This is my first Editor’s Choice for OrthoBuzz as new Editor-in-Chief of JBJS. I am following the example of my esteemed predecessor, Vern Tolo, who recently issued an Editor’s Choice warning about our failure to improve the management of patients with fragility fractures in terms of appropriate diagnosis and treatment of underlying osteoporosis. That is a failure of under-treatment. I want to focus on a potential issue of overtreatment.
In the July 2, 2014 JBJS, Leroux et al. describe the risk factors for repeat surgery after ORIF of midshaft clavicle fractures. The study analyzed 1,350 patients treated with surgery between 2002 and 2010 in Ontario. It is important to note that this analysis includes years after 2007, when JBJS published the seminal multicenter RCT on this topic by the Canadian Orthopaedic Trauma Society (COTS). The essence of that study was that ORIF with plate fixation results in a lower rate of nonunion and better functional outcomes predominantly in patients who have completely displaced fractures with about 2 cm of shortening or displacement.
Since that publication, we have seen an explosion in the operative treatment of midshaft clavicle fractures in North America. However, all too often the inclusion criteria derived from the seminal RCT are not referenced in individual patient decision making, and the presence of a clavicle fracture–regardless of degree of displacement–becomes an indication for surgical management.
The findings of the Leroux study should help put a hard stop to this! These researchers found a 24.6% incidence of repeat surgery in this cohort of patients. The most common reoperation was isolated implant removal (18.8%), and the incidence of major complications included nonunion (2.6%), deep infection (2.6%), pneumothoraces (1.2%), and malunion (1.1%). Risk of reoperation was increased in female patients and in those with major medical comorbidities. Limited surgeon experience increased the risk of reoperation for infection.
The orthopaedic surgery community must heed these data and act upon them. We should not misinterpret the COTS study to “encourage” a patient to opt for surgery if he or she has a midshaft clavicle fracture with less than 2 cm of shortening or displacement. The technical aspects of surgery for midshaft clavicle nonunion is not that different than that for a fresh fracture, so avoidance of nonunion must be thoughtfully discussed with the patient before recommending surgical fixation.
The bottom line that Leroux et al. provide is that surgery for a midshaft clavicle fracture is not a guaranteed success and that surgeon experience matters. And beyond clavicle fractures, let’s be sure we use our literature during shared decision making in an accurate and appropriate manner. That is a basic tenet of professionalism that we all should subscribe to.
As the lead investigator of the COTS study (1), Michael D. McKee has been instrumental in the trend to operative fixation of displaced midshaft clavicular fractures worldwide.
It is therefore with some surprise to find he is also a co-author later to a meta-analysis (2) of operative vs non-operative management of displaced midshaft clavicular fractures, in which though there is still significant evidence of lower non-union rates in operative fixation group, the authors cautioned readers that this “information should not be used to justify an indiscriminate approach to surgical fixation of all clavicular fractures”.
Obremskey in his commentary of this article (3) picked up the change in tone
and congratulated him “for his part in the initial study and for his clear view of this timely topic”.
In response to another article (4) which “do not support routine primary open reduction and plate fixation for the treatment of displaced midshaft clavicular fractures”, McKee commented that this result was “very similar and complementary, not contradictory, and some clear facts emerge.” (5). He pointed out that many will respond well to non operative treatment and stated that operative fixation has at best modest improvement in functional outcome though results are variable between studies. He suggested consideration for individual and possibly cultural difference to response to pain and disability, and asserted readers to “use this information in a clear, nonbiased fashion to assist our patients in making the appropriate therapeutic choice”
In the OTA Annual Meeting Oct 12, 2013. Dr McKee continued to stress the need for a well informed discussion and stated “the choice to proceed with operative intervention for a displaced mid-shaft fracture of the clavicle will be a decision made between surgeon and patient.”
Unfortunately, work is certainly cut out for Dr McKee as he promoted his message of moderation and shared decision-making in dealing with patients having displaced midshaft fracture. Even now, various authors continued to reanalyse data, using his research to justify the operative movement in the international community of orthopaedic surgeons with little consideration of risks and complications associated with the procedure. It takes more that a few cautionary tales (7) to rein in the massive tide of unrestrained enthusiasm for operative management of displaced midshaft clavicle fracture.
