicle invasion to allow for optimal presurgical planning.

Introduction

Technological advances along with access and availability have increased utilization of multiparametric magnetic resonance imaging (MRI) in the detection, staging, and treatment planning phases of prostate cancer [1], [2]. At present, radical prostatectomy remains the mainstay surgical therapy for localized prostate cancer. With regard to surgical planning, accurate preoperative assessment of local disease extension in prostate cancer is paramount to the urologic surgeon’s treatment decisions [3]. Specifically, by evaluating for extraprostatic extension (EPE) and seminal vesicle invasion (SVI), preoperative pelvic MRI can potentially affect decision-making regarding a nerve-sparing approach and extent/width of resection. Such surgical nuances clearly can affect quality of life and functional outcomes in men after the operative procedure.

With this in mind, MRI has increasingly become the widespread modality of choice for prostate imaging [4], although prior studies have presented conflicting and inconsistent results [1], [5], [6], [7], [8], [9], [10], [11], [12]. Some of the limitations of prior studies include heterogeneous cohorts, different MRI sequences such as 1.5T vs. 3T MRI, and endorectal vs. surface body coil MRI [5], [6].

A few prior studies have evaluated interobserver agreement of prostate MRI albeit with smaller sample sizes and varying degrees of radiologic experience [8], [9], [11]. Therefore, the goal of this study was to critically evaluate a series of MRIs performed for surgical planning with initial review by fellowship-trained body radiologists and subsequent rereview (in a blinded fashion) by a senior radiologist. We aim to determine if variability issues persist in a tertiary referral center with fellowship-trained radiologists and, if so, the relative accuracy of preoperative MRI imaging.

Section snippets

Patients and methods

A single institution retrospective review identified 269 individuals who underwent prostatectomy from 2010 to 2015 for suspected localized prostate cancer. Of this cohort, 233 (86.7%) were able to obtain a preoperative multiparametric pelvic 3T surface body coil MRI. Multiparametric MRI was routinely ordered for all patients but was not performed in those with contraindications (pacemaker, metal fragments, claustrophobia) or cost with insurance reimbursement issues. Protocols for image

Results

The cohort had a median age of 62 years old with a median PSA of 6.2. Of the 233 patients, 143 (61.4%) had clinical stage T1 disease and the remaining 90 (38.6%) patients had clinical T2 disease. The vast majority of prostatectomies, 228 (97.9%), were performed robotically. Over 80% of the cohort had intermediate or high-risk prostate cancer, and 8% of patients had positive subcentimeter LNs for malignancy not appreciated on preoperative MRI (Table 1).

The initial MRI interpretation was

Discussion

Decisions regarding nerve-sparing surgery (and extent of nerve preservation) for prostate cancer can be influenced by preoperative imaging. Prostate MRI is one of the imaging modalities utilized for preoperative assessment of locoregional involvement of prostate cancer. Its utility, however, is predicated on the quality of the study as well as the ability of the radiologist to interpret the study with a high degree of accuracy.

Our study specifically explored interobserver agreement and

Conclusion

Even at an academic medical center with fellowship-trained body radiologists, interobserver agreement to evaluate local extent of disease on prostate MRI is fair to relatively poor. We report, however, improved diagnostic characteristics for pelvic EPE when a senior member of the body radiology team theoretically with more experience reads the MRI. These findings underscore the importance of uniformity when defining criteria for EPE invasion to allow for appropriate surgical planning.

Acknowledgments

The authors thank the Ken and Bonnie Shockey Fund for Urologic Research at Penn State Health for supporting this work.

References (17)

There are more references available in the full text version of this article.

Cited by (15)

  • Prostate Cancer Local Staging with Magnetic Resonance Imaging

    2024, Radiologic Clinics of North America
  • Bioptic intraprostatic chronic inflammation predicts adverse pathology at radical prostatectomy in patients with low-grade prostate cancer

    2020, Urologic Oncology: Seminars and Original Investigations
    Citation Excerpt :

    In spite of the potential role of mpMRI and Genomic tests in improving risk stratification in ISUP GG 1 and 2 CaP patients [31], these tests are expensive. Furthermore, mpMRI demonstrated a limited accuracy in predicting extracapsular disease in low-risk patients and a significant inter-reader variability [32–35]. In this scenario, the identification of easily available, validated information that pathologists can provide at no additional costs when reporting PBx, like the Irani scores, is particularly attractive.

  • Improved Detection of Clinically Significant Prostate Cancer With Software-assisted Systematic Biopsy Using MR/US Fusion in Patients With Negative Prostate MRI

    2018, Urology
    Citation Excerpt :

    As such, we may have underestimated the detection rates of clinically significant PCa. The utilization of mpMRI comes with inherent limitations: difficulty with accuracy of the biopsy sites via MRI and the fusion software, as well as the known interobserver variability that exists between radiologist interpretations of MRI data.29 Re-examination of the mpMRI in the 6 cases with clinically significant PCa, with knowledge of the biopsy outcome, demonstrated at least one instance where the original clinical interpretation of mpMRI could have resulted in missed diagnosis.

View all citing articles on Scopus
View full text