Controversy continues about the impact of ACL reconstruction, postop rehab

In part II of this round table, participants discuss the data about ACL reconstruction, degenerative bone disease.

For this round table, Donald C. Fithian, MD, gathered an international panel of experts to answer the question, “Does ACL reconstruction cause knee arthritis?” Last month, the panelists looked at the data regarding whether ACL reconstruction prevents the development of subsequent OA and agreed that randomized controlled trials are needed.

Participants also reviewed data indicating that both ACL injury and meniscal injuries are associated with degenerative changes of the knee joint. It remains unclear, however, to what degree surgical ACL reconstruction can delay OA.

Part II considers additional questions on this topic.

MODERATOR Donald Fithian, MD Fellowship Director San Diego Knee and Shoulder, and Kaiser Permanente-San Diego Co-Director, Total Joint Registry San Diego, California, USA PARTICIPANTS Karl P. Benedetto, MD ESSKA President Department of Trauma Surgery and Sports Medicine Academic Hospital Feldkirch, Austria Stefan Lohmander, MD, PhD, Lund University, Department of Orthopedics Department of Orthopaedic Surgery Lund, Sweden Lynn Snyder-Mackler, PT, Professor of Physical Therapy University of Delaware Newark, Delaware, USA Freddie H. Fu, MD Professor of Orthopedic Surgery, Chairman of Orthopedic Surgery, University of Pittsburgh School of Medicine and University of Pittsburgh Medical Center, Pittsburgh, Member of Orthopaedics Today International Editorial Board John C. Richmond, MD, Chairman of Orthopedic Surgery, New England Baptist Hospital, Boston

Donald Fithian, MD: Is there a relationship between activity level and the degree of arthritic change on radiographs at follow-up?

Karl P. Benedetto, MD: There is some relation between activity level at follow-up and the degree of arthritic change on radiographs. But more important than the activity level is the type of sports that the patient is doing following his or her injury. Those doing high-risk pivoting sports significantly do worse when compared with those doing nonrisk pivoting sports. But this again has significant influence on secondary meniscus lesions.

Stefan Lohmander, MD, PhD: There are no controlled, prospective studies to allow a firm conclusion. However, several observational studies suggest that higher postinjury activity is associated with a higher degree of radiographic OA. The design of these studies precludes any firm conclusions. Again, controlled, prospective studies could give an answer.

John Richmond, MD: I agree that activity level is a likely contributor to degenerative joint disease(DJD), and so I warn my patients, particularly if they have had some degree of meniscal resection. I take almost a diametrically opposed approach to Stefan in counseling patients, in that I regard my role as a physician is to keep their knee from buckling either by activity modification, bracing and rehabilitation, or surgery.

Any of these treatments will work to reduce the risk of DJD if the menisci can be preserved. Since most young athletes are unwilling to prevent buckling episodes through nonoperative means, then ACL reconstruction is the best option to reduce the risk of late DJD.

“I am aware of no evidence that return to high-demand sports predisposes athletes to the development of OA.” — Lynn Snyder-Mackler

Lynn Snyder-Mackler, PT: Karl Peter has a nice article on this. I believe his data show that activity level is only related to arthritis in those who do not undergo reconstruction. Those who are reconstructed have better and higher levels of function. I think Scott Dye is a bit of an evangelist on this subject without much evidence.

People come to us for reconstruction in order to return to activity. Counseling them to reduce their activity level will only work for a small percentage of patients. Ultimately, preventing giving-way episodes should decrease the incidence of OA if the die is not already cast at the time of injury. In that case, they really are no worse off OA-wise and have a much better quality of life and activity level.

Benedetto: My personal experience, looking at our data at six-year and 13-year follow-up, has been that patients with nonoperative treatment of an ACL injury had a higher incidence of secondary meniscus lesions. If they went on with knee-demanding sport activities, they had a higher incidence of OA.

The torn ACL with cartilage damage and probably with bruising bone will have influence in OA regardless of whether the patient has been treated operatively or conservatively. But we did not look at that time at our patients to identify whether there was any influence of morphology type. Certainly the secondary meniscus lesion is one of the major risk factors for developing OA.

“Restoring knee stability through ACL reconstruction is an important factor in decreasing the incidence of degenerative changes after rupture of the ACL.” — Freddie H. Fu

Freddie H. Fu, MD: There is literature supporting the premise that high-level physical activity is associated with degenerative OA of the knee (McAlindon, Am J Med 1999; Spector, Arthritis Rheum 1996; Turner, Brit J Sports Med 2000). It has also been suggested that functional disability following ACL tear is correlated with a high physical activity level (Fithian, Orthop Clin North Am 2002).