It is time to go back to the table and start a real discussion with our patients again, with better information.
References
1.Canadian Orthopaedic Trauma Society. Nonoperative treatment compared with plate fixation of displaced midshaft clavicular fractures. A multicenter, randomized clinical trial. J Bone Joint Surg Am. 2007;89:1-10.
2.McKee RC, Whelan DB, Schemitsch EH, McKee MD. Operative versus nonoperative care of displaced midshaft clavicular fractures: a meta-analysis of randomized clinical trials. J Bone Joint Surg Am 2012;94:675-84.
3. Obremskey WT. Should I Ever Fix a Clavicular Fracture? J Bone Joint Surg Am. 2012;94:e52(1)
4. Robinson CM et al. Open Reduction and Plate Fixation Versus Nonoperative Treatment for Displaced Midshaft Clavicular Fractures A Multicenter, Randomized, Controlled Trial. J Bone Joint Surg Am. 2013;95:1576-84
5. McKee MD. Displaced Fractures of the Clavicle: Who Should Be Fixed? J Bone Joint Surg Am. 2013;95:e129(1-2)
6. McKee MD The Operative versus Non-operative Treatment of Common Upper Extremity Injuries: What Does Evidence Based Medicine Tell Us?
Fractures of the Clavicle. OTA Annual Meeting Oct 12, 2013. http://ota.org/media/78916/12-McKee.pdf Accessed July 7, 20141
7. Bain GI, Eng K, Zumstein MA. Fatal Air Embolus During Internal Fixation of the Clavicle: A Case Report JBJS Case Connect, 2013 Mar 13;3(1):e24
Dear Dr. Goh- thank you for your thoughtful contribution to the discussion. I could not agree more that Dr. McKee and the authors of the COTS study have emphasized moderation in the reccomendation for ORIF for mid-shaft clavicle fractures and have repeatedly cautioned against mis-interpretation of the study. Unfortunately, it appeats that the message has not been widely heard or accepted. This dialogue may help remedy the situation.
sincerely,
marc swiontkowski
editor
After forty nine years of practice, I could not agree more, that clavicle fractures are of tenpin over treated
The 2-cm threshold for shortening and displacement may be a useful guideline, but orthopaedic surgeon Grant Jones from Ohio State pointed out in a recent item in Orthopedics This Week (OTW) that experienced surgeons reading standard X-rays don’t always agree on the extent of shortening/displacement. Sixteen orthopaedists reviewed the X-rays (AP and 30-degree cephalad views) of 30 patients with clavicle fractures. As Dr. Jones told OTW, “In the critical range of 1 to 2 cm shortening, where one decides on surgery versus non-surgical management, there was poor agreement” among these fellowship-trained subspecialists. Dr. Jones suggests that surgeons add the degree of comminution and degree of anterior and superior displacement to shortening evaluations when making the decision for or against surgical intervention. Another alternative: using contralateral clavicle radiographs to more precisely evaluate the degree of shortening.
Lloyd Resnick, developmental editor
JBJS
In a recent talk on ORIF of clavicle fractures Dr. McKee suggested that winging of the ipsilateral scapula is one of the signs he uses in deciding whether the patient would benefit from surgery.
Gabriel Gluck MD
Manassas, VA
Great choice, Marc. Operating on displaced, unstable, and shortened (> 2cm) clavicular fractures theoretically lessens complications of nonunion and shoulder dsykinesia in 1 of 6 patients (McKee). Re operation for nonunion and infected nonunion in healthy adults has ramped up my nonunion business, unfortunately. The problem with an infected clavicular nonunion is that the treatment does not generate very good results. Trying to rid the clavicle of infection, regenerate a biological friendly environment after prior stripping the vascular supply, and healing under a precarious sub q coverage is difficult at best. It would be interesting to see if problems with fixation, healing, and infection are related to avoidance of modern plating techniques (lag screw fixation with neutralization plating, compression plating, contouring plates to fit individual anatomy, and balanced fixation) and the substitution of 3.5 mm locked thicker reconstruction plates. Two OTA meetings ago generated a debate in which one of the speakers stated that he fixes all clavicular fx. On the other extreme, some (Court-Brown) only operate on the failed non operatively treated clavicular fractures/nonunions. The non operatively generated nonunion at least does not have an unfriendly environment of infection/osteomyelitis or dysvascularity from prior surgery. I hope this article generates further debate and pause. Also it would be interesting to see operative rates in people doing oral boards. Thanks.
