Patellofemoral Update focuses on the causes, prevention and treatment of patellofemoral disorders. The blog is sponsored by The Patellofemoral Foundation whose mission is to improve the care of individuals with anterior knee pain through targeted education and research. The Patellofemoral Foundation offers additional online education resources on its website.

BLOG: How to set a medial patellofemoral complex reconstruction to length

jack farr headshot
Jack Farr
Adam Yanke headshot
Adam B. Yanke

Anatomy of the medial knee is intricate. While most literature has highlighted the medial patellofemoral ligament, it is only one, yet very important, part of this complex.

The proximal portion of the medial patellofemoral complex (MPFC) extends from just distal to the adductor tubercle to the distal quadriceps tendon, constituting the medial quadriceps tendon-femoral ligament (MQTFL). As such, this portion of the MPFC may be reconstructed by soft tissue attachment to the medial distal quadriceps tendon, obviating the need for patella drilling (which carries a small, but worrisome risk of patella fracture), or to the patella itself as in more traditional medial patellofemoral ligament (MPFL) reconstruction.

For distal MPFC attachment to the patella, a subperiosteal tunnel may be created at the patellar attachment site for a sling or weave soft tissue fixation. When using a two-armed graft, the more proximal arm in the femoral socket is routed to the quadriceps tendon and the more distal into this subperiosteal patellar tunnel to avoid drilling into the patella itself.

John Fulkerson headshot
John P. Fulkerson

The native MPFC is subjected to low loads and has mean tensile strength of approximately 200 N. Once MQTFL reconstruction attachments have healed, they have adequate strength with no failures in an initial small series.

In either case, whether reconstructing the MPFL or the MQTFL or both, “setting the reconstruction to length” is key. In other words, once the MPFL or MQTFL reconstruction graft has been attached securely on one side, the surgeon must decide how to attach the graft on the other end. If the two arms are close together, they share nearly identical attachment length changes. If they are placed at the mid-waist of the patella and distal quadriceps, the length changes will be different as shown by Redler and colleagues in a laboratory study of the effect of various degrees of patellar alta. The MQTFL is proximal to the MPFL attachment site. As such, the two arms may need to be fixed in different degrees of flexion and thus, the rationale for femoral fixation first. If the graft is placed anatomically, length change in the graft should be minimal throughout flexion and extension with the goal of increasing laxity with flexion.

Locating an optimal femoral fixation point is key to creating proper length changes through range of motion: optimal is anatomometric (lax in flexion) or at minimum, isometric. Schöettle and colleagues popularized fluoroscopic parameters to select the femoral site. Several others have shown both in patients and cadavers, this may lead to suboptimal length changes in up to one-third of cases. Elias and colleagues have shown in lab the deleterious effects of even a 5 mm error and DeVries Watson and colleagues confirmed this clinically. While Servien and colleagues and McCarthy and colleagues showed no difference in relatively short outcomes of well-placed and suboptimally placed femoral attachments, there is concern that, while stable, an over-constrained patella will lead to chondral wear long term. (Remember that the Hauser procedure yielded “good stability outcomes,” yet led to delayed severe patellofemoral arthritis in many cases).

Therefore, triple-checking is recommended. First, select an optimal posterior skin incision site using Schöttle’s “point” as it is usually posterior to one’s instinct and many patients have adipose tissue preventing adductor tubercle palpation. After dissection to localize the tubercle, LaPrade and colleagues found the MPFL mean of 3.8 mm distal to the adductor tubercle, Fujino and colleagues found 10.6 mm, Wijdicks and colleagues found 8.9 mm, Smirk and Morris found 10 mm and Viste and colleagues found 10 mm. These differing measurements may be related to the specific knees, the extent of dissection (compare Tanaka and colleagues and Smigielski dissections of MPFC to earlier MPFL dissection) and the definition of the tubercle (a point, a circle, the broad adductor attachment). In light of these differences and the work of Elias and colleagues showing that 5 mm or greater error is concerning, radiographic initial placement may be used when the tubercle is broad and ill-defined and an adductor-based initial placement otherwise. These may still be off optimal placement (3.8 mm to 10.6 mm in the literature cited above). Remember too that the upper portion (MQTFL) of the MPFC originates at the most proximal end of the femoral MPFC origin, essentially at the distal aspect of the adductor tubercle itself.

First, there is confusing terminology in the literature. There is discussion of length changes when referring to the ligament. The ligament is essentially one length as there is minimal elasticity noted (without displacement forces). It does not change length, but rather when the attachment sites become closer (attachment sites distance/length is shorter in flexion), it becomes lax. When the distances become greater as the knee extends and the patella is out of the intercondylar notch, the ligament comes out to length. While Amis and colleagues measured tension at length of 5N, this is with displacement testing. In the clinical setting, the normal patella is displaceable one to two trochlear quadrant widths, so there is no tension in the MPFL until the patella reaches the end of the check-rein.

The sentence “Set the MPFL length to proper tension” should be abandoned.

