Diagnosis and treatment of patellofemoral disorders must be individualized

Patellofemoral Update

After reading an article from Pagenstert and colleagues on lateral retinacular release vs. lengthening, I was compelled to respond with a letter to the editor. I was surprised and disappointed that an orthopedic surgeon would design a study using a 22-year-old technique that was said to assure an “adequate” release (the “90° turn-up” test), even cutting the vastus lateralis tendon if necessary to achieve this goal, and leave it unrepaired. Severing the vastus lateralis tendon is known to cause iatrogenic medial patellar subluxation and permanent disability in a high proportion of patients. Indeed, about one third of their release-only patients have this totally predictable fate.

In the authors’ reply to my letter, I was relieved to learn that they were ignorant of the permanent harm and disability their study would cause to the release-only cohort. However, I was disappointed to find out they had based their rigid adherence to the 90° turn-up test only on a diagram in a chapter of a 2008 textbook. Apparently, they had neither read nor heeded the advice within the text of that very same chapter, which reads, “The proximal release is not extended into the muscle fibers of the vastus lateralis or quadriceps tendon.”

Furthermore, they said there was no information available in international publications warning about over-release by cutting the vastus lateralis tendon. Yet a cursory search of Medline revealed several articles containing this information predating 2008. For the sake of brevity, I only cite a 1995 study article by Marumoto and colleagues published in the American Journal of Sports Medicine.

Many members of the International Patellofemoral Study Group encouraged me to write this review to:

I consider these issues logically simple. After proper nonoperative treatment has failed and a tight lateral retinaculum (LR) is the cause of symptoms, then a LRR is logical. Even if it is not the only cause of symptoms, it is a logical first step. If the LR is not tight, then don’t release it. A satisfactory LRR should not cut the muscle or tendon of the vastus lateralis. Remembering that the goal of a LRR is to normalize a tight LR, then a reasonable endpoint is a medial glide test of more than one quadrant, or more than one fingerbreadth. Another reasonable endpoint is a 60° tilt-up test. It is safe to release the LR distally to the joint line if necessary to achieve these endpoints.

The diagnosis and treatment of all patellofemoral disorders must be individualized. To quote Scott F. Dye, MD, It is an “intellectually and clinically constrained notion that a single, simple, structural, surgical ‘90° turn-up test – one size fits all approach’ works in a system as complex as the living human knee.” By individualizing treatment, the LRR also can achieve satisfactory results in approximately 60% of patients with isolated patellofemoral osteoarthritis. Likewise, in certain patients with recurrent patellar dislocation, a LRR can achieve excellent long-term results.

Finally, the fault for the publication of the Pagenstert and colleagues study does not lie solely with the authors. Institutional review boards were established to avoid causing permanent harm to patients during clinical studies, yet this study was allowed. In addition, there is another factor at work – the tyranny of evidence-based medicine. Journal editors and their appointed reviewers seem so eager to publish prospective double-blinded comparative studies having higher levels of evidence that clinical judgment, experience and historical perspective are forgotten. This is not a new idea. Others have noticed this intrusion of evidence-based clinical guidelines and protocols, including Latov and Hieb.

I thank Healio.com for helping to widen the audience for this discussion and the members of International Patellofemoral Study Group and the Patellofemoral Foundation for their encouragement and support.

Gasser S. Arthroscopic lateral release of the patella with radiofrequency ablation. In: Jackson DW, ed. Master Techniques in Orthopaedic Surgery: Reconstructive Knee Surgery. 3rd ed. Philadelphia: Lippincott, Williams & Wilkins; 2008;1-11.

Hieb LD. Why your doctor is out of date. Journal of American Physicians and Surgeons. 2005;10(1):69-70. Available online at: www.jpands.org/vol16no3/hieb.pdf

Latov N. Evidence-based guidelines: Not recommended. Journal of American Physicians and Surgeons. 2005;10(1):18-19. Available online at: www.jpands.org/vol10no1/latov.pdf.

Marumoto JM. A biomechanical comparison of lateral retinacular releases. Am J Sports Med. 1995;23(2):151-155.

Merchant AC. Arthroscopy. 2013;doi:10.1016/j.arthro.2013.01.002.

Pagenstert G. Arthroscopy. 2012;doi:10.1016/j.arthro.2011.11.004.

Pagenstert G. Arthroscopy. 2013;doi:10.1016/j.arthro.2013.01.001.

Monday, December 23, 2013

Patellofemoral Update

Trochleoplasty for patients with recurrent patellar instability has been debated the past few years. The procedure is becoming more common in Europe, while orthopedic surgeons in the United States have a more cautious approach, although a few surgeons are doing the procedure. One can ask if there is a need for trochleoplasties when there are other well-established procedures.

