Contemporary elbow surgery: Consider your abilities and patient expectations

In Part 2 of this two-part Round Table discussion on elbow surgery, I pose questions to some of the leaders and innovators in the field to gather their opinions and experience with elbow arthroplasty.

In part 1, we addressed the sports medicine and trauma aspects of elbow surgery. Here we take on elbow arthroplasty through case examples and how our participants would treat these patients.

William N. Levine, MD
Moderator

Round Table Participants Moderator William N. Levine, MD Vice Chairman Education Professor, Orthopedic Surgery Director Sports Medicine Associate Director, Center for Shoulder, Elbow & Sports Medicine Columbia University Medical Center New York, N.Y. Michael D. McKee, MD, FRCSC Professor, Division of Orthopedics Upper Extremity Reconstruction Service St. Michael’s Hospital University of Toronto Toronto, Ontario Ken Yamaguchi, MD Sam and Marilyn Fox Distinguished Professor of Orthopaedic Surgery Chief, Shoulder & Elbow Service Washington University School of Medicine St. Louis, Mo. Bernard F. Morrey, MD Professor of Orthopedics Mayo Clinic Rochester, Minn.

William N. Levine, MD: What is your operative algorithm for a 50 year-old right hand dominant (RHD) female with rheumatoid arthritis who has 40° to 100° flexion/extension arc, severe pain, and moderate joint space obliteration on radiographs?

Michael D. McKee, MD, FRCSC: This is certainly a difficult problem for an individual who, in my country, can reasonably expect to live another 30 to 35 years. There are relatively few arthroplasties that would have a track record of lasting this long even in a low demand female rheumatoid patient. My standard elbow arthroplasty is the Coonrad-Morrey semi-constrained linked prosthesis (Zimmer) and I use this prosthesis for the majority of my elbow arthroplasties. However, there is increasing evidence that an unlinked or unconstrained prosthesis may be better for younger patients although there is no long-term follow-up of these prostheses at the present time. Certainly there is an increased short-term complication rate, especially with regards to instability, when using these newer prostheses. It remains to be seen if the theoretical long term benefits will be borne out. I would have a discussion with this patient regarding the prosthetic arthroplasty options for her but I would probably recommend that we proceed with the standard semi-constrained total elbow arthroplasty (TEA) using the 4-inch humeral stem, and explain to the patient that at some point in her life she will most likely require a revision of this prosthesis.

Bernard F. Morrey, MD: At this age, I would consider an arthroscopic intervention if the joint appears to be reasonably well preserved. If there is moderate joint space obliteration, arthroscopic synovectomy is less reliable. I would mention, however, a recent review of Mayo’s data shows that more aggressive arthroscopic debridement has proven reliable to increase the arc of motion. This is a finding that has not been appreciated in the previous literature. In any event, I do not think arthroscopy would be reliable in this patient. If the patient is very active and feels as though there is a need to aggressively use the extremity, I would discuss interposition arthroplasty. This has been very successful in improving motion when the elbow is stiff and has been moderately successful in relieving pain. If the patient wants to relieve pain at all costs and is willing to accept the limitations of a total elbow, then I would offer her a total elbow as the most reliable procedure; (Figures 1 and 2).

Preoperative AP radiograph (a) of a 63 year old female with long-standing rheumatoid arthritis. Preoperative lateral radiograph (b) of same patient. Images: Morrey B

Same patient from figures 1A,B. a) Postoperative AP radiograph 12 years after primary total elbow arthroplasty with the semi-linked Coonrad-Morrey prosthesis. b) Lateral 12 years s/p Total elbow arthroplasty. Images: Morrey B

The technique would include a very aggressive and complete release of the anterior capsule and very possibly a release of the flexor and extensor origins from the epicondyles. I would explain to the patient that there is an excellent chance of improving flexion and that extension would be only moderately improved. Additional features, of course, to consider are the shoulder and hand involvement which is almost always present.

