What’s new in Genitourinary Imaging – December 2020

3 years ago


MRI-based Bosniak Classification of Cystic Renal Masses, Version 2019: Interobserver Agreement, Impact of Readers’ Experience, and Diagnostic Performance

Bai X, Sun SM, Xu W, Kang HH, Li L, Jin YQ, Gong QG, Liang GC, Liu HY, Liu LL, Chen SL, Wang QR, Wu P, Guo AT, Huang QB, Zhang XJ, Ye HY, Wang HY.

Radiology. 2020 Sep 22:200478. doi: 10.1148/radiol.2020200478. Epub ahead of print. PMID: 32960726.


This group conducted a study to evaluate the interobserver agreement and the diagnostic performance of 2019 Bosniak classification system compared to the 2005 version. This was a retrospective study that included 207 patients with cystic renal masses and surgical pathology. These cystic renal masses were evaluated independently by 8 radiologists using both the 2019 and the 2005 versions of the Bosniak criteria. The kappa value for interobserver agreement was 0.64 vs 0.50 for the 2019 and 2005, respectively, which was significantly different (p < 0.001). There was as similar interobserver agreement between senior and junior radiologists for the 2019 version with a kappa value of 0.65 vs 0.64. The diagnostic performance of the 2019 version was better with higher specificity (83% vs 68%) and no significantly different sensitivity (89% vs 84%). The investigators concluded that the 2019 version of the Bosniak criteria improved interobserver agreement, which was not affected by level of reader experience, and specificity for the detection of cystic renal malignancies, while maintaining sensitivity.


Diagnostic Algorithm to Differentiate Benign Atypical Leiomyomas from Malignant Uterine Sarcomas with Diffusion-weighted MRI

Abdel Wahab C, Jannot AS, Bonaffini PA, Bourillon C, Cornou C, Lefrère-Belda MA, Bats AS, Thomassin-Naggara I, Bellucci A, Reinhold C, Fournier LS.

Radiology. 2020 Nov;297(2):361-371. doi: 10.1148/radiol.2020191658. Epub 2020 Sep 15. PMID: 32930650.


This group conducted a study to determine MRI features that improved the ability to differentiate uterine sarcoma from atypical leiomyomas. MRI features from T2 weighted imaging, DWI, and ADC maps were used to effectively predict and differentiate between pathologically proven uterine sarcomas from atypical leiomyomas. 156 women were retrospectively evaluated and the predictive MRI criteria for malignancy were defined as high DWI signal (greater than the endometrium), ADC ≤ 0.905 x 10-3, lymphadenopathy, and peritoneal implants. Authors also found that global or focal area of low T2 signal with low intermediate DWI signal less than the endometrium or lymph nodes correlated well to benign tumors. The authors tested their diagnostic algorithm an internal training set (51 sarcomas and 105 leiomyomas) and two external validation sets (45 sarcomas and 56 leiomyomas). The algorithm yielded high sensitivities and specificities in differentiating sarcoma from atypical leiomyomas in the training set (98% and 94%, respectively) and external validation sets (88% and 100%, respectively).


Right-sided scrotal varicocele and its association with malignancy: a multi-institutional study

Itani M, Kipper B, Corwin MT, Burgan CM, Fetzer DT, Shenoy-Bhangle AS, Althubaity A, Loehfelm TW, Middleton WD, Fananapazir G.

Abdom Radiol (NY). 2020 Nov 5. doi: 10.1007/s00261-020-02840-9. Epub ahead of print. PMID: 33151361.


The group conducted a retrospective, multi-institutional study (including 5 large institutions) to evaluate if patients with isolated right-sided varicocele on scrotal ultrasound had additional evidence of malignancy in the abdomen and pelvis. Patients with existing cancer diagnosis in the abdomen and pelvis, epipdymo-orchitis, intratesticular or intrascrotal mass, or prior scrotal surgery were excluded. Diagnosis of varicoceles were confirmed by the radiologist based on pampiniform plexus vein diameter ≥ 2 mm with augmented flow or increases by ≥ 1 mm with Valsalva. Amongst the 210 patients included in the study, there was no underlying malignancy to explain right sided varicocele concluding that adult isolated right varicoceles are highly unlikely to be the only presenting finding of an abdominal malignancy and that the finding of an isolated right varicocele does not warrant further imaging of the abdomen or pelvis.


Non-contrast MR angiography versus contrast enhanced MR angiography for detection of renal artery stenosis: a comparative analysis in 400 renal arteries

Lal H, Singh RKR, Yadav P, Yadav A, Bhadauria D, Singh A.

Abdom Radiol (NY). 2020 Nov 2. doi: 10.1007/s00261-020-02836-5. Epub ahead of print. PMID: 33136181.


This group conducted a study comparing the diagnostic performance of non-contrast MRA (ncMRA) with contrast enhanced MRA (ceMRA) for the evaluation of renal artery stenosis (RAS). The goal was to establish ncMRA as a viable alternative to ceMRA, which could be useful for the evaluation of the renal vasculature in patients who do not wish to receive intravenous contrast.  They assessed 400 renal arteries in 201 patients using 3D fat suppressed inflow inversion recovery balanced steady state free precession (SSFP) technique and 3D fast spoiled gradient echo (FSPGR) for ncMRA and ceMRA respectively. ncMRA underdiagnosed a grade 3 stenosis as grade 2 stenosis in 2 arteries when compared to ceMRA. 72 patients were diagnosed with hemodynamically significant stenosis (grade 3 or 4) by both ncMRA and ceMRA. The study showed that the SSFP-based ncMRA was non-inferior to the FSPGR-based ceMRA for detection of hemodynamically significant renal artery stenosis.