The goal is anatomometric placement of a graft with attachments fixed when the attachment points are farthest apart, that is, the distance that is the longest should be “setting the graft to length.” As this is most sensitive to femoral attachment site, place a temporary suture through the patellar or quadriceps sling attachment site and test the provisional femoral placement (set by anatomy and/or fluoro) with the knee in 20° to 30° degrees for a patellar arm and 30° to 60° for a quadriceps arm. The suture is pulled “out to length.” As it is not attached at the patella/quadriceps, there is no tension between the attachment sites. Mark the exit of the suture at the sling(s) proposed attachment sites. Place the knee through full range of motion, monitoring the excursion of the suture. By marking it at the degree of knee flexion with maximal distance between the attachment sites, the mark on the suture should move out of the sling (anatomometric) or not move (isometric) as the knee moves from extension into flexion. Oka colleagues found that when the length changes were incorrect, the femoral pin was typically moved distally or posteriorly, at which point it would demonstrate the correct length change pattern.

Alternatively, if the graft has already been attached at the chosen anatomic site of the distal adductor tubercle, the graft(s) (one or two arms) may be brought under the VMO and through the proposed fixation tunnels (subperiosteal patellar and/or distal quadriceps/VMO tendon at level of the proximal patella) and then brought out to maximum length by flexion/extension of the knee with the patella position monitored arthroscopically (to assure no distortion of a normal tracking pattern), and sutured carefully at this point of maximum length. Thereby, excessive graft tightness may always be avoided, particularly with attention to Elias and colleagues’ recommendation of mild lateral displaceability of the medial complex reconstruction to prevent overtensioning.

The concept of “setting the reconstruction graft to length” plays an important role in preventing overtightening, which is potentially much more devastating than residual mild laxity.


Jack Farr, MD, is medical director of OrthoIndy Cartilage Restoration Center of Indiana and director, OrthoIndy Sports Medicine Fellowship, Indiana Orthopedic Hospital in Indianapolis. He is also professor of orthopedic surgery and a volunteer at Indiana University Medical School.

Adam B. Yanke, MD, PhD, is assistant professor of orthopedic surgery and assistant director of the Cartilage Restoration Center at Rush University Medical Center in Chicago.

John P. Fulkerson, MD, is a clinical professor of orthopedic surgery at the University of Connecticut School of Medicine and practices at Orthopedic Associates of Hartford in Farmington, Connecticut. He is also president of The Patellofemoral Foundation.


Disclosures: Farr, Yanke and Fulkerson report no relevant financial disclosures.


Amis AA, et al. Knee. 2003; doi:10.1016/S0968-0160(03)00006-1.

DeVries Watson NA, et al. Iowa Orthop J. 2015;35:13-19.

Elias JJ, et al. Am J Sports Med. 2006;doi:10.1177/0363546506287486.

Elias JJ, et al. Knee Surg Sports Traumatol Arthrosc. 2017;doi:10.1007/s00167-017-4799-9. [Epub ahead of print].

Fujino K, et al. Knee Surg Sports Traumatol Arthrosc. 2015;doi:10.1007/s00167-013-2797-0.

Fulkerson JP, et al. Arthrosc Tech. 2013;doi:1016/j.eats.2013.01.002.

LaPrade RF, et al. J Bone Joint Surg Am. 2007;doi:10.2106/JBJS.F.01176.

Matsushita T et al. Am J Sports Med. 2018;doi:10.1177/0363546517752667.

McCarthy M, et al. Iowa Orthop J. 2013;33:58-63.

Oka S, et al. Knee Surg Sports Traumatol Arthrosc. 2014;doi: 10.1007/s00167-014-3192-1.

Redler LH, et al. Arthroscopy. 2016; doi:10.1016/j.arthro.2017.08.256

Sanchis-Alfonso V, et al. Knee Surg Sports Traumatol Arthrosc. 2017;doi:10.1007/s00167-015-3905-0.

Schöttle PB, et al. Am J Sports Med. 2007;doi:10.1177/0363546506296415.

Servien E, et al. Am J Sports Med. 2011;doi:10.1177/0363546510381362.

Smirk C, et al. Knee. 2003;doi:10.1016/S0968-0160(03)00038-3.

Stephen JM, et al. Am J Sports Med. 2012;doi:10.1177/0363546512449998.

Tanaka MJ, et al. J Bone Joint Surg Am. 2016;doi:10.2106/JBJS.15.01182.

Tanaka MJ. Sports Med Arthrosc Rev. 2017;doi:10.1097/JSA.0000000000000143.

Tateishi T, et al. Knee Surg Sports Traumatol Arthrosc. 2011;doi:10.1007/s00167-010-1235-9.

Wijdicks CA, et al. J Bone Joint Surg Am. 2009;doi:10.2106/JBJS.H.00909.

Viste A, et al. Surg Radiol Anat. 2014;doi:10.1007/s00276-014-1270-1.

See more from Patellofemoral Update