High rate of fracture in long-bone metastasis: Proposal for an improved Mirels predictive score

Abstract

Introduction: Pathologic fracture is the most feared complication in long-bone metastasis. Various radiographic tools are available for identifying at-risk patients and guide preventive treatment. The Mirels score is the most frequently studied and widely used, but has been criticized, many patients not being operated on until the actual fracture stage. We therefore conducted a French national multicenter prospective study: (1) to determine the proportion of patients operated on at fracture stage versus preventively; (2) to compare Mirels score between the two; and (3) to identify factors for operation at fracture stage according to Mirels score and other epidemiological, clinical and biological criteria.

Hypothesis: Simple discriminatory items can be identified to as to complete the Mirels score and enhance its predictive capacity.

Material and methods: A non-controlled multicenter prospective study included 245 patients operated on for non-revelatory long-bone metastasis, comparing patients operated on for fracture versus preventively according to body-mass index (BMI), ASA score, Katagiri score items and the 4 Mirels items.

Results: One hundred and twenty-six patients (51.4%) were operated on at fracture stage: 106 (84.1%) showed high risk on Mirels score (score>8), and 15 (11.9%) moderate risk (score=8). On multivariate analysis, 4 independent factors emerged: in increasing order, advanced age (OR=1.03; 95%CI 1.01-1.06), VAS pain score>6 (OR=1.47; 95%CI 1.02-2.11), WHO grade>2 (OR=2.74; 95%CI 1.22-6.15), and upper-limb location (OR=5.26; 95%CI 2.13-12.84).

Discussion: The present study confirmed that more than half of patients with long-bone metastasis are operated on at actual fracture stage, in agreement with the literature. Several studies highlighted the weakness of the Mirels score as a predictive instrument. Comparison between preventive and fracture-stage surgery showed that upper-limb location, intense pain, advanced age and impaired functional status were associated with fracture-stage surgery, and should be taken into account alongside the original Mirels criteria. This improved scoring instrument remains to be validated in a prospective study.

Level of evidence: IV, prospective cohort study without control group.

Keywords: Impending fracture; Long-bone metastasis; Mirels score; Pathological fracture.