Mirels Score for the Impending Pathologic Fracture
A metastatic long bone lesion fixed before it breaks means less blood loss, a shorter stay and a better chance of walking again than fixing the same bone after it fractures.
Key terminology
- Impending pathologic fracture
- This is a bone weakened by tumor that has not yet broken. Treating it before it breaks is the whole decision this plate covers.
- Mirels score
- This bedside score adds 4 features of a metastatic long bone lesion, each worth 1 to 3 points. A higher total means a higher chance the bone will break.
- Peritrochanteric
- This region is the upper femur around the greater and lesser trochanters, just below the femoral neck. It takes the highest load in the body and scores highest for site.
- Functional pain
- This is pain brought on by loading the limb, such as walking or lifting. It signals the bone is failing mechanically and scores highest for pain.
- Lytic
- This lesion destroys bone and looks dark or punched out on a film. Lytic lesions weaken bone the most and score highest for lesion type.
- Blastic
- This lesion lays down abnormal dense bone and looks white on a film. Prostate cancer is the classic cause, and blastic lesions score lowest for type.
- Prophylactic fixation
- This surgery fixes a bone before it breaks, usually with a long intramedullary nail for the femur. Radiation to the area typically follows.
Takeaways
- The Mirels score grades an impending pathologic fracture by adding up 4 scored features. Each feature scores 1 to 3. Site scores upper limb 1, lower limb 2 and peritrochanteric 3. Pain scores mild 1, moderate 2 and functional pain 3, and the lesion scores blastic 1, mixed 2 and lytic 3. Size scores 1, 2 or 3 for less than one-third, one-third to two-thirds, or more than two-thirds of the bone's width, for a total of 4 to 12 (Mirels, Clin Orthop Relat Res 1989;(249):256-264, PMID 2684463; Hoban et al, JSES Int 2022;6:675-681, PMID 35813136).
- A Mirels score of 9 or more favors prophylactic fixation before radiation. Of 78 irradiated lesions, those that broke within 6 months had a mean score of 10 vs 7. A score of 7 or less can be radiated alone, 8 had about a 15% fracture risk and 9 about 33%. Mirels drew the line at 8, and most later work uses 9, so 8 is a judgment call (Mirels, Clin Orthop Relat Res 1989;(249):256-264, PMID 2684463; Hoban et al, JSES Int 2022;6:675-681, PMID 35813136; Damron et al, Clin Orthop Relat Res 2016;474:643-651, PMID 26169800).
- Fixing a femur before it breaks goes better than fixing it after. In one surgeon's series, 97 impending fractures had less blood loss (438 vs 636 cc) and a shorter stay (7 vs 11 days) than 85 completed fractures. More went home instead of to a facility (79% vs 56%) and more walked without support (35% vs 12%). The series was retrospective and not randomized (Ward et al, Clin Orthop Relat Res 2003;(415 Suppl):S230-S244, PMID 14600615).
- Mirels is a sensitive screen with poor specificity. 53 raters from residents to tumor surgeons scored 12 real femoral cases. Sensitivity was 91% and specificity 35%, and the score beat clinical judgment at every experience level. Agreement was good for site and only fair for size and pain, the 2 parts that need judgment (Damron et al, Clin Orthop Relat Res 2003;(415 Suppl):S201-S207, PMID 14600611).
- Radiation follows fixation because the implant does not treat the tumor. In 64 stabilizations for metastatic disease, normal use of the limb was reached in 53% with postoperative radiation vs 11.5% with surgery alone. Second operations at the same site were more frequent without radiation. The authors flag possible selection bias, and no trial has tested it (Townsend et al, J Clin Oncol 1994;12:2345-2350, PMID 7669102).
- Upper limb lesions break at lower Mirels scores than femoral lesions. In the humerus a cutoff of 9 caught only 14.5% of fractures, and a cutoff of 7 raised sensitivity to 81% (specificity fell to 32%). A Scottish series of 46 upper limb lesions found 76% broke, and most broke with scores of 8 or less. Both groups suggest considering fixation at 7 in the arm (Evans et al, Clin Orthop Relat Res 2008;466:1279-1284, PMID 18357496; Hoban et al, JSES Int 2022;6:675-681, PMID 35813136).
- The length of cortex destroyed may predict a femoral fracture better than Mirels. A Dutch radiotherapy trial followed 110 femoral lesions, and only 14 broke. Only axial cortical involvement >30 mm and circumferential involvement >50% predicted fracture, and Mirels was not specific enough (p = 0.36). The authors concluded most classic risk factors overestimate fracture risk (Van der Linden et al, J Bone Joint Surg Br 2004;86:566-573, PMID 15174555).
- CT-based rigidity analysis predicts femoral fracture more accurately than Mirels and is not widely available. In 78 lesions left unfixed, CT rigidity analysis had sensitivity 100% vs 66.7% and specificity 60.6% vs 47.9% for Mirels (confidence intervals overlapped). A 2026 systematic review still calls Mirels a useful first-line screen and found no evidence to replace films and clinical judgment (Damron et al, Clin Orthop Relat Res 2016;474:643-651, PMID 26169800; Salaja et al, J Bone Oncol 2026;60:100790, PMID 42699563).