Description

Van der Linden et al used a modification of Mirels score for predicting the risk of pathologic fracture in a femur with metastatic cancer. This can help identify a patient who may benefit from more aggressive management. The authors are from Leiden University in The Netherlands.


Patient selection: metastasis to the femur

 

Parameters:

(1) anatomic location within the femur

(2) degree of pain using a Likert scale from 0 (no) to 10 (worst imaginable)

(3) radiographic appearance

(4) percent of cortical circumference involved

 

Parameter

Finding

Points

anatomic location

distal femur

2

 

shaft

2

 

proximal femur

3

pain

0 or 1

0

 

2, 3 or 4

1

 

5, 6 or 7

2

 

8. 9 or 10

3

radioagraphic appearance

blastic

1

 

lytic

3

 

mixed

2

percent of cortical circumference involved

< 33%

1

 

33 - 67%

2

 

> 67%

3

 

total score =

= SUM(points for all 4 parameters)

 

Interpretation:

• minimum score: 4 (in the footnote to Table II the minimum score is given as 5, but this assumes that the minimum pain score is 1)

• maximum score: 12

• A score < 9 was not associated with pathologic fracture.

• A score >= 9 may or may not be associated with fracture.

 

Performance:

• The score is 100% sensitive but only has a specificity of 13%. This can result in an overestimate of the risk for fracture.

 

Risk factors better predictive of pathologic fracture:

(1) axial cortical involvement > 30 mm

(2) circumferential cortical involvement > 50%


To read more or access our algorithms and calculators, please log in or register.