Is the Hills-Sachs lesion best treated with a surface replacement? Sumant Krishnan, M.D. of The Shoulder Center in Dallas says “yes” while Anthony A. Romeo, M.D. of Midwest Orthopaedics at Rush disagrees. Who gets the best of this debate? Read on…
This week’s Orthopaedic Crossfire® debate is “The Hill-Sachs Lesion is Best Treated With a Surface Replacement.” For the proposition is Sumant Krishnan, M.D. of The Shoulder Center in Dallas, Texas. Against the proposition is Anthony A. Romeo, M.D. of Midwest Orthopaedics at Rush in Chicago, Illinois. Moderating is John Brems, M.D. of the Cleveland Clinic.
Moderator Brems: “We’re debating an actual case today. It is a 32-year-old amateur boxer who has had numerous anterior dislocations; they now occur daily and nocturnally. The clinical exam is consistent with isolated recurrent anterior instability. He has failed to respond to nonoperative management and has had no prior surgery.”
Dr. Krishnan: “To be clear, a Hill-Sachs lesion is a posterolateral impression fracture of the humeral head with an anterior subcoracoid dislocation. Fifty years ago Michel Latarjet described the expansion of the glenoid by attaching a corocoid process to that to treat all manner of bone problems around the shoulder. But there are lesions that cannot be treated that way…a significant Hill-Sachs lesion involves the location and size of the lesion and the patient’s age.”
“There is a contact point as the humerus comes into abduction and external rotation in which it tracks on the glenoid. Nearly a decade ago Yamamoto and Romeo helped popularize the glenoid track, i.e., that the humeral head is a contact point to the glenoid as the arm is moving. If the Hill-Sachs lesion is within that track—generally 18mm medial to the rotator cuff footprint—then it will engage.”
“As for size, if you look at the percentage of articular surface involvement, and consider that this surface is half of a sphere, we realize that this is a 50 degree articular surface involvement. But most of the biomechanical studies with regard to engagement of significant Hill-Sachs lesions involve a percentage of the humeral head diameter.”
“Patient age is probably the most important component in defining what is significant about a Hill-Sachs lesion. A study in the British Journal of Radiology found a positive correlation between the bony defect size and the age of the patient. In a retrospective study of more than 400 consecutive multiple dislocators, researchers found that if the patient was over 29 there was a higher degree of depth and size of lesions on the humeral head compared to patients younger than 29. Perhaps it’s related to relative osteopenia or multiple trauma.”
“There are complications with regard to grafting defects. A size mismatch can be very ineffective with regard to articular surface restoration. There’s fixation failure with screws if you’re trying to fix a bulk/sized allograft. Dr. Romeo’s partner, Dr. Cole, said that the best grafts are those that are fresh…fresh, frozen osteochondral allografts that are less than 28 days old. With regard to arthroplasty complications, there can be a size mismatch. We do know with surface replacements that if you don’t match the radius of curvature (the articular surface to where the rotator cuff inserts) you might engender a cuff difficulty.”
“So in our case here the lesion is within the glenoid track. Whether it’s a humeral head defect or some type of angular measurement, it’s approaching between 30-50% of the articular surface. He is 32 years old, but there is pre-existing dislocation arthropathy. Dr. Bob Samuelson helped us understand a classification for dislocation arthropathy relative to the humeral head and glenoid osteophytes. In this 32 year old there are cartilaginous changes, both on the glenoid and humeral sides. So in this case perhaps it’s best to proceed with a surface replacement.”
Dr. Romeo: “When we have these individuals with a fairly violent event there can be significant bone lesions on both sides of the joint. The question becomes, ‘When is the humeral side important to address?’ In the concept of the glenoid track is the fact that the glenoid will fall into that defect. You can address that two ways. You can widen the glenoid per Latarjet or corocoid transfer—or you can fill the defect to avoid the glenoid from hitting into the hole that is created from this injury.”
“We know that for the lesion to occur that at some point they all engaged…and we try to understand when this happens—especially below 90 degrees of abduction. We want to know how much this plays a role in recurrent instability.”
“How often is this really an issue? A group in Japan felt that it was about 7%. If you take care of a large rugby population that number goes up dramatically. Most of the time our treatment is glenoid-based, but it is occasionally humeral-based as we cannot resolve the entire issue by bone-grafting on the glenoid side when we have a large humeral defect.”