Many surgeons are excited about medial patellofemoral ligament (MPFL) reconstructions. Some studies with long-term follow-up seem to have good results; however, it is not a solution for all patients with patellar instabilities. We have to be aware of the underlying pathomorphology. In a recent meta analysis, Shah and colleagues found a complication rate of 26% after MPFL reconstruction. A major complication was postoperative instability with recurrent apprehension, which was 32% of all complications. Howells and colleagues found that 15 of 25 patients with persistent recurrent symptoms after reconstruction of the MPFL had moderate trochlear dysplasia. Patients with severe trochlear dysplasia already had been excluded.

Wagner and colleagues found a negative relationship between the degree of trochlear dysplasia and outcome after MPFL reconstruction. The situation seems to be that when the native MPFL ruptures, without any major trauma, it is likely an injury caused by continuously stress on the ligament. By making the ligament stronger, we can introduce abnormal joint forces leading to either pain at the insertion side, osteoarthritis (OA) or patella stress fractures.

 

Alternatively or in combination with an MPFL reconstruction, we can transfer the tibial tubercle either medially to reduce the tibial tubercle–trochlear groove distance (TT-TG) distance and/or distally to correct a patella alta. However, it is difficult to ignore the studies from Carney and colleagues and Nakagawa and colleagues, who observed that good to excellent results after transfer of the tibial tubercle declined after 10 years to 15 years, with some patients having increasing pain. Nakagawe and colleagues found radiological osteoarthritic changes grade 2 or worse in 42% of patients after Elmslie-Trillat procedures (no anteriorization at time of medial tibial tubercle transfer).

This brings us back to the question: is there any indication for the trochleoplasty procedure? The procedure is based on the principle of restoring abnormal anatomy in cases with trochlear dysplasia. We know trochlear dysplasia is a main factor for patellar instability and apparently it seems logical to do the trochleoplasty procedure. Trochlear dysplasia is characterized by overstuffing too much bone in the trochlea, resulting in a flat of shallow configuration in the most proximal part of the trochlea. This configuration has impact on many factors. Mostly, the trochlear groove is medialized and this causes an increase of the TT-TG distance.

It is important to notice that in these cases, the abnormal TT-TG is not caused by an external placed tibial tubercle, but an asymmetric trochlea because of a medialized trochlear groove. Furthermore, the patella tilt is caused by overstuffing of the trochlea, forcing the patella to articulate on it lateral facet simply because the patella has no groove to be contained. When a patient has trochlea dysplasia, you will find the trochlea inclination angle is reduced, meaning there is a deficiency of the lateral part of the trochlea which is supposed to give osseous support to the patella. These factors can all be normalized by doing a trochleoplasty and the procedure also has a positive impact on the sulcus angle and eventual patella alta. Depending on the operative trochleoplasty technique, the TT-TG distance can be reduced by 5 mm to 10 mm.

Most surgeons have a reasonably skeptical approach on the trochleoplasty procedure based on several sound arguments. First, the procedure is technically demanding and should only be done by a selected and dedicated group of surgeons who have special knowledge about the biomechanics of patellar instability and have a routine for more standard patellar stabilizing procedures, such as MPFL reconstructions and osteotomies of the tibial tubercle.

Second, this is a procedure that involves the osteochondral interface. Radin and Rose showed this is a delicate structure and impacts on this can potentially lead to arthritis. Schöttle and colleagues studied the cartilage viability after the Bereiter trochleoplasty and found tissue in the trochlear groove remained viable, with retention of distinctive hyaline architecture and composition. However, some pathological changes were found, as lacunae were seen progressing from the subchondral bone through the calcified layer and, in some cases, even into the basal layer of the cartilage. This can be a sign of cartilage degeneration giving profound impact of subchondral bone disruption in mostly young patients. It was hypothesized that the cartilage flake probably undergoes microfractures while being pressed down into the new groove, resulting in these lacunae and cluster formation.  Pathologic changes were found in the area just beneath the subchondral bone, therefore, showing processes characteristic of epiphyseal fracture healing.

It was concluded that even though a few minor changes in the calcified layers were identified, the results of the microscopic findings in conjunction with the clinical and radiological results seen at 2-year and 5-year follow-up after trochleoplasty could be expected to persist long term. However, further histological studies with a longer follow-up are recommended. If subchondral bone and cartilage stiffness occur in the trochleoplasty area, it is likely that this might eventually lead to OA.

