All right. I'd like to thank Anna and IOSSM for inviting me to give the talk about arthroscopic debridement today. I'm going to go through these slides fairly quickly. No disclosures related to this topic. Normal articular cartilage has a normal homogeneous staining pattern for glycosaminoglycans as you see here in this slide. In addition, the superficial layer, if intact, cartilage works really well at protecting the subchondral bone. This tide mark here, as long as that's not penetrated, will tend to maintain a good healthy distribution of force across the bone and really work well. When it's damaged, however, that's when the problems arise and it's difficult sometimes to figure out whether there's actually a cartilage lesion based on a standard MRI unless there's bone edema. If you look at the prevalence of arthroscopic findings by this study where we looked at 1,000 arthroscopies, you see that in patients less than 40 years of age, there was only a 5.3% prevalence of grade three or four lesions. These were usually unrecognized until arthroscopy. Similarly, in this larger study, a retrospective study where they looked at 53,000 hyaline cartilage lesions, they found once again in patients that were less than 40 years of age, grade four lesions were very low in prevalence, overall prevalence. These small lesions can be hard to recognize prior to arthroscopy. That's one of the problems we have. You look at this nice review by Jody Buckwalter. It looks at the presence of bone marrow lesions. If there's really no bone marrow lesion, an MRI is not going to pick up an articular cartilage finding. You see here in this area here, we see some bone marrow irregularity. That suggests there is some cartilage damage there, but without that in small lesions, you're not going to see much by MRI. If you do find out there's mechanical symptoms and synovitic change effusions, then we know by clinical evaluation that there probably is some cartilage damage or meniscus pathology, then we have a hint that things are going on inside the knee. Typical arthroscopy by chondroplasty refers to taking the shaver and gently removing any irregular segments as you see here. We want to remove those cartilage flaps. As you see, I'm removing this little flap here. That's all you want to do. Tendency to turn the shaver towards the articular cartilage should be avoided. You don't go deeper than you need to with the arthroscopic chondroplasty. As you can see, I've also performed a concomitant arthroscopic meniscectomy. We can perform a lot with an arthroscopy, but the chondroplasty procedure really refers to removing those irregular flaps, damaged flaps of cartilage down to the normal smooth surface. You can see here, if you do it appropriately, you don't have to take a lot of cartilage, but you can certainly stabilize the damage and prevent it from extending further, which is typically my goal with arthroscopic chondroplasty. You can also remove, obviously, loose bodies. As you see here, a large loose body has been removed from the posterior compartment of the knee. This is causing locking and catching. Arthroscopic procedures certainly can have some improvements if there's mechanical symptoms that a patient has prior to surgical intervention. You look at this study by Levy, 15 young college soccer players, small acute lesions of 42 millimeters squared had really positive response with 100% return to soccer at a very short period of time following arthroscopy. You can have some pretty good results if there's small lesions with acute histories, long periods of history certainly can cause bigger problems post-operatively. In this study by Dennis Crawford and colleagues out of Portland, 53 of the actual 86 patients eligible for questionnaires filled out the questionnaires. We see that the mean age was 37, BMIs were 27, most were women. We see that there were grade two or three lesions that were typical findings and KL scores were in the zero to two range in most of the patients. 3.3 centimeters squared size lesions demonstrated significant improvements in almost all the patient recorded outcomes measures in this study. When they did regression analysis to assess further, they found that lower pre-op scores actually gained more after chondroplasty interestingly. They also correlated by regression equations whether these pre-op scores were associated with any particular findings and KL scores that were slightly higher than zero in males were typically the most positive response to arthroscopic chondroplasty. By contrast, high ICRS grade lesions down to bone, as you see here in this picture here, down to bone, had worse results with arthroscopic chondroplasty, something that would make sense. Lesion size, age, obesity had no impact on outcomes interestingly. Only eight patients out of this large series went on to second surgical interventions in the follow-up period of 31 months. This study looked at 193 bipolar radiofrequency-based chondroplasties over a 38 month period of time. They did second look arthroscopies in this study, which was very nice, in 15 patients. We looked at 25 lesions in that 15 patients. These were patients who had some recurrent and newer injuries. They went back and looked inside. They found over this period of time that they could assess whether they responded well to treatment with standard radiofrequency chondroplasty. They also did look at lesion size, location, grade, and stability of the lesions at second look arthroscopy. At second look arthroscopy, as opposed to the initial lesion size of 625 millimeters squared, they went down to 300 millimeters squared. They found some improvements in size of the lesion at second look arthroscopy in that small subset of patients. They looked at three that they felt to be unstable. There was progression of that damage beyond the initial lesion to the surrounding articular cartilage, but most of them did not change in size, where eight were found to be of no significant change in size. Eight actually had partial filling with stable repair tissue. Six actually had complete fill, which was a very interesting finding in this study. Lesions of the tibiofemoral joint did much better than the patellofemoral joint in this study. This study looked at 60 patients, and they randomized them to mechanical shaving alone, as I was just showing you in my little video, versus using a monopolar radiofrequency probe to further stabilize lesion along with chondroplasty. They looked at them preoperatively, and in 12 months with MRI, also used tegnylisone, Cincinnati, and IKDC scores to take a look at these patients. The end scores were set to 12 or 24 months in this study. They had a good number of patients come back for follow-up, 28 in the shaving group versus 28 in the monopolar radiofrequency group, and they found that the demographics were very similar between the two groups. Postoperative MRIs were very similar, with no incidence of avascular necrosis, which is one of the concerns using a heat probe on bone. Nineteen months postoperative follow-up was the mean