Ken Yamaguchi, MD: Operative management for this type of patient is highly dependent on the specific radiographic findings. Additionally, the stability of the elbow is also very important to assess. A patient with a Grade II or even IIIA joint space obliteration and retained elbow stability, can be a candidate for synovectomy. If one is performed, I believe it is extremely important to retain the radial head. In addition, the physician should make sure that the patient has had maximum medical management with disease modifying agents such as methotrexate and Enbrel (etanercept, Amgen, Wyeth) etc. In patients with more significant joint space loss, especially in the context of loss of bony architecture from erosions, and/or with instability, a total elbow arthroplasty may have to be considered even in this young patient.

The treatment algorithm for a TEA in our hands is highly dependent on the radiographic findings, especially with respect to the radial head. Patients who have a reasonably intact radial head — the bony architecture is intact and free of significant erosive disease — are candidates for an unlinked total elbow arthroplasty. The type we use allows for retention of the native radial head against an anatomic prosthetic capitellum in addition to joint replacement on the ulnar humeral side (Figures 3 and 4). The retention of a native radial head on a prosthetic capitellum allows for a balanced, stable elbow and we have found the unlinked constructed to be highly successful in this population. I believe it would be particularly advantageous to have an unlinked arthroplasty in a younger female, especially in the context of rheumatoid arthritis patients who are much more active and coping with their disease much better with the advent of new disease modifying agents.

Preoperative AP radiograph (a) of a patient with severe rheumatoid arthritis. Preoperative lateral radiograph (b) of same patient. Images: Yamaguchi K

Post-operative AP radiograph (a) of patient from Figure 3 following unlinked total elbow arthroplasty. Post-operative lateral radiograph (b) of same patient. Images: Yamaguchi K

Levine: Please provide your current favored surgical approach for TEA. How do you manage the triceps and have you changed your approach over the years?

McKee: My current favorite approach for TEA is a triceps-split approach. In this approach the triceps is split directly in the midline and peeled medially and laterally off the olecranon (which can be done with a small flake of bone) as one might perform detachment of the gluteus medius in a Hardinge approach to the hip for hip arthroplasty. The triceps is sewn back to itself and through drill holes in the olecranon at the conclusion of the procedure. I have used this approach consistently in my practice and I have been happy with it although I have seen the occasional case of triceps dehiscence post operatively, typically in non-compliant patients.

At the present time my preferred elbow prosthesis is the semi-constrained Coonrad-Morrey prosthesis. The advantages of this prosthesis include its simplicity, its well-established track record, its adaptability and its immediate stability allowing early post operative motion and rehabilitation.

I do have some experience with nonconstrained prostheses. I think that some of the newer components that are available will probably find a role in elbow surgery especially for younger, more active patients with reasonable soft tissue support to enhance the initial stability of these components. I also believe that at some point in the future a hemiarthroplasty for distal humeral fractures would be ideal once the problem of immediate post operative stability is solved.

Morrey: I manage the triceps with the Bryan/Morrey triceps reflecting approach in all instances in which it cannot be left intact. What has been lost in recent efforts to reassess the technique of exposure is the reason the Bryan/Morrey approach was developed.

At Mayo, poor results, at least in our hands, were being observed with a triceps split, with the tongue of fascia turndown procedure, the so-called Van Gorder or Campbell procedure or a transverse release of the triceps with a fleck of bone. All of these procedures were considered unsatisfactory by my senior colleague, Dick Bryan. Therefore, the “in-continuity reflection” technique was developed. The original article describing this technique also reported strength and complications to be less with this exposure than with the others mentioned. For this reason we continue to favor this exposure. What has changed over the years is the reattachment.

We have always very meticulously reattached the triceps with a crisscross suture into bone and a crisscross suture down the tendon. A second transverse suture is designed to closely apply the Sharpey’s fibers to their origin on the olecranon to avoid the interposition of synovial fluid at that interface. If synovial fluid enters the site of attachment the triceps can be in continuity but the loss of a firm mechanical attachment will render the triceps extremely weak. An additional feature is the recognition that if the tissue is so thin that the triceps reflection does not appear to provide a sufficient sleeve of tissue, we will split Kocher’s interval, mobilize the anconeus, and displace the triceps anconeal mechanism from lateral to medial. This centralization prevents the extensor mechanism from slipping off to the lateral aspect of the joint which in our experience is one cause of failure of the triceps reattachment. One last point is that today we use a very specific set of locked stitches along with the bone tunnels.

Thus, the key in my mind is a technique of reattachment.

Yamaguchi: Our preferred approach is a triceps-on procedure. We keep the triceps in continuity with its insertion on the olecranon and perform the total elbow procedure through medial and lateral windows. This requires significant soft tissue release on the medial and lateral sides including complete release of the flexor-pronator origin on the medial side and the lateral extensor mechanism on the lateral side. In return, however, you have a strong triceps postoperatively and can allow early unprotected motion with the expectation of very minimal triceps weakness in the future. This has been our approach for the last 5 years and is used for both linked and non-linked situations. In general, our practice has seen a marked shift toward unlinked procedures.

As previously stated, the indications for linkage vs. unlinkage depend on the state of the radial head or whether the radiocapitellar articulation can be reconstructed. Our indication for an unlinked arthroplasty is an intact radial head or one that can be reconstructed well with radial head replacement, relatively younger or more active patients, and patients in whom reasonable stability is demonstrated at the end of the procedure.

With the prosthesis I currently use, the elbow can be easily converted from an unlinked to a linked situation if necessary at the end of the procedure or even postoperatively and therefore we are able to lean towards an unlinked prosthesis without too much risk.

Levine: How would you manage a 65 year-old RHD female with a complex comminuted intra-articular distal humeral fracture? When do you favor open reduction internal fixation (ORIF) vs. elbow arthroplasty? With the option of distal humeral replacement alone does this change your indications for ORIF vs arthroplasty? Would you still favor a total elbow over a distal humeral arthroplasty given the lack of long-term data for hemiarthroplasty?

McKee: We recently performed randomized clinical trial which is now available online in The Journal of Shoulder and Elbow Surgery, comparing total elbow arthroplasty versus open reduction and internal fixation for elderly individuals with comminuted intra-articular distal humeral fractures. I would point out that the population in our study was predominantly female and that the mean age of the individuals was 79 years of age. This emphasizes my most important point in this discussion which is that total elbow arthroplasty should be reserved only for elderly patients with this type of injury.

My basic approach to a 65 year old would be to attempt to repair this fracture with internal fixation (Figures 5 and 6). Technical tricks that would enhance the possibility of this being successful would include pre-contoured plates, countersunk headless articular screws and a “fragment specific” type of approach to obtain distal fixation which is as solid as possible to allow early motion. There may well be a role for locking plates in the care of this injury but I think that the principles I have already mentioned would be predominant over the use of locking plates alone.

Preoperative AP radiograph (a) of a 65 year old with comminuted distal humeral fracture. Preoperative lateral radiograph (b) of same patient. Images: McKee M

Postoperative AP radiograph of patient from figure 1 following open reduction internal fixation of fracture. Images: McKee M

I would do the procedure through a triceps split approach leaving the olecranon intact so that intraoperatively, if I found that I could not fix the fracture adequately, I could then convert to a semi-constrained TEA without the added complication of having to deal with an olecranon osteotomy. Other individuals, who have skill in distal humeral hemiarthroplasty for fracture, may elect to do this through an olecranon osteotomy type of approach. In our randomized trial, conversion from attempted fixation to arthroplasty occurred five times out of 20 cases randomized to fixation, a percentage which is consistent with other papers.

Frankle’s review from 2005 found fixation to be unsuccessful in approximately 25% of similar individuals. I do believe that at some point we will probably use distal humeral hemiarthroplasty exclusively for these injuries but, given the lack of data concerning this procedure, it must be considered experimental at this time. My own prosthesis of choice is the semi-constrained Coonrad-Morrey total elbow arthroplasty and I believe that, on average, in a 79-year-old individual who has this procedure performed for fracture; it will be the only operation they will ever need on their elbow. I would stress again that a 65-year-old woman could reasonably expect to live another 20 years and that this procedure is reserved for elderly individuals and I would do my best to try and repair this fracture primarily.

