T2-weighted hypointense lesions within prostate gland: Differential diagnosis using wash-in rate parameter on the basis of dynamic contrast-enhanced magnetic resonance imaging–Hystopatology correlations

Francesco Pierconti, Pierfrancesco Bassi, Anna Lia Valentini, Benedetta Gui, Alessandro Cina, Francesco Pinto, Angelo Totaro, Lorenzo Bonomo

Research output: Contribution to journalArticle

Abstract

BACKGROUND AND AIMS: Dynamic contrast enhanced magnetic resonance improves prostate cancer detection. The aims of this paper are to verify whether wash-in-rate parameter (speed of contrast uptake in dynamic contrast enhanced magnetic resonance) can help to differentiate prostate cancer from non-neoplastic T2-weighted hypointense lesions within prostate gland and to assess a cut-off for prostate cancer diagnosis. METHODS: Prospective, monocentric, multi-departmental study. Thirty consecutive patients underwent T2-weighted and dynamic contrast enhanced magnetic resonance, and re-biopsy. T2-weighted hypointense lesions, >5mm in size, were noted. Lesions were assessed as cancerous (showing mass effect, or no defined margin within transitional zone) and non cancerous (no mass effect) and were compared with histopathology by 2×2 tables. Wash-in-rate of each lesion was calculated and was correlated with histopathology. Student's t-test was adopted to assess significant differences. Receiver operating characteristic (ROC) analysis was employed to identify the best cut-off for wash-in-rate in detecting prostate cancer. RESULTS: At re-biopsy, cancer was proven in 43% of patients. On T2-weighted MRI, 111 hypointense lesions ≥5 mm in size were found. Sensitivity, specificity and accuracy of T2-weighted MRI were 80% (±12.4 CI 95%), 74.6% (±10.1 CI 95%), and 76.5% (±7.9 CI 95%), respectively. Mean WR was 5.8±1.9/s for PCa zones and 2.96±1.44/s for non-PCa zones (p<0.00000001). At ROC analysis, the best area under curve (AUC) for wash-in-rate parameter was associated to 4.2/s threshold with 82.5% sensitivity (CI±7.07), 97.2% specificity (CI±4.99) and 91.2% accuracy (CI±5.27). Eighteen false positive lesions on T2-weighted MRI showed low wash-in-rate values suggesting non-cancer lesions, while in 5/8 false negative cases high wash-in-rate values correctly suggested prostate cancer. Nine lesions with surgically proven cancer were not included in the saturation biopsy scheme, in 2/9 cases the only site of cancer. CONCLUSIONS: Wash-in-rate parameter allows to differentiate prostate cancer from non-neoplastic lesions, helping cancer detection in areas not included in the biopsy scheme.
Original languageEnglish
Pages (from-to)3090-3095
Number of pages6
JournalEuropean Journal of Radiology
DOIs
Publication statusPublished - 2012

Keywords

  • Prostate cancer
  • magnetic resonance

Fingerprint

Dive into the research topics of 'T2-weighted hypointense lesions within prostate gland: Differential diagnosis using wash-in rate parameter on the basis of dynamic contrast-enhanced magnetic resonance imaging–Hystopatology correlations'. Together they form a unique fingerprint.

Cite this