Prospective Evaluation of PI-RADS Version 2.1 for Prostate Cancer Detection

Walker SM, Mehralivand S, Harmon SA, Sanford T, Merino MJ, Wood BJ, Shih JH, Pinto PA, Choyke PL, Turkbey B.

AJR Am J Roentgenol. 2020 Sep 2:1-6. doi: 10.2214/AJR.19.22679. Epub ahead of print. PMID: 32877244.


This group completed a prospective study of 110 patients who underwent mpMRI followed by targeted fusion and systematic biopsy. The MRI studies were interpreted based on ACR PIRADs 2.1 criteria. A total of 171 lesions were detected by mpMRI and biopsied with 33.3% (57 of the 171 lesions) having clinically significant prostate cancer (Gleason grade group > 1). 5.7% of PIRADs 2, 14.8% of PIRADs 3, 44.2% of PIRADs 4, and 80% of PIRADs 5 had clinically significant prostate cancer. 16 patients had normal mpMRI findings with 18.8% having clinically significant prostate cancer. The conclusion was that cancer detection rate increased with an increased PIRADs score.


Prostatitis, the Great Mimicker of Prostate Cancer: Can We Differentiate Them Quantitatively With Multiparametric MRI?

Uysal A, Karaosmanoğlu AD, Karcaaltıncaba M, Akata D, Akdogan B, Baydar DE, Ozmen MN.

AJR Am J Roentgenol. 2020 Nov;215(5):1104-1112. doi: 10.2214/AJR.20.22843. Epub 2020 Sep 9. PMID: 32901562.


This group conducted a study to determine efficacy of using quantitative pharmacokinetic parameters and ADC values to differentiate prostatitis from prostate cancer. They conducted a retrospective review of patients with biopsy proven prostate cancer and prostatitis and compared mean ADC and multiple pharmacokinetic parameters. Of the 138 patients evaluated, 44 had prostatitis with significantly higher extravascular extracellular space volume fraction (Ve). There was a 92.7% overall diagnostic accuracy prostatitis with ADC, reverse volume transfer constant (Kep), and time to peak (TTP) as the most predictive combination. The analysis of the signal intensity-time curves show that the type 3 enhancement curves were not seen in any cases of prostatitis. The group argues that the use of these parameters has no additional burden to patients and can be used as an additional tool in differentiating prostate cancer from prostatitis.


Can MRI features differentiate ovarian mucinous carcinoma from mucinous borderline tumor?

Kaga T, Kato H, Hatano Y, Kawaguchi M, Furui T, Morishige KI, Matsuo M.

Eur J Radiol. 2020 Nov;132:109281. doi: 10.1016/j.ejrad.2020.109281. Epub 2020 Sep 12. PMID: 32961452.


This group performed a study to determine if ovarian mucinous carcinoma and mucinous borderline tumor could be differentiated based on MRI features. They found that mucinous carcinomas at presentation were usually larger tumors (219.7 mm vs 177.4 mm, p < 0.05) with larger maximum mural nodule diameters (41.7 mm vs 6.6 mm, p < 0.01). ADC values were also significantly lower in mucinous carcinomas measuring 1.20×10-3 vs 1.61×10-3 (p < 0.05). The group also found that number of loculi, honeycomb sign, stained glass appearance, fluid-fluid level, thickened septa larger than 5 mm, peritoneal dissemination, or T2 hyperintense microcytes were not effective differentiating factors between ovarian mucinous carcinoma and mucinous borderline tumor.


Dynamic 68Ga-PSMA-11 PET/CT for the primary evaluation of localized renal mass: a prospective study

Golan S, Aviv T, Groshar D, Yakimov M, Zohar Y, Prokocimer Y, Nadu A, Baniel J, Domachevsky L, Bernstine H.

J Nucl Med. 2020 Oct 23:jnumed.120.251272. doi: 10.2967/jnumed.120.251272. Epub ahead of print. PMID: 33097628.


This group conducted a prospective case series in 27 patients with 29 enhancing renal masses to determine the ability of 68Ga-PSMA -11 PET/CT for the evaluation of renal masses. Increased PSMA expression has been demonstrated in extra prostatic tissues and specifically renal masses. The aim of the study was to determine if there is a difference in the SUV values between malignant and benign lesions in the kidneys. 24 of the 29 enhancing renal masses were pathologically confirmed as malignant and 20 of those lesions had positive PSMA staining. Both median SUV mean and SUV max were higher for malignant renal lesions (SUV mean – 2.3 for benign and 6.8 for malignant, SUV max – 3.8 for benign and 9.4 for malignant). The washout coefficient was significantly lower in malignant lesions. Based on the results, the group concluded that PSMA could serve as a promising tracer to the evaluation of renal masses.

  • Share