Design: Retrospective diagnostic accuracy study with four independent readers
Population: 423 lytic long-bone lesions in skeletally mature patients, mean age 62
Primary outcome: Malignancy on biopsy or 2-year radiographic stability
Effect: Sensitivity 97.1%, specificity 67.4%, NPV 92.4%; OR 9.7 per score point

Bone-RADS is an American College of Radiology scoring system for lytic bone lesions, built from five radiographic features and a history of primary cancer. This retrospective study scored 423 lytic long-bone lesions in adults, using biopsy or two years of radiographic stability as the reference.

Each one-point rise in the score was associated with much higher odds of malignancy (OR 9.7). For separating possibly benign from possibly malignant, sensitivity was 97.1% and negative predictive value 92.4%, with specificity of 67.4%. Agreement between readers on the four-tier score was only moderate (weighted κ 0.52), but better for the simple benign-versus-malignant split.

For orthopaedic surgeons who see incidental lytic lesions on fracture-clinic films, this supports Bone-RADS as a triage step: a low score is reassuring, while a high score warrants prompt cross-sectional imaging and referral.

Consider scoring incidental lytic long-bone lesions with Bone-RADS to guide urgency of referral.
Look for the features linked to malignancy: aggressive margins, endosteal erosion, pathological fracture and soft-tissue mass.
Do not rely on a low score alone in patients with a known primary cancer.
Expect false positives; specificity was about two-thirds.

Why it matters

Incidental lytic lesions are common, and a reproducible triage score helps decide which need urgent work-up.

The statistics, in plain English

A sensitivity of 97% means the score flagged almost all cancers. A specificity of 67% means about one in three benign lesions was also flagged, so many referrals will turn out benign.