Part I: Round table participants consider how the mechanisms of reconstruction can contribute to joint deterioration in some patients.
ORTHOPAEDICS TODAY INTERNATIONAL 2005; 8:1
Part II: [Controversy continues about the impact of ACL reconstruction and postop rehabilitation]
Anterior cruciate ligament (ACL) injury, if undiagnosed and untreated, has been associated with disability, meniscus tears and arthritis in studies of patients presenting with chronic ACL deficiency.
As noted in several early studies, the goals of ACL reconstruction are to prevent recurrent knee injury, symptoms of instability and pain, sports disability, and to prevent or delay deterioration of the joint (radiographic osteoarthritis, or OA). Although subsequent work has shown that ACL reconstruction can reduce anterior knee laxity and the risk of subsequent re-injury, reconstruction does not guarantee return to pre-injury knee function, nor necessarily prevent the development of OA on follow-up radiographs.
Dale Daniel and others have reported that patients who “cope” with an ACL-deficient knee, while they must accept a reduced level of activity, actually can have less severe degenerative changes on radiographs than patients who underwent either immediate or delayed reconstruction. Daniel’s report and the subsequent 10-year follow-up data showing similar findings using the same patient cohorts continue to stir controversy about whether ACL reconstruction actually prevents subsequent knee degeneration.
Daniel’s study was only the first of several reports to compare OA in reconstructed vs. unreconstructed knees. Most subsequent reports of prospective studies have reported either no difference in OA between reconstructed knees or that unreconstructed knees had slightly less OA than reconstructed knees (Fink, 2001; Fink, 1996; Fithian, 2002; von Porat, 2004; Lohmander, 2004). However, virtually every published work on this subject has bias in one direction or another, so it is difficult to draw definitive conclusions about the effects of ACL reconstruction on the risk of subsequent OA.
We have assembled an international panel of experts to answer the question, “Does ACL reconstruction cause knee arthritis?” and to discuss mechanisms by which reconstruction may actually contribute to joint deterioration. The question is not so much whether ACL surgery is efficacious but whether it has lived up to all our expectations for returning athletes to sports with a minimum risk of subsequent problems.
Such a critical appraisal will naturally make some people uncomfortable because of the challenge it represents, but it is worth considering whether further refinements of technique and rehabilitation would be useful. The discussion brought out several unanswered questions and identified areas in need of further research.
Donald Fithian, MD
Moderator
|
Moderator |
|
Donald Fithian, MD |
|
Participants |
|
|
Karl-Peter Benedetto, MD Stefan Lohmander, MD, PhD John Richmond, PhD |
Lynn Snyder-Mackler, PT Freddie H. Fu, MD |
Donald Fithian, MD: Surgeons, patients and the lay press often cite prevention of arthritis among the reasons for elective reconstruction of the ACL. Is there evidence in the published literature to support this?
For this discussion, we can define arthritis as the appearance of degenerative changes on radiographs — what has been described as radiographic OA and is used in most published literature of ACL outcomes.
Stefan Lohmander, MD, PhD: There is no support in the scientific literature that ACL reconstruction prevents the development of subsequent OA. The published reports are most often retrospective and observational, with considerable loss to follow-up, and most often do not use validated and standardized outcome measures for either radiographic change or patient-relevant change. There are no published randomized controlled trials (RCTs) comparing the outcome of surgery with no surgery; there are very few if any RCTs comparing different surgical techniques; there are very few if any studies in this area using blinded observers.
This is, in my view, a sad state of affairs for an intervention practiced on some 100,000 young patients each year in the United States, an intervention associated with a significant investment in rehab effort, morbidity and adverse events. Can we do better?
Lynn Snyder-Mackler, PT: I do not believe it is entirely accurate to conclude from Daniel’s studies that reconstructed patients should expect more arthritis than those who don’t undergo reconstruction. That study was published in 1994 and the subsequent 10-year follow-up paper published in 2002 tells a mixed story. Patients in that study who had worse symptoms and pathology wound up undergoing ACL reconstruction, so the groups were not really comparable.
Those who had early reconstruction had worse initial pathology; those who had late reconstruction had repetitive episodes of giving way prior to reconstruction and, we can presume, significantly more joint deterioration at the time of ACL reconstruction. Those who did not have reconstruction were either true copers or modified their activities such that they did not have episodes of giving way.
In addition, many of the surgeries performed in the study by Daniel and coworkers are obsolete. The postoperative rehabilitation was Draconian. I think all that we can say is that those with worse pathology had more arthrosis, and that bad surgery may be worse than no surgery.
Fithian: Actually, Lynn, this is not entirely true. In Daniel’s study, the early reconstructed patients had fewer meniscal tears and joint surface pathology than nonreconstructed patients who had arthroscopy during the study. You are correct, however, that late-reconstructed patients had the worst joint pathology of all groups at the time of surgery. Furthermore, the patients who had early surgery were younger and were more active (total level 1 and 2 sports hours) than patients who were treated conservatively.
Snyder-Mackler: I think we need a consensus that all individuals with ACL injury are not the same and that surgical decision-making is often influenced by factors that impact Don’s question. For example, patients’ reconstruction tend to be younger, more active and have worse pathology (eg, meniscal damage, chondral damage or concomitant other ligament injury) than those who do not undergo reconstruction.
Those who undergo late reconstruction have repetitive episodes of giving way and generally have extended the pathology. So the populations are not the same. I believe a true randomized trial would be impossible to do in the United States and probably elsewhere.
I have spent the better part of 10 years studying true copers – those who can compensate fully for ACL rupture and even decades after injury have no OA. They are remarkably rare. Most people are noncopers; a fair number have some potential to compensate but do better if they have specific rehabilitation. (Fitzgerald, Axe and Snyder-Mackler 2000).