follow-up time. They once again looked at those outcomes and measures and saw no significant difference between the groups in outcomes, which is very interesting. There was no difference in mean improvement between groups either. This study looked at…several studies have looked at whether there's good to excellent findings after arthroscopic debridement in arthritic knees, and we found that they have very similar kind of findings, as long as it wasn't poor alignment, long duration of disease, or radiographic changes which were significant, then patients tend to do pretty well in the good to excellent group for about two years. If we look at one of the groups in particular, the Baum-Gaetner group, and we look at that group in that study in particular, 49 knees and 44 patients, with a fairly significant number with radiographic findings which were severe, we found that poor clinical findings or results from arthroscopic contraplasty were demonstrated with malaligned knees and severe radiographic changes. Long duration of symptoms, as you would expect, were correlated with these poor outcomes as well. They found improvements in patients with shorter duration of symptoms and mechanical symptoms, so loose spies, irregular degenerative meniscus tears, which were causing flaps of tissue to cause mechanical symptoms, did correspond well with outcomes following contraplasty. Also, where there were mild to moderate radiographic findings, we certainly found better improvements in this study. Crystal deposition diseases did improve better at the arthroscopic contraplasty as well. A good number of the patients didn't demonstrate any significant benefit, so 39% of the patients did not improve, and another 9% had temporary improvement. Once again, the follow-up, though, demonstrated poor findings. Good to excellent findings, once again, were 52% of the patients, once again, with shorter duration of symptoms and mechanical symptoms, but there was a number that dropped to 40% when they looked at final follow-up, so it actually did drop off significantly at final follow-up to 40% overall, having some positive response at 33 months time following a surgical intervention. Another study looked at the 126 patients that had failed medical management following arthroscopic debridement. They looked at the health-related outcomes measured using the WOMAC quality of life assessment score, as well as SF36. They also had an interesting part of the study where they looked at whether the surgeons could actually predict whether patients were going to do better or worse after surgery using fellows, as well as attending staff to use independent evaluations preoperatively, standardized clinical assessments with clinical signs and symptoms, and plane radiographs prior to surgery as well. They did resect arthroscopic contral flaps, as I showed you in my video, meniscus tears as well, but no abrasion contraplasty was performed. We looked at this study. There was really a poor rate of the surgeon's ability to determine who was going to really benefit from this type of surgery, so really 54% in the fellows' hands, 59% in the staff's hands, so really no difference despite having more experience in the staff. If they looked at these patients, the people with medial compartment wear typically did pretty poorly. Unstable meniscus tears were correlated with findings that were difficult to predict as well. The three biggest variables that are associated with improvement in this study were interestingly medial joint line tenderness and a positive Steinman test, which is that rotational move when you assess the patients for meniscus tear at flexion greater than 90 degrees. In addition, unstable meniscus tears were one of the positive predictors for improvement following this surgery. In addition, they looked at whether or not patients responded well to the level of damage in the knee. This study by Robert Jackson, stage 1 and 2 patients where they had minimal findings by radiographic assessment and chondral assessment did very well, whereas there was variable findings in stage 3 disease and stage 4 disease, as you would expect and tends to be the typical finding in all the studies I've seen, had poor results. Once again, stage 1 disease to 2 disease did really well. 3 disease was variable. As you see, 49% did have some positive response, but a good number, 51%, did not do well. In stage 4 disease, the majority of patients did poorly. If you look at this MEDLINE-based Cochrane analysis using a systematic review, they used two independent reviewers to assess all titles and abstracts for eligibility in the study. They did risk of bias in the included studies and pooled the outcomes studies with a random effects model. We looked at outcomes using function and pain relief and dichotomized based on short-term, less than six months, and long-term, greater than two years findings, whether they were actually positive results or negative results. Interestingly, seven of these randomized controlled studies included 805 patients were included in the study, but none really showed significant minimal clinically important differences in either treatment arm in a long-term arm versus the short-term arm. The short-term arm did show some improvement, but it was not statistically significant for MCID. Arthroscopic surgery did not show significant improvement in pain scores, either in either the short-term or long-term results in this study. There was a heterogeneity of low to moderate findings for outcomes and measures. If you look at these studies overall, good to excellent findings were typically demonstrated with patients with shorter duration of symptoms, well-contained lesions, and single lesions that were not kissing in nature. Six months to two years, you could demonstrate some improvements, but those results tended to trend significantly in downward trend after two years. Small lesion sizes of less than two to three centimeters squared did better. KL scores of zero to two certainly will do better than KL scores that are higher. Mechanical symptoms can be improved, obviously, with arthroscopic debridement, so locking and catching type symptoms certainly is an indication to perform arthroscopic chondroplasty and arthroscopic surgery overall. Longer duration of symptoms, contracted knees, osteofix formations, multiple lesions, kissing lesions certainly are the problems that we'll be worried about using arthroscopic chondroplasty. Treat these may not give you positive results. Greater than two years, you're going to see your results drop off significantly, and larger lesion sizes are going to do more poorly in most people's hands. KL scores of three to four are certainly going to be a warning sign against doing this type of surgery. You may need to do something more definitive, such as an osteotomy or a partial replacement or full replacement type surgery, and really deal with those patients with kissing lesions in a more definitive way. Thank you.

arthroscopic debridement

articular cartilage

MRI

prevalence and outcomes

lesion size