Chapter 72 Stiffness
Prevention and Treatment
Anterior cruciate ligament (ACL) reconstruction has evolved into a highly successful procedure, with recent studies reporting good outcomes in more than 90% of the patients.1–3 Although the patellar tendon continues to be the most popular type of graft in North America, the quadruple hamstrings graft is emerging as the “other gold standard.” Multiple comparison studies document no difference in the outcomes between the two types of graft.4–6 Perhaps the two main reasons for the equal success rates are advances in the fixation methods of tendon grafts as well as an increase in our understanding of the biology of healing of ACL grafts.
However, complications following ACL reconstruction do occur, and motion loss is one of the most common. The reported incidence is between 2% and 11%,7,8 and its management can be quite frustrating for both the patient and the surgeon. In this chapter we will discuss factors that are associated with development of arthrofibrosis after ACL reconstruction, propose strategies to avoid this complication, and present treatment options.
Etiology
Genetic Predisposition
The tendency of certain patients to develop excessive scarring following trauma or surgery is well known. A history of arthrofibrosis from previous surgery or trauma should alert the surgeon. The exact reason behind this excessive connective tissue proliferation is not known. Several mechanisms have been proposed. Two cytokines, the platelet-derived growth factor-ß (PDGF-ß) and the transforming growth factor-ß (TGF-ß), have a central role in the healing process. Over-expression of TGF-ß has been associated with unresolved inflammation and fibrotic events.9 In animal models, neutralization of its isoforms (TGF-ß-1 and TGF-ß-2) has reduced scarring.10 Interestingly, exogenous addition of the isoform TGF-ß-3 achieved the same effect. More recently, a possible association between arthrofibrosis and certain human leukocyte antigen (HLA) types has been suggested. Skutek et al11 performed HLA typing in a pool of patients who developed primary arthrofibrosis following ACL reconstruction. Patients with secondary reasons for arthrofibrosis, such as prolonged immobilization, infection, or other surgical complications, were excluded. In their patient group, the phenotype HLA-Cw*08 was detected significantly more often compared with the control group. Additionally, in the same group of patients the phenotypes HLA-Cw*07 and HLA-DQB1 were detected significantly less frequently compared with the control group. The importance of the fat pad in the fibrotic process following surgical trauma to the knee has long been recognized.12 Adipose tissue is capable of releasing cytokines in an endocrine, autocrine, and paracrine manner.13 Ushiyama et al14 demonstrated that the infrapatellar fat pad produces a variety of growth factors and pro-inflammatory cytokines such as basic fibroblast growth factor (bFGF), vascular endothelial growth factor (VEGF), tumor necrosis factor alpha (TNF-α) and interleukin (IL)-6, much like an endocrine organ. Additionally, Murakami et al15 found elevated concentrations of PDGF-ß and TGF-ß in the fat pad after ACL reconstruction. From this research, it is evident that the knee fat pad is capable of mounting a response much like the articular synovium and enhances the inflammatory reaction to the surgical insult. Preservation and minimal disturbance of this structure during ACL reconstruction may minimize excessive scarring and motion loss.
Surgical Factors
The timing of surgery following acute ACL tear and its association with motion loss have been the subject of many studies and an issue of controversy. Although some authors believe that early reconstruction (i.e., within 2 weeks from the injury) does not affect the ultimate knee range of motion (ROM) that the patient achieves,16–18 it seems that the majority of surgeons favor a delay varying from 1 to 3 weeks to allow resolution of the acute inflammation and restoration of ROM.7,19–21 Furthermore, even in studies where no motion complications were documented after early intervention, no advantage in terms of the outcome of the reconstruction was identified. Shelbourne et al22 found that patients who had delayed ACL reconstruction at a mean of 40 days after injury had earlier return of quadriceps strength and were able to progress to sport-specific rehabilitation sooner than patients who had their knee reconstructed early at a mean of 11 days after injury. It must be realized, however, that significant variability exists among patients in the intensity of the observed inflammation following acute ACL tear, and this is at least partially related to the energy of injury. Rather than relying on timetables or strictly followed protocols, we believe that the decision on surgical timing should be based on clinical observation of subsidence of the posttraumatic inflammation, restoration of ROM, and normalization of gait. The patient is advised to follow a classic RICE (rest, ice, compression, and elevation) regimen and is referred to physical therapy (i.e., prehab). This allows the patient to become familiar with the physical therapist and the exercises and equipment that will be used after the surgery, become emotionally prepared, and make other necessary arrangements for the upcoming surgery. These are all factors that contribute to correct patient education and, in our opinion, enhance compliance and chances for a successful outcome. The issue of timing becomes even more important when the ACL tear is combined with other ligament injuries. Associated medial collateral ligament (MCL) injury has been recognized as a combination that is particularly prone to loss of motion. The location of the MCL tear influences the return of motion, and patients with proximal (above the joint line) injury are more likely to experience motion loss postoperatively.23 Low-grade injuries are successfully treated with an initial period of functional bracing to allow healing of the MCL. This waiting period is used to prehab the knee before the ACL reconstruction. In cases of associated severe grade III MCL or multi-ligamentous injury, priority is generally given to early restoration of knee stability. Attention during surgery is given to anatomical repair of the MCL, especially in injuries where the ligament is avulsed from its tibial attachment. Fixation of the superficial MCL near the joint line will prevent the normal posterior displacement of the ligament during knee flexion and will result in postoperative loss of flexion. Generally, in cases of multi-ligament reconstruction, slow return of motion should be anticipated and treated aggressively in the postoperative period.
