Design: Prospective method-comparison study
Population: 328 men undergoing 3 T multiparametric prostate MRI
Primary outcome: Agreement of ellipsoid volume with manual segmentation
Effect: ICC 0.985; mean absolute difference 5.5 cc; underestimation above 80 cc

This prospective European Radiology study, published on 28 September, compared prostate volume from the PI-RADS ellipsoid formula (length × width × height × 0.52) with manual segmentation of every axial slice in 328 men having 3 T MRI for suspected cancer.

Agreement was excellent (ICC 0.985), with a mean absolute difference of 5.5 cc. The ellipsoid method underestimated glands over 80 cc. Adjusting the 0.52 factor by gland size did not help meaningfully. For PSA density cut-offs of 0.10, 0.15 and 0.20, the ellipsoid volume classified patients the same way as segmentation with sensitivity 0.86 to 0.95 and specificity 0.88 to 0.97.

This supports continued use of the quick formula. In large glands, underestimated volume will overestimate PSA density, which could push borderline patients towards biopsy.

Keep using the PI-RADS ellipsoid formula for routine prostate volume.
Measure from both sagittal and axial T2-weighted images as PI-RADS describes.
In glands over 80 cc, note that volume may be underestimated and PSA density overestimated.
Do not adjust the 0.52 factor; tailoring it did not help.

Why it matters

A borderline PSA density in a large gland may be a measurement artefact.

The statistics, in plain English

An intraclass correlation of 0.985 means the two methods agree almost perfectly across the whole range. Sensitivity and specificity here measure agreement with segmentation at each cut-off, not accuracy for cancer.