Send to

Choose Destination
Am J Sports Med. 2012 Aug;40(8):1871-9. doi: 10.1177/0363546512449998. Epub 2012 Jun 22.

The medial patellofemoral ligament: location of femoral attachment and length change patterns resulting from anatomic and nonanatomic attachments.

Author information

Mechanical Engineering Department, Imperial College London, London, United Kingdom.



Incompetence of the medial patellofemoral ligament (MPFL) is an integral factor in patellofemoral instability. Reconstruction of this structure is gaining increasing popularity. However, the natural behavior of the ligament is still not fully understood, and crucially, the correct landmark for femoral attachment of the MPFL at surgery is poorly defined.


To determine the length change pattern of the native MPFL, investigate the effect of nonanatomic femoral and differing patellar attachment sites on length changes, and recommend a reproducible femoral attachment site for undertaking anatomic MPFL reconstruction.


Descriptive laboratory study.


Eight cadaveric knees were dissected of skin and subcutaneous fat and mounted in a kinematics rig with the quadriceps tensioned. The MPFL length change patterns were measured for combinations of patellar and femoral attachments using a suture and displacement transducer. Three attachments were along the superomedial border of the patella, and 5 femoral attachments were at the MPFL center and 5 mm proximal, distal, anterior, and posterior to this point. Reproducibility of attachment sites was validated radiographically.


The femoral attachment point, taking the anterior-posterior medial femoral condyle diameter to be 100%, was identified 40% from the posterior, 50% from the distal, and 60% from the anterior border of the medial femoral condyle. This point was most isometric, with a mean maximal length change to the central patellar attachment of 2.1 mm from 0° to 110° of knee flexion. The proximal femoral attachment resulted in up to 6.4 mm mean lengthening and the distal attachment up to 9.1 mm mean shortening through 0° to 110° of knee flexion, resulting in a significantly nonisometric graft (P < .05).


We report the anatomic femoral and patellar MPFL graft attachments, with confirmation of the reproducibility of their location and resulting kinematic behavior. Nonanatomic attachments caused significant loss of isometry.


The importance of an anatomically positioned MPFL reconstruction is highlighted, and an identifiable radiographic point for femoral tunnel position is suggested for use intraoperatively.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center