Morrey: In this patient if the fracture is badly comminuted and the patient is willing to accept functional limitations, I strongly favor joint replacement. This is done by leaving the triceps intact. In our experience, the technique is short and complications are fewer than what is reported in the literature for ORIF. The concern, of course, is the potential for wear, loosening or other mechanical failures of a joint replacement over time. Nonetheless, today, in spite of improved techniques of ORIF, I favor joint replacement in badly comminuted fractures since even with better philosophies of fixation, the problem with healing and avascular resorption makes this a less reliable intervention in the older patient with a badly comminuted fracture.

Today, the hemireplacement may also be a viable option. In my practice I have not performed a hemireplacement when the medial or the lateral epicondyles are fractured and require reconstruction. If, however, the attachment of the collateral ligaments is intact, and the fracture is a distal or shear type of articular fracture, then a hemireplacement is an attractive option and would be my treatment of choice.

Yamaguchi: In general, any patient younger than 70 to 75 years should be treated with ORIF whenever possible. A good ORIF is favored over an arthroplasty unless the patient is older than 70 to 75 years old. Otherwise, total elbow arthroplasty is reserved for those patients with pre-existing arthritis of the joint or when an ORIF is not possible. Total elbow arthroplasty is also generally favored over distal humeral replacement alone as these are generally going to have issues of stability unless the lateral and medial columns can be reconstructed in a reasonable fashion.

Distal humeral replacement is optimal in those patients with reasonable medial and lateral columns for whom an nonreconstructable articular surface situation is present. Shear fractures, for instance, comminution shear fractures of the articular surface are an ideal situation where distal humeral replacement may be preferred. Additionally, the patients should be relatively younger where there would be concerns about longevity of the total elbow replacement.

It should be noted that in the context of fracture, a reconstruction with the use of a distal humeral displacement is a complicated and long operation that only the most experienced surgeons probably should attempt.

For more information:

• William Levine, MD, professor of orthopedic surgery and chief of sports medicine can be reached at Columbia at New York-Presbyterian Hospital, 622 West 168th St., PH 11, New York, NY, 10032; 212-305-0762.; e-mail: wnl1@columbia.edu. He receives grant research support from Arthrex, Inc.
• Michael D. McKee, MD, FRCSC, can be reached at 55 Queen St. E., #800, Toronto, Ontario, Canada, M5C 1R6.; 416-864-5880; e-mail: mckeem@smh.toronto.on.ca. He receives institutional support from Zimmer and miscellaneous support from Stryker Biotech.
• Bernard F. Morrey, MD, can be reached at 200 1st St., SW, Rochester, MN 55905; 507-284-3659; morrey.bernard@mayo.edu. He receives royalties from Aircast, SBI, Stryker and Zimmer.
• Ken Yamaguchi, MD, can be reached at Barnes-Jewish Hospital, Washington University School of Medicine, One Barnes Hospital Plaza, West Pavilion Suite 11300, Saint Louis, MO 63110; 314-747-2534; e-mail: yamaguchi@wudosis.wustl.edu. He receives royalties from Zimmer and Tornier.

References:

• Ahmad CS, Park MC, ElAttrache, NS. Elbow medial ulnar collateral ligament insufficiency alters posteromedial olecranon contact. Am J Sports Med. 2004; 32(7):1607-1612.
• Frankle, MA, Herscovici D Jr, DiPasquale TG, et al. A comparison of open reduction and internal fixation and primary total elbow arthroplasty in the treatment of intraarticular distal humerus fractures in women older than age 65. J Orthop Trauma. 2003;17(7) 473-480.
• Kamineni S, ElAttrache NS, O’Driscoll SW, et al. Medial collateral ligament strain with partial posteromedial olecranon resection. A biomechanical study. J Bone Joint Surg (Am), 2004;86-A(11): 2424-2430.
• Kamineni S, Hirahara H, Pomianowski S, et al., Partial posteromedial olecranon resection: a kinematic study. J Bone Joint Surg (Am). 2003;85(6):1005-1011.
• McKee, MD, Hirji, E, Schemitsch, L et al. Patient-oriented functional outcome after repair of distal biceps tendon ruptures using a single-incision technique. J Shoulder Elbow Surg. 2005;14(3): 302-306.