However, I believe that it is difficult to conclude that a high activity level after ACL reconstruction accelerates the development of OA of the knee. There may be confounding variables in this patient population that may result in a selection bias (ie, associated chondral or meniscal pathology). Thus, the general damage to the knee (ie, cartilage, meniscus injuries) certainly has a significant impact on a patient’s postoperative activity levels. Therefore, the impact of the postoperative activity level on the degenerative changes of the knee is difficult to assess by epidemiologic investigations.

Fithian: Some very good studies, though not randomized controlled trials (RCT), have shown that reconstructed patients return to a higher level of activity than nonreconstructed patients. Does the ability to return to high-demand sports after ACL injury and reconstruction predispose athletes to the development of arthritis?

Lohmander: What is our best assessment of the effect of surgery vs. no surgery on ability to return to preinjury activity? Are there any RCTs or only observational studies? I think it is very difficult to answer this question without better studies.

Benedetto: There is no evidence that this is true but again this is multifactorial. It depends on the muscle strength and even more it depends on the coordination ability. High-demand sports after ACL injury and reconstruction have to be differentiated. By my personal experience — being involved in taking care of the Austria National Ski team for more than 15 years — I just can say that there is a big difference in whether one of the world cup athletes comes back as a slalom runner or as a down hill racer.

Snyder-Mackler: I am aware of no evidence that return to high-demand sports predisposes athletes to the development of OA. Stefan makes the excellent point that we have no RCTs, and John makes the point that the desire to return to sports is the precise reason patients have ACL reconstruction.

Reconstruction reliably returns most individuals to the athletic field, the best data showing a drop of one to two levels on the Tegner scale as an average, which is misleading. About 65% return to the same level and some never return – so averaging may not be valid.

Richmond: The return to high-demand sports is a likely contributor to DJD following ACL reconstruction. This is due to the higher joint reactive forces in sporting activities applied to a joint that has not been restored to normal biomechanically, as well as the biochemical alterations in synovial fluid and articular cartilage from repetitive hemarthrosis and bone bruising.

Fu: There are sufficient data in the literature to conclude that high-level physical activity is associated with the development of OA (McAlindon, Am J Med 1999; Spector, Arthritis Rheum 1996; Turner, Brit J Sports Med 2000).

However, I believe we cannot extrapolate these observations on patients undergoing ACL reconstruction. In this patient population, many confounding variables exist. Thus, the return to activity after ACL reconstruction very much depends on the general damage to the knee sustained at the injury (cartilage defects, meniscal injuries, etc.). Further controlled observational studies would help to answer this question.

Fithian: Is there evidence of on-going cell-mediated joint degradation following ACL reconstruction?

Benedetto: There is some evidence but it is extremely difficult to quantify and to make some prognosis.

“Several groups have reported increased levels of cytokines, proteases and cartilage matrix degradation products following injury to ACL and or meniscus.” — Stefan Lohmander

Lohmander: Yes, there certainly is such evidence. Several groups have reported increased levels of cytokines, proteases and cartilage matrix degradation products following injury to ACL and or meniscus. These levels are suggested to be associated with increased joint metabolism. They can be observed very early after injury, and they continue to be observed long after the injury, albeit at lower levels.

There are no comparative data on such markers from different types of interventions following injury (Lohmander et al, Arthritis Rheum 1993; Arthritis Rheum 1999; Arthritis Rheum 2003).

Here, also, I think it is worth asking whether there are any decent reports on the rate of adverse outcomes following ACL reconstruction. What is our best assessment of this risk, and what effect might it have on the reliability of reconstruction to prevent late OA?

Snyder-Mackler: Beynnon et al have some data that suggest that there are still markers of degradation two years after ACL reconstruction.

Richmond: Elevated cytokine levels seem to indicate that the degeneration of a joint following any mechanical perturbation is in part biochemically mediated.

Fu: I believe that restoring knee stability through ACL reconstruction is an important factor in decreasing the incidence of degenerative changes after rupture of the ACL; however, it is not the only factor. We looked at the changes in the synovial fluid cytokine and keratan sulfate concentrations after ACL rupture and examined the samples with enzyme-linked immunosorbent assays.

We found that keratan sulfate levels remained considerably elevated months after injury (Cameron, AJSM 1997). This suggests that posttraumatic OA after ACL rupture may not be purely biomechanical in origin, but also biochemical.

Fithian: Does ACL reconstruction as it is currently performed restore knee kinematics to normal? Here I am asking not just about translations but also about joint center of rotation, center of contact and the dynamic movement of these parameters as the knee is ranged both actively and passively.

Snyder-Mackler: HereI assume we are talking about translations, which are not restored completely to normal. But the kinematics resulting from modern ACL reconstructions performed by experienced surgeons are exponentially better approximations of normal.

Many of the surgeries performed in the study by Daniel and coworkers (eg, MacIntosh) are obsolete. The postop rehab was draconian. I think what we can say is that those with worse pathology had more arthrosis and that bad surgery may be worse than no surgery.

“There is a big difference in whether one of the world cup athletes comes back as a slalom runner or as a down hill racer.” — Karl P. Benedetto

Benedetto: ACL reconstruction neither as a single bundle nor as a double bundle reconstruction restores completely normal kinematics. I think, it has been proven that correct ACL reconstruction significantly reduces A-P translation. But this does not implicate normal kinematics

Richmond: Single bundle ACL reconstruction clearly does not restore normal kinematics to the joint. Freddie has looked at whether a double bundle technique is more successful in approximating normal knee kinematics.

Lohmander: It is my understanding of the literature that even the best reconstruction as practiced today does not restore the knee to ‘normal’ kinematics.

Fu: Much of the basic science work concerning the knee kinematics of the ACL-deficient knee and the ACL-reconstructed knee has been performed here at Pittsburgh. In our biomechanical study, “The effectiveness of reconstruction of the anterior cruciate ligament with hamstrings and patellar tendon” (JBJS 2002), we showed that reconstructions with either bone-patellar tendon-bone or hamstrings resist the in situ force in response to anterior tibial loads similar to the intact ACL; however, the in situ force in response to a combined rotational load with internal and valgus tibial torque was markedly reduced in the reconstructed knees using the transtibial approach and placing the femoral graft in the 11 o’clock position.

We then looked at the effect of placing the femoral tunnel at the 10 o’clock position compared to the 11 o’clock position (Loh, Arthroscopy 2003). The results showed that resistance to A-P translation was similar, but the resistance to rotatory loads was better in the 11 o’clock position. This position, however, did not restore full knee stability to the level of the intact knee.

We also looked at a biomechanical analysis of an anatomic (double bundle) reconstruction that separately reconstructed the anteromedial and the posterolateral bundles. We found that the in situ forces against both anterior and rotatory loads were better restored in the double bundle reconstructions compared to the single bundle group (Yagi, AJSM 2002).

This basic science research has led us to believe that a more anatomic approach to ACL reconstructions may better restore the kinematics of the ACL-reconstructed knee to normal. At our center, we are now performing anatomic (double bundle) ACL reconstructions that specifically restore the AP kinematics (anteromedial bundle) and the rotational kinematics (posterolateral bundle). Our early clinical results have been promising; however, longer term follow-up and controlled clinical trials are needed to validate this claim.

Fithian: Has gait or motion analysis revealed residual movement abnormalities in patients who have undergone ACL reconstruction?

Richmond: My bias is that most patients do return to near normal in gait and muscle firing patterns, but I defer to Lynn as the expert.

“Many patients are under rehabilitated. Many doctors, without evidence, have suggested shorter and shorter rehab times and focus on exercise that does not appreciably train the quadriceps.” — Lynn Snyder-Mackler

Snyder-Mackler: Yes, there are residual movement abnormalities in reconstructed knees, but these are largely related to muscle weakness. We and Paul Devita have shown persistence of aberrant gait patterns after ACL reconstruction, but strongly related to quadriceps weakness. (Lewek, Clin Biomech 2000).

Patients who remain weak after surgery walk just as they did before surgery, with less knee excursion, lower knee moments and significant quadriceps hamstring cocontraction, a dangerous stiffening pattern that, on face, could contribute to the development of OA via increases in joint compression and impact forces.

Fu: It has been suggested that patients after ACL reconstruction demonstrate abnormal gait patterns. In particular, the quadriceps avoidance pattern has been a frequently observed phenomenon. It appears that the gait pattern normalizes within the first year after the ACL reconstruction (Knoll, Knee Surg Sports Traumatol Arthrosc 2004).

However, it must be emphasized that many gait analyses provide limited information. Thus, the investigation of muscle strength and the anterior tibial translation has been well examined. Recent surgical advances, such as the ACL double bundle reconstruction, aimed at improving the rotational stability of the ACL deficient knee.

As of now, there is no reliable measurement for the rotational knee stability. Current investigations in our laboratory focus on the investigation of rotational knee stability in both ACL deficient and normal knees.

Fithian: Does postoperative rehabilitation have any effect on the development of arthritis? What is the importance of postoperative rehabilitation in restoring knee function?

Lohmander: As far as I know, there are no conclusive data either for or against. Bruce Beynnon and colleagues are performing interesting work in this area, which is not yet published.

Benedetto: Certainly it does. Immobilization and high aggressive rehabilitation causes damage in the long-term result. It is important to have some guidelines for the rehab protocol but they have to be adjusted to each patient individually. It should include closed chain activity, underwater therapy for mobilization and early and intensive coordination.

Snyder-Mackler: I believe our research and that of others suggest strongly that adequate rehabilitation of the knee can affect the development and progression of knee OA. First, failure to restore adequate knee extension ROM can result in patellofemoral OA, presumably as a result of high PF contact forces from walking on a flexed knee.

We have demonstrated that weakness of the quadriceps is compensated for by high quadriceps hamstring and quadriceps gastrocnemius cocontraction, and results in a flexed knee gait pattern. Many patients are under rehabilitated. Many doctors, without evidence, have suggested shorter and shorter rehab times and focus on exercise that does not appreciably train the quadriceps. Then, because they are afraid of the strain of quadriceps exercise on the graft, they don’t test the quads and return people to activity with quad weakness. This is dangerous. We need strict criteria for allowing return to activity.

Stefan mentioned the Beynnon study that is still not published but with which I am very familiar. Ultimately there was no difference in any variable between their two rehab programs (laxity and markers of OA). Both groups had about a 2-mm increase in knee laxity at two years vs. intraoperative values, and both groups had elevation of one of the markers still at two years.

Richmond: I am absolutely convinced that adequate postop rehabilitation reduces the risk of subsequent OA. Immobilization, extension loss, overly aggressive open kinetic chain quad exercises have all been proven to be deleterious to the articular surfaces and have been eliminated from most ACL postop rehab protocols.

One must be careful about attributing the high rates of DJD in those patients operated on in the 1980s to the surgery per se, since many of our postop treatment protocols contributed to the risk of DJD. I would anticipate that those patients treated in the late 1990s will likely fare better in the long-term with the change in rehab techniques.

Fu: Thereis consensus that adequate postop rehabilitation results in earlier return to the previous activity level, greater range of motion, superior knee proprioception and improved muscle strength. Therefore, it must be assumed that optimized rehabilitation regimens also reduce the incidence of late knee degeneration. However, this concept has not been supported by reliable and significant data in the literature, and controlled long-term clinical trials will be necessary to allow a conclusive answer to this question.

“I am absolutely convinced that adequate postop rehabilitation reduces the risk of subsequent OA.” — John C. Richmond

I do agree with the others that an inadequately rehabilitated ACL-reconstructed knee will predispose the knee to early degeneration. It is important that full range of motion is attained with restoration of a normal gait and good muscular control around the knee joint.

Fithian: The question of whether patients who undergo ACL reconstruction will have more, less or the same risk of late-term arthritis than nonreconstructed patients has not been resolved in the literature. It is not possible to answer the question directly without a randomized trial.

There is a strong consensus among the panelists that intrinsic bias affects the results of any given published study. A randomized prospective clinical trial now being conducted at Lund University in Sweden will be the first truly randomized study to look at late degenerative changes and related outcomes following ACL injury. We await its findings with interest.

Other deficiencies of the current literature also have been highlighted by this discussion. Several panelists, for example, indicated a belief that newer techniques of reconstruction and of rehabilitation will provide better outcomes than older approaches. There is good reason to share their belief, but there is much to be said for clinical studies that could prove that our technological and conceptual innovations are actually delivering on this promise.

Indeed, inherent in that argument is the recognition that techniques we have been using are not altogether satisfactory. Further refinements to the technique of ACL reconstruction, such as multiple-bundle reconstructions and techniques intended to restore rotational stability need to to be compared to existing techniques to show their effectiveness in preventing arthritis and other bad outcomes.

Finally, efforts designed to optimize lower limb control after ACL reconstruction should be studied in order to develop better guidelines for return to sport after surgery.

For more information:

• Daniel DM, Stone ML, et al. Fate of the ACL-injured patient: A prospective outcome study. Am J Sports Med. 1994;22:632-44.
• Dye SF, Wojtys EM, et al. Factors contributing to function of the knee joint after injury or reconstruction of the anterior cruciate ligament. Instr Course Lect.1999;48:185-98.
• Fink C, Hoser C, et al. Long-term outcome of conservative or surgical therapy of anterior cruciate ligament rupture. Unfa-llchirurg.1996;99:964-9.
• Fink C, Hoser C, et al. Long-term outcome of operative or nonoperative treatment of anterior cruciate ligament rupture —is sports activity a determining variable? Int J Sports Med. 2001;22:304-9.
• Fithian DC, Paxton LW, et al. Fate of the anterior cruciate ligament-injured knee. Orthop Clin North Am. 2002;33:621-36, v.
• Von Porat A, Roos EM, et al. High prevalence of osteoarthritis 14 years after an anterior cruciate ligament tear in male soccer players: a study of radiographic and patient relevant outcomes. Ann Rheum Dis. 2004;63:269-73.
• Lohmander LS, Östenberg A, et al. High prevalence of knee osteoarthritis, pain and functional limitations in female soccer players twelve years after anterior cruciate ligament injury. Arthritis Rheum. In press.