“You can’t fix all bone lesions with a soft tissue operation. When there is a Hill-Sachs lesion that is less than 15% we manage the glenoid pathology exclusively. When it’s 15-25% we will incorporate our management on the arthroscopic side if there’s minimal glenoid bone injury. If the glenoid bone problem is significant then we treat it on the glenoid side with a coracoid transfer. If the lesion is more than 25% we must consider the possibility that we won’t be successful with just our soft tissue or arthroscopic treatment alone. We try to identify the glenoid track to make sure we understand that the glenoid is going to fall into this defect.”
“If it is a smaller lesion you can fill that defect with the soft tissues of the capsule and the infraspinatus and get a good result. This was demonstrated by Pascal Boileau where he looked at 47 patients who were around 29 years of age, and found that the capsule healed in many of these patients. Boileau provided us with a nice clinical algorithm. If you have a bone lesion we get a CT scan; if you see minimal damage on the glenoid side we can address a significant humeral side lesion (under 25%) with a remplessage.”
“We use fresh osteochondral grafts and we have a lot of experience with this around the shoulder. These are not frozen or preserved. They are tested and after 14 days they are allowed to be placed into your patient by 28 days. If you are in the shoulder, the cartilage will survive at a very high rate. So we’re doing a biological restoration of the shoulder. We can match these defects very well, and it’s important to try to match them so we don’t distort the overall anatomy. I typically address the glenoid side arthroscopically in the lateral decubitus position. Then we make a split posteriorly in the same position (arm out of traction) and we can address that Hill-Sachs lesion with an osteochondral graft that is fresh with live cartilage. We check to ensure that’s it is tracking well, we move it back and forth, and we will sink the screw. Our healing rates approach 100%.”
“The best study yet is from Canada…the 30% defect model from George Athwal and his group; their work on remplessage, osteochondral graft or partial replacement is outstanding. Joint stiffness is reproduced normally with the two techniques of the bone as well as the graft and surface replacement, but not so well with the remplessage. With the osteochondral graft we get all the benefits and we restore anatomy. The one problem we have with partial resurfacing is while it looks good and does a lot of nice things, we can’t match it up perfectly with the normal anatomy.”
Moderator Brems: “This was a real case. If this person was 60 years old would that change what you do?”
Dr. Romeo: “A 60 year old would make us think more along the lines of joint replacement, either complete surface replacement—not partial—or hemiarthroplasty. Because you get dislocation arthropathy, if we lower that number to 40 and below then I would stick with my argument.”
Moderator Brems: “Sumant, if this patient had a surface replacement would you let him return to boxing?”
Dr. Krishnan: “J.P. Warner has probably done the best study on anatomic replacement on the humeral side with just a hemiarthroplasty. He demonstrated that the contact point of the glenoid—if it’s truly anatomic—is no different from a natural shoulder. And the glenoid wear that they have started to demonstrate over 5-10 years has not been the ‘one in four develop glenoid arthritis. So I would let him return to boxing.”
Moderator Brems: “Tony?”
Dr. Romeo: “From the beginning I would assume that he is returning to boxing. So the graft would be part of that.”
Moderator Brems: “What about the timing of his return?”
Dr. Krishnan: “As soon as the cup can stabilize that humeral head—three to six months.”
Dr. Romeo: “We are treating most of these patients on the glenoid side, so for a boxer I would let him back to drills and shadow boxing at three months…six months until he could compete.”
Moderator Brems: “Do imaging techniques adequately allow you to assess graft incorporation?”
Dr. Romeo: “Yes, but be patient. If you take a picture of these at three months you often see the gap and there is some osteopenia. You shouldn’t interpret that as a nonunion because the patients are clinically stable…if you follow them long enough you will see that all incorporate.”
Dr. Krishnan: “We agree that the most effective advanced imaging both preop and postop is a 3D CT for this implant.”
Moderator Brems: “At what point would you convert to a stemmed humeral implant from a surface replacement?”
Dr. Romeo: “In my algorithm I rarely use a surface replacement. I use a limited or smaller stemmed replacement. But if there was enough deformity of the humeral head such that the surface replacement cannot sit anatomically with a nice mechanical fit then I would convert.”
Moderator Brems: “Cement or non cement?”
Dr. Romeo: “Uncemented.”
Dr. Krishnan: “Uncemented.”
Moderator Brems: “Thank you both.”
Please visit www.CCJR.com to register for the 2014 CCJR Winter Meeting, December 10 – 13 in Orlando.