Graft malpositioning due to nonanatomical tunnel placement either in the tibia or the femur is one of the most common errors in the surgical technique of ACL reconstruction and is believed to be responsible for a high percentage of graft failures. Noyes and Barber-Westin24 reviewed a series of 114 consecutive ACL revisions and found that 30% involved cases of improper graft placement. Errant tunnel placement subjects the graft to excessive strains and, depending on its stiffness, can lead either to loss of motion or plastic deformation and elongation of the graft. The normal ACL has a complex anatomy, which the current, essentially cylindrical grafts are unable to reproduce. In theory, during reconstruction, an attempt is made to place the graft in an isometric position and at the same time avoid impingement of the graft on the surrounding anatomical structures. Hefzy et al25 found that of the two fixation points of the graft, the one that most affects graft isometry is the femoral. They identified a zone near the center of the femoral insertion of the normal ACL that was the most isometric, as defined by a length change of 2 mm or less. The axis of this zone has a nearly vertical orientation with the knee extended. Its width varies from 3 to 5 mm and tapers from proximal to distal. Anterior placement of the femoral tunnel was one of the most common errors during ACL reconstruction. Placing the graft too far anteriorly results in a graft that is lax in extension and tight in flexion. Often the end result is a joint with limited postoperative flexion. Attempts to regain full flexion in such a knee, as during postoperative rehabilitation, will subject the graft to excessive strains and compromise its integrity. Recognizing the problems associated with anterior femoral tunnel placement, some surgeons use the over-the-top position. Attachment of the graft in the over-the-top position essentially reverses the situation and results in a graft that is tight in extension but lax in flexion. Currently most surgeons prefer to place the entrance of the femoral tunnel high in the notch at the 10- or 2–o’clock position, at the proximal end of the zone described by Hefzy et al,25 where this zone is wider. Depending on the graft choice, a 1- to 2-mm back wall is left to allow safe fixation of the graft. We use three methods to identify the center of the femoral tunnel and place the femoral guide pin. This point may be selected using freehand technique, femoral over-the-top offset guides, and (less frequently) an isometer.
In the study of Hefzy et al,25 altering the tibial attachment site had a smaller effect on isometry of the graft. Nevertheless, the tibial attachment site is important if one is to avoid impingement of the graft on the intercondylar roof or the posterior cruciate ligament (PCL). Sapega et al26 noted that even in the normal ACL, only a few fibers are truly isometric (length change of 1 mm or less) over the full ROM of the knee. In their cadaveric study the fibers of the anteromedial bundle of the ACL demonstrated the least deviation from isometry. Many authors27–30 shared the same view and placed the tibial tunnel in the anteromedial footprint of the tibial ACL insertion in an attempt to re-create these fibers. However, anterior placement of the tibial tunnel anterior to an imaginary line and tangential to the intercondylar roof (Blumensaat’s line) with the knee in full extension results in roof impingement of the graft and loss of extension. Howell and Taylor31 found poor results in patients with impinged grafts in terms of extension loss and graft failure. They reported a 100% failure rate in the severely impinged grafts where the entire articular opening of the tibial tunnel was anterior to the slope of the intercondylar roof. The tibial insertion of the native ACL has an eccentric anterior extension32 that cannot be re-created without impingement of the graft in the intercondylar roof. Jackson and Schaefer33 reported a series of patients who presented with loss of extension following ACL reconstruction with patellar tendon autograft. Second-look arthroscopy revealed the presence of a fibrous nodule, reminiscent of a “cyclops,” in front of the ACL graft, which blocked extension. Excision of the nodule resulted in improvement of knee extension in all patients. Marzo et al34 described the same pathology in a group of patients following ACL reconstruction with patellar tendon or hamstring autograft. In both instances it was believed that the fibrous nodule was the result of anterior tibial tunnel placement and impingement of the graft in the intercondylar roof. Romano et al35