Lohmander: There is a randomized trial ongoing in Lund, with outcome at two and five years stratified for age and activity level.
John Richmond, PhD: The meniscus remains the most important protector of the joint surfaces in an ACL-injured joint. Any treatment that might be employed that increases the chance of future meniscal tearing and/or meniscectomy will increase the risk of degenerative changes within the knee. To support this there are numerous studies that link meniscectomy in the ACL-injured or reconstructed knee.
To further support meniscal preservation, our long-term prospective study (Am J Sports Med 2002) showed less than 10% of ACL-reconstructed patients with intact menisci had any radiographic signs of degeneration at 10 years postsurgery, while maintaining a high activity level (mean Tegner score of 6.8). Since meniscal injury is low in the ACL reconstructed population, I believe that this supports the premise that ACL-reconstruction reduces the likelihood of late degenerative joint disease (DJD).
With that as a starting point, there are many factors that contribute to risk DJD, perhaps the greatest being the postoperative immobilization that we employed in the early and mid-1980s. Nothing is more certain to ulcerate the articular surface than a hemarthrosis followed by several weeks of immobilization. Add weeks of aggressive quad rehabilitation, followed by return to high impact sports, and you have a formula for inducing DJD. This alone may explain the higher DJD rate in the reconstructed patients in the landmark study of Daniel and Fithian (Am J Sports Med 1994).
While there has been no prospective, randomized controlled trial to look at reconstruction vs. nonoperative treatment, such a study would not be unethical if meniscal injury is more likely without surgery.
Karl-Peter Benedetto, MD: I do not believe we can inform patients that the indication on ACL reconstruction is based on arthritis prevention. There is no single prospective study that compares patients having sustained not only an ACL injury but also the same complete pathology of the knee. Unfortunately, I strongly believe that such a study never will be available.
Also our study included just patients with clinically and arthroscopically documented rupture of the anterior cruciate ligament. No one was looking at the morphology or associated pathology of the knee. Alignment of the opposite healthy knee was never measured either by long-standing films or by weight-bearing X-rays, full extension and 40° reflexion. Whether a patient had an additional medial collateral ligament sprain has not been documented.
We only considered whether there was a complete tear of the MCL. At that time we also were not looking to determine whether those patients had an additional posterolateral injury. The slope of the tibial plateau also has been not focused on, and we do not know whether those patients had full extension or slight hyperflexion. Also a bone bruise was not documented at that time as all patients just had undergone arthroscopy.
I’m aware that there are some limitations in our study, but I’m doubtful whether a prospective study will ever come up comparing apples to apples regarding pathology of the knee.
What we certainly could find out and document was that an intact meniscus plays an important role in those patients who had undergone conservative or operative treatment. At least those patients who were having secondary partial meniscectomy had a significantly higher rate of osteoarthritic changes.
It just happened that those patients who avoided rotation while loading the knee had significantly fewer secondary meniscus lesions and therefore less OA.
Freddie H. Fu, MD: There is evidence in the literature that both ACL injury and meniscal injuries are associated with degenerative changes of the knee joint (Gelber, Ann Intern Med. 2000; Gillquist, Sports Med 1999). However, it remains unclear to what degree surgical ACL reconstruction can delay OA of the knee.
In a seven-year follow-up study including 72 patients undergoing surgical ACL reconstruction, Johma et al demonstrated that patients undergoing surgery of acute ACL injuries had fewer radiographic signs of OA than patients undergoing surgery for chronic ACL insufficiency (Johma, Clin Orthop 1999), hence indicating the potential benefit of early ACL reconstruction. However, more research is needed to investigate the long-term effect of ACL injury as well as ACL reconstruction on the development of late OA.
Fithian: The role of meniscal damage in the development of DJD after ACL injury is well established. Given that surgery is effective at preventing subsequent meniscus injury, why might arthritis appear to be more common in reconstructed than in nonreconstructed knees?
Richmond: The lowest rate of radiographic DJD is in those patients treated nonoperatively and who do not buckle. They avoid meniscal resection, hemarthrosis and usually high-demand sports, all contributors to potential articular surface deterioration. If all our patients were willing to adjust their lifestyles in that way, only those patients who had instability with ADL would benefit from ACL surgery.
Snyder-Mackler: I do not agree that the literature shows more arthritis in those who were reconstructed when comparable cohorts are compared.
Some experts have suggested that the high prevalence of bone bruising with ACL rupture may represent occult injury to the subchondral bone that ultimately leads to degeneration. Perhaps even small damage to the menisci that happens in most ACL ruptures is enough to alter the hoop stresses and thereby reduce the protective effects of the meniscal cartilage.
Lohmander: There is very strong face validity in retaining a functional meniscus, especially in the ACL-injured knee. So why isn’t ACL reconstruction — that in observational studies appears to provide some protection against subsequent meniscal injury — associated with less OA? I’m afraid we have no good answer at this time, other than to note that this again questions the rationale of some of our current practice.
Benedetto: ACL reconstruction does not guarantee that there will not be a subsequent meniscus lesion if the patient is going on with high-risk pivoting sports. Furthermore, it is unclear whether the secondary meniscus lesion may not have happened if the patient not had an ACL injury.
Fu: The knee that suffers an ACL rupture is by no means a normal knee. In fact, it is a knee that has suffered a large amount of trauma and may have many associated injuries. To validate the above statement, matched cohorts with adequate controls are needed to show if reconstructing the ligament indeed causes the arthritis.
See the March/April issue for part II of the virtual round table, “Does ACL reconstruction cause knee arthritis?” as well as a complete list of references.