Clifford B. Jones, MD, FACS
Clinical Professor, Michigan State University/CHM
Adjunct Professor, Van Andel Institute
In Denver Colorado we have many athletic patients with displaced, midshaft clavicle fractures who present initially predisposed to operative treatment. we find that if time is taken to present the data regarding operative vs non operative treatment, the majority select the non operative approach. patients are generally pretty smart if you give them the actual facts, data, percent of nonunions, change in constant score at one year etc.. many report to us anecdotally that they were initially told by an ED physician or Orthopedist, that the only treatment option was operative.
Thank goodness for follow up studies!
Wade Smith MD
Englewood, Colorado
When my residents present a patient with more than 2 cm of shortening, I tell them to take another film. Amazing how the measurement changes.
Michael Bosse, MD
Charlotte, NC
Mike Bosse is on target. It seems that those wanting to operate measure overlap, while others measure from the point of one fragment to the matching apex lucency of the other fragment and do not call a butterfly fragment a “Z-type” fracture.
Marc, I don’t think that clavicle fractures are the only fracture that is being over-treated by surgery these days.
I am reluctant to say very much publicly as people already accuse me of being the next Gus Sarmiento, but I must say I am disturbed by the very poorly defined indications for surgical treatment for a lot of these fracture patients, particularly distal radius fractures and ankle fractures.
I continue to argue (not too successfully!) for non-surgical management for some of these patients, but there is no question it’s an uphill battle. I think the more influential people like yourself call attention to our progressively more elastic indications for surgery the better it will be not just for patient outcome, but also for healthcare costs.
James P. Waddell MD, FRCSC
Professor, Division of Orthopaedic Surgery
University of Toronto
The following comment about possible overtreatment of FAI with hip arthroscopy and osteoplasty was compiled from an email conversation between Paul Reiss, MD and JBJS Editor-in-Chief Marc Swiontkowski, MD.
Although I have not performed hip arthroscopy and am currently retired from active practice, in the capacity of an independent medical examiner (IME) for labor & industry (L&I) cases, I have had occasion to see several patients who have had hip arthroscopy accompanied by osteoplasty for femoroacetabular impingement (FAI).
A disturbing pattern has emerged: short-term improvement followed by significant worsening over two to three years. While radiographs do not always show post-operative worsening by Tonnis criteria, MRI imaging, when performed, invariably shows worsening of the pre-surgical hip joint.
Series to date, such as JBJS 2010 (92) 8: 1697-1706, are of relatively short follow-up, and while MRI imaging was used pre-operatively in almost all of those patients, postoperatively only the Tonnis classification from plain radiographs was reported. The Harris Hip Score, also used in that study, is a valid test, but it’s based almost entirely on subjective criteria. If post-op MRIs of those patients were taken, it would be nice to know the results.
Furthermore, FAI is a “diagnosis” that has been shown to be morphologically present in more than 90% of the general young-adult population (JBJS 2013 (95) 13: e90).
I am old enough to remember the cautionary tale of the total open meniscectomy, which showed short-term subjective improvement, but long-term objective worsening. I am concerned that the current practice of hip arthroscopy for degenerative acetabular labra and osteoplasty for FAI has not been established as safe through long-term follow-up and comparison of pre-op and post-op MRI imaging.
I realize that the L&I patients I see for IMEs would be expected to have poorer subjective results, but because of that, many are getting post-op MRIs. The post-op MRIs I’ve seen are worse without exception. I can think of no reason why the MRIs of non-L&I patients would be different.
I am aware that there is considerable opinion in favor of “hip preservation,” but I worry that patients are being harmed by a procedure the effects of which have not been properly studied. While orthopedics is essentially a consensus-driven specialty, our primary focus should be to do no harm. If someone published a series that included pre- and post-op MRIs, patients and referring orthopaedists could make more-informed decisions about the risks, benefits, and tradeoffs related to these procedures.
I have had the same observation in patients who present for IME’S. The longer term outcome seems to be the recommendation for THR. This seems to be a radical end point for what would have been and probably is a “hip sprain” injury. Thanks.