Third, postoperative detachment of the osteochondral flap or chondrolysis may have serious consequences. Fortunately there have not been any reported cases yet. Today, about 20 case series have now demonstrated good short- to mid-term results of the trochleoplasty procedure. In the past three studies, the trochleoplasty has been combined with an MPFL reconstruction. The obtained results seems to be encouraging in patients with trochlear dysplasia (Dejour grade B-D), resulting in consistent statistically significant improvement in all applied knee function scoring systems. Ntagiopaulos and colleagues found no radiologically signs of OA (follow-up: 7 years; range: 3 years to 9 years). This is contrary to Knoch and colleagues, who reported radiologic osteoarthritic changes grade 2 or worse in 30 % (follow-up: 8 years; range: 4 years to 14 years). In this study, patellofemoral pain became worse in 33% of patients and improved in 49% of patients, but the median Kujala score was 94.

Many factors are involved in anterior knee pain. We should pay attention to the study from Keser and colleagues, who found a significant correlation between trochlear dysplasia and anterior knee pain. Also it is interesting that Goutallier and colleagues successfully used the recession trochleoplasty procedure in cases with trochlea dysplasia, where the patella already had been stabilized successfully but the patient continued to have pain.

We know trochlear dysplasia predisposes patients to OA and this is whether or not the patella is unstable. Albee elevation trochleoplasty increases the forces in the patellofemoral joint and causes secondary OA.  Therefore, it is likely to hypothesize that when the forces in an overstuffed dysplastic trochlea are reduced, by doing a deepening or depression trochleoplasty, the risk of OA might be reduced.

In addition to the above mentioned factors, there are other factors involved. For example, the procedure is technically demanding. There is a risk of arthrofibrosis and there is spare data on the long-term follow-up. The differences observed in postoperative stiffness and pain may be attributed to different surgical techniques, rehabilitation protocols and indications. The persistence of instability and the need for reoperation after trochleoplasty procedures are rarely recorded.

In my opinion, we are only at the dawn and hopefully new patient-related outcomes scores, such as the Norwich Patella Instability score and Banff patellar instrument, will be useful instruments to help obtain more precise and comparative data in the future. However, it is difficult to neglect the good clinical results obtained until now. Obviously, trochleoplasty is indicated in some cases, however, is still difficult to say if it is a salvage procedure or if it is also a procedure for primary cases with perhaps less pronounced degrees of trochlear dysplasia. We need to sort out the differences between the types of trochleoplasty techniques including more data on arthroscopic techniques. 

Shah JN. Am J Sports Med. 2012; doi:10.1177/0363546512442330.

Howells NR. J Bone Joint Surg Br. 2012; doi:10.1302/0301-620X.94B9.28738.

Wagner D. Knee Surg Sports Traumatol Arthrosc. 2013;doi:10.1007/s00167-012-2015-5.

Carney JR. Long-term evaluation of the Roux-Elmslie-Trillat procedure for patellar instability: a 26-year follow-up. Am J Sports Med. 2005;33(8):1220-1223.

Nakagawa K. Deterioration of long-term clinical results after the Elmslie-Trillat procedure for dislocation of the patella. J Bone Joint Surg Br. 2002;84(6):861-864.

Radin EL. Role of subchondral bone in the initiation and progression of cartilage damage. Clin Orthop Relat Res. 1986;(213):34-40.

Schöttle EB. Cartilage viability after trochleoplasty. Knee Surg Sports Traumatol Arthrosc. 2007;15(2):161-167.

Blønd L. Knee Surg Sports Traumatol Arthrosc. 2013;doi:10.1007/s00167-013-2422-2.

Banke IJ. Knee Surg Sports Traumatol Arthrosc. 2013;doi:10.1007/s00167-013-2603-z.

Nelitz M. Am J Sports Med. 2013;doi:10.1177/0363546513478579.

Ntagiopoulos PG. Am J Sports Med. 2013;doi:10.1177/0363546513482302.

Knoch F. Trochleaplasty for recurrent patellar dislocation in association with trochlear dysplasia. A 4- to 14-year follow-up study. J Bone Joint Surg Br. 2006;88(10):1331-1335.

Keser S. Knee Surg Sports Traumatol Arthrosc. 2008; doi:10.1007/s00167-008-0571-5.

Goutallier D. Retro-trochlear wedge reduction trochleoplasty for the treatment of painful patella syndrome with protruding trochleae. Technical note and early results. Rev Chir Orthop Reparatrice Appar Mot. 2002;88(7):678-685.

Smith TO. Knee Surg Sports Traumatol Arthrosc. 2013;doi:10.1007/s00167-012-2359-x.

Hiemstra LA. Am J Sports Med. 2013; doi:10.1177/0363546513487981.