Secondary Logo

Journal Logo

FEATURED ARTICLES

Editor’s Spotlight/Take 5: What Is the Best Evidence to Guide Management of Acute Achilles Tendon Ruptures? A Systematic Review and Network Meta-Analysis of Randomized Controlled Trials

Leopold, Seth S. MD1

Author Information
Clinical Orthopaedics and Related Research: October 2021 - Volume 479 - Issue 10 - p 2113-2118
doi: 10.1097/CORR.0000000000001953

The number of operations that all orthopaedic surgeons can do is decreasing as our profession grows increasingly subspecialized by the year [3]. But there remain more than a few operations that most call-taking surgeons would rather keep than hand off in the morning. Straightforward ankle fractures and hip fractures come right to mind.

As do Achilles tendon ruptures. I’m certain that foot and ankle experts believe they’d tackle these more aptly than would, say, the average ham-fisted arthroplasty specialist (ahem), just as the latter might move more capably through an on-call hip case. They’re likely both right, but each probably still can work the other side of the street pretty well.

For that reason, this month’s Editor’s Spotlight article should receive a lot of attention. It answers a question common to the practices of many readers: Which approach is best for treating a patient with an acute Achilles tendon rupture [6]? It derives its answers using the most robust of research approaches: the network meta-analysis.

Meta-analyses add a measure of external validity—that is, generalizability—to randomized trials by pooling results across diverse study centers. This is important because the randomized controlled studies that fuel meta-analyses often suffer from all of the shortcomings of local study designs: They may be small, there may be important differences in technique across surgeons or centers, and each hospital’s population may differ in important ways from those of the other included studies. Meta-analyzing, that is, pooling data across studies from a variety of surgeons and centers, can offset these limitations.

But meta-analyses themselves have an important shortcoming: They can pool data only from studies that compared interventions directly, head-to-head. If you want to compare treatment A to treatment B, or treatment B to treatment C, and there have been randomized trials making those comparisons, you can meta-analyze them. But let’s imagine you want to compare treatment A to treatment C, and no individual trials have compared them directly. If the only tool you have is traditional meta-analysis, you’re stuck.

This bit of real-world rubber-meets-road is where network meta-analyses shine. This study design lets us compare results of treatments that have not been evaluated head-to-head. In the above example, a network meta-analysis would allow us to learn whether or not treatment A outperforms treatment C. In this month’s Clinical Orthopaedics and Related Research®, a group that includes both clinical and methodological subspecialists from the University of Ottawa under the guidance of internationally known methodologist and medical sociologist Ian D. Graham PhD and surgeon Brad Meulenkamp MD, found that most treatments that a surgeon is likely to offer a patient in this setting—functional rehabilitation, conventional open surgery, or so-called minimally invasive surgery (MIS)—do not differ from one another in terms of the risk of rerupture. The only treatment with a higher risk of rerupture was primary immobilization. They also found that MIS repair had a lower risk of complications resulting in surgical treatment than any other intervention evaluated, though I think that finding needs to be probed just a bit. Stick around a moment for more on that.

Usually, in our Editor’s Spotlight/Take 5 section, I interview one author of the featured paper. Because this one has so many facets—on-call realities and surgical techniques, but also how evidence is disseminated and why it so often suffers from slow uptake—I’m going to interview two of them, Dr. Meulenkamp for his perspectives on surgical practice and Dr. Graham for his thoughts on where the science comes from, how we use it, and why we sometimes don’t.

If reading network meta-analyses is new to you, it’s worth coming along for the ride on this one [6] for several reasons. First, the good ones—like the one in this month’s CORR®—aggregate the best-available evidence: high-quality randomized trials. Then, the well-reported ones—again, like this one here—help the reader to understand how much confidence to have in comparisons that were not made in any head-to-head way, with easy-to-understand visuals (Fig. 1). They also provide metrics that make sense to nonmethodologists, like the number of patients one would need to treat with one approach to prevent a complication had some other choice been made (called the number needed to treat), and plain-language characterizations of effect direction (words like probably superior, possibly superior, definitely inferior, and the like). They also adjust the strength of their recommendations in light of the quality of the source studies, which, in the current study was variable, but on the key topic of complications, was pretty good.

Fig. 1
Fig. 1:
The network geometry for risk of rerupture, taken from Meulenkamp et al. [6], is shown. The reader can get a clear sense for how robust the comparisons are, both in terms of the numbers of studies making any comparisons a reader might be interested in and how many patients inform those comparisons. Node size is proportional to the number of participants in the specified treatment arm and is indicated by “n =” below the treatment name. Greater edge (connecting line) thickness denotes an increasing number of studies informing an indicated comparison and is specified with the number adjacent the edge. Reprinted with permission from Meulenkamp B, Woolnough T, Cheng W, et al. What is the best evidence to guide management of acute Achilles tendon ruptures? A systematic review and network meta-analysis of randomized controlled trials. Clin Orthop Relat Res. Published online June 28, 2021. DOI: 10.1097/CORR.0000000000001861.

I began by saying that most of us probably can (and many of us probably do) repair Achilles tendons; I’m guessing most of us do not use MIS approaches. Do we need to? Read the Take 5 interview with Dr. Meulenkamp (foot and ankle expert) and Dr. Graham (medical sociologist and methods guru) that follows to find out. In the interview, we’ll also cover how and why high-quality evidence—like the evidence in this month’s Editor’s Spotlight article [6]—does not always improve patient care as much as it ought to, and what we can do to fix that important problem.

Take 5 Interview with Brad Meulenkamp MD, FRCSC, and Ian Graham PhD, FCAHS, FNYAM, authors of “What Is the Best Evidence to Guide Management of Acute Achilles Tendon Ruptures? A Systematic Review and Network Meta-Analysis of Randomized Controlled Trials”

Seth S. Leopold MD:Congratulations on an important and well-presented study. If you can, put yourself in the shoes of readers who are call-taking general orthopaedic surgeons. Do your findings mean that they shouldn’t repair Achilles tendons unless they use MIS approaches? And, while you’re on that subject, what does “MIS” mean, anyway? In many parts of our specialty, that definition may be—at least to some degree—in the eye of the beholder. It’s been my experience that to the person feeling the sharp end of the knife, many MIS operations feel plenty “invasive.”

Brad Meulenkamp MD, FRCSC: Maybe it’s best to work backwards on these questions. When we sat down to hash out our study design, the definition of “MIS” was an important consideration because we needed to ensure we were pooling apples with apples. Otherwise, it makes it very tough to contextualize and generalize the study findings to surgeons' practices. Open surgical approaches were defined as those in which a longitudinal open incision was used with paratenon reflection to allow end-to-end Achilles repair. The limited (MIS) approaches included those methods that avoid this full open reflection that have been more traditionally associated with wound complications, and included limited transverse incision techniques, device-assisted, and suture-shuttling techniques. Certainly, these may not be felt by the patient to be any less invasive, but the reasoning was to separate or compartmentalize methods that have been developed to avoid the wound-healing problems that we have seen in the past and certainly impact the patient’s surgical experience.

To your first point, I am that call-taking surgeon! I’m not one to use “always” or “never” statements, and I think context is everything. I have little doubt that many surgeons have great results with open surgical repairs, and I think the data back that up. And full-disclosure, until recently, if I was repairing an Achilles, I would have been an open-repair guy. It’s what I was taught and what I know, and I didn’t feel the need to learn a new technique. But it’s those few times that you run into the wound dehiscence or necrosis or infection that are so impactful on those unfortunate patients. And so, if I know there is a tool that can help minimize those issues, minimize return trips to the OR, while providing at least as good an outcome, I feel now I have that reason to evolve in my management.

Dr. Leopold:I found several of your findings somewhat surprising, which is a bit unusual in the context of a data-pooling exercise, but I’m curious: What surprised you most, and, more importantly, how have your discoveries changed either how you practice or how you teach (whether residents or general orthopaedic surgeons) about the treatment of Achilles tendon ruptures?

Dr. Meulenkamp: As a surgeon who treats many of these injuries (and I like to think I know the evidence about them reasonably well), the most surprising to me was the parity between treatment arms. Perhaps therein lies my surgeon bias, but having all the data on the table, I am quite confident that the rerupture risk debate has been settled for these injuries and that surgery should not be offered for that reason alone, as long as a valid functional rehabilitation program can be offered and followed by the patient.

This is the message I have been pushing on our trainees, but I also still urge them to take a balanced approach when discussing options with patients. While I feel we can state similar rerupture risk with relative certainty, I also think shades of gray remain when we look at other outcomes, particularly when it comes to return of strength. I haven’t taken surgery off of the discussion table yet, but it continues to lose ground in my practice and teaching. As to how I have changed, as mentioned above, for the minority of patients who undergo Achilles tendon repairs in my practice, I have shifted to an MIS approach. I think there are enough data to convince me that if there is an acute Achilles to be repaired with surgery, I can do my part to help prevent a return trip to the OR with this approach.

Dr. Leopold:Dr. Graham, I’ve seen that a large part of your research program focuses on how the best-available evidence sometimes—often—does not wind up influencing or even informing practice very much[1, 7]. What can journals do differently to help the discoveries we publish, like yours this month, have a larger impact on patients’ lives? And are there other entities apart from journals that have a role to play in increasing the effectiveness of this kind of knowledge translation?

Ian Graham PhD, FCAHS, FNYAM: Great questions. When thinking about how to influence the uptake of research findings by patients, healthcare providers, or by any target audience, some questions to ask include: What is the knowledge translation goal? Is it to simply increase awareness of the discovery? Is it to influence patient or provider decision making (that affects their behavior/practice)? Is it to influence policy making? Is it to enlighten researchers and influence the future research they conduct?

To determine the knowledge translation goal, one needs to determine whether the discovery is reliable and valid (does the research need to be replicated or is it ready for prime time?) and whether the strength of the evidence for the discovery is sufficient to justify the goal (is the evidence of sufficient strength to warrant changing behavior/practice/policy?). Inappropriate widespread dissemination of premature findings can be counterproductive and even cause harm. Therefore, we should always be thinking about judicious knowledge translation.

From a knowledge translation perspective, we often consider knowledge syntheses (like our systematic review and network meta-analysis) as the most appropriate form of evidence to justify the goal of changing behaviors, practices, or policies as they contextualize and integrate research findings within the global body of knowledge on the topic. As you have already pointed out, this reduces the likelihood of being misguided by the findings from individual clinical trials. So, if you are interested in discoveries impacting patient lives in positive ways, the journal should continue to prioritize the publishing of all types of knowledge synthesis (scoping reviews, systematic reviews, network meