Clinical Faceoff: The Role of Arthroscopic Partial Meniscectomy in the Treatment of Meniscal Tears

Treating symptomatic degenerative meniscal tears used to be relatively straightforward. Some surgeons would recommend physical therapy and a cortisone injection followed by surgery if nonoperative treatment did not relieve pain, while others recommended surgery as soon as they identified a tear. Because tears are so common, and because they increase in frequency as part of normal aging [2], how we care for patients with this diagnosis is a serious question. If tears occur as part of normal aging and so often are asymptomatic, we need to be careful when we recommend surgery to patients with degenerative meniscal tears in order to avoid a surgical procedure for patients who may not need it. In fact, we now have numerous randomized [10]—and sometimes placebo-controlled [9]—surgical trials on the topic, and most of those call into question when and whether arthroscopic partial meniscectomy should be performed, particularly if the patients are older, have degenerative (rather than traumatic) tears, or if underlying osteoarthritis (OA) is part of the picture. However, not all surgeons believe that those trials generalize well, and some remain enthusiastic about the potential for arthroscopic partial meniscectomy to relieve symptoms in many patients who undergo the procedure.

We asked two prominent orthopaedic surgeons who have published extensively on the subject matter to help clarify this hotly debated topic: Dr. Robert Marx from the Hospital for Special Surgery and Dr. Raine Sihvonen from the University of Tampere in Finland. Dr. Marx is a world-renowned sports medicine surgeon, and has been a leader in the pursuit of evidence-based medicine on this topic, having written several books, chapters, and countless research papers. Dr. Sihvonen is a leading voice questioning the use of arthroscopic partial meniscectomy, and lead author on one of the most-important large randomized controlled trials on the topic, a placebo-controlled trial published in the New England Journal of Medicine [9].

Bruce A. Levy MD: If a patient presents with mild OA on a radiograph and a meniscal tear on MRI, what are your “triggers” to recommend arthroscopic partial meniscectomy versus nonoperative care? Is it the type of tear, duration of symptoms, painful effusions, or painful mechanical symptoms like catching or locking?

Robert G. Marx MD: There are several patient and imaging factors that suggest a patient may benefit from arthroscopic partial meniscectomy rather than further nonoperative management. The key is determining whether the symptoms and physical exam are more likely to be related to OA. Age is a negative prognostic factor, and generally, I recommend nonoperative treatment for patients older than 50-years-old prior to performing arthroscopic partial meniscectomy. Patients who have pain that keeps them up at night, or who are only able to walk short distances (five to 10 blocks) are more likely to have OA-related pain and still have pain after arthroscopic partial meniscectomy. On physical examination, patients who have a fixed-flexion deformity on the affected side compared to the normal side are less likely to benefit from arthroscopic partial meniscectomy.

It is critical to obtain good weight-bearing radiographs, including an AP view as well as a flexed AP view, each of which must be parallel to the joint line. The presence of radiographically visible arthrosis makes arthroscopic partial meniscectomy less predictable. In my opinion, arthroscopic partial meniscectomy should generally not be done in the setting of severe joint space narrowing. There are many factors on MRI that must be evaluated. A radial split in the posterior horn clearly seen on coronal images will tend to respond well to arthroscopic partial meniscectomy. Reading the report without viewing the images is never sufficient.

Raine Sihvonen MD: It remains unclear whether any type of patient, symptom, or tear characteristics factor into whether patients benefit more from arthroscopic partial meniscectomy than from nonsurgical treatment. One study demonstrated that it was not possible to determine whether a meniscus tear would be stable or unstable based on symptoms or clinical examination [1], and so in practice, that determination generally is based on the treating surgeon’s personal experience or gut feeling. Further, we do not know that the presence of so-called mechanical symptoms is associated with a higher likelihood that arthroscopic meniscectomy will improve a patient’s symptoms, since we do not know whether those symptoms arise from the meniscus or from uneven chondral surfaces, as recent evidence suggests [7]. In fact, the best evidence we have—from randomized clinical trials [3, 5, 9]—has not found mechanical symptoms to be associated with improvement after surgery. Although those trials excluded patients with locked knees, that presentation is quite rare in the patient population we are discussing. Although tempting, the decision of whether to recommend surgery should not be based on opinion, but rather on scientific evidence. The obligation to gain more evidence rests now on the shoulders of those arguing for arthroscopic partial meniscectomy.

Dr. Levy: A substantial number of randomized clinical trials [3, 5, 9] have called into question the efficacy of arthroscopic partial meniscectomy in middle-aged patients (especially with mild or no OA) and, as Dr. Sihvonen points out [7], some have even demonstrated that arthroscopic partial meniscectomy fails to alleviate painful mechanical symptoms. Considering these findings, have you changed your practice?

Dr. Marx: Two studies published in the New England Journal of Medicine [5, 6] influenced my practice. These papers, despite their flaws, clearly demonstrated that patients with OA did not benefit from knee arthroscopy. I have also occasionally seen in my practice that some patients had disappointing results after knee arthroscopy, and in retrospect, it was likely that their pain was more related to arthritis than their meniscal tear. More recently, studies have claimed that arthroscopic surgery for meniscus tears is ineffective [8]. However, in my clinical experience, patients whose clinical picture fits well for meniscal tear without OA usually have dramatic pain relief at 2 weeks after surgery, which is maintained for years. When reading such randomized trials, it is difficult to imagine that the patients included resemble the ones that I operate on. Like every surgical procedure, the outcomes after arthroscopic partial meniscectomy are highly related to the indications for surgery. If the indications for surgery are not clearly established, and rigidly adhered to, patient outcomes (with respect to pain relief and improved function) will generally be poor.

Dr. Sihvonen My practice has changed dramatically. For example, our institute performed approximately 500 arthroscopic partial meniscectomies in 2007. Last year, we performed fewer than 30. We have focused on patient education that reassures the patient that these symptoms often are self-limiting, are generally a part of normal aging, and may be managed with ways other than surgery, such as home-based exercise. As mentioned above, more trials are needed to find those rare patients who possibly benefit from arthroscopic partial meniscectomy. But in my opinion, as long as we do not have evidence to support arthroscopic partial meniscectomy, it should not be done unless performed within a randomized clinical trial.

Dr. Levy: When patients, residents, or fellows ask you about the current data on arthroscopic partial meniscectomy, how do you explain the role of research studies in defining medical treatment guidelines, especially if it differs from your own personal experience? If we are not “following the evidence,” how can we differentiate ourselves from other kinds of providers who approach things “empirically” (such as those providers who believe in magnets or copper bracelets for arthritis pain).

Dr. Sihvonen: Expert opinion, which arises from years of experience, is unfortunately vulnerable to various biases; it does not consider the possibility that symptoms may fluctuate over time, and that conditions associated with normal aging, like meniscal tears, may have symptoms that wax and wane, or that abate over time. Expert opinion also does not generally account for the placebo effect, which can be strong as surgical interventions are concerned [11]. When a surgeon has a feeling that patients improve after a certain surgical operation, the only way to make this feeling a fact is to evaluate every patient with validated outcomes tools over a reasonable period of time to ensure that apparent early successes can endure. Then, we must consider what the results would be without surgery, and compare those results to what has been realized by using surgery. To make sure that the patients in both treatment groups are similar, researchers should randomize the patients. In other words, to really know whether the intervention works, we need well-designed randomized clinical trials. As it turns out, for this diagnosis, we have many of them, and the weight of evidence from those trials does not appear to support arthroscopic partial meniscectomy.

Dr. Marx: I agree with Dr. Sihvonen that randomized controlled trials are the highest level of evidence in clinical research. However, randomized controlled trials work best for well-defined clinical problems to compare interventions. The scenario of a meniscal tear in the setting of degenerative knee changes can be very heterogeneous [4]. While randomized controlled trials are the gold standard for clinical research, having participated in such a trial [4], I have learned that the single answer provided by a randomized controlled trial does not necessarily apply directly to widely varying individual patients. Furthermore, trials comparing nonoperative management to surgery are fraught with crossovers that complicate data analysis and make it more difficult to arrive at clinically and practically useful conclusions. We must strive to use the best available evidence for all decisions involving patients, and in this particular area, we need to best define which patients are highly likely to benefit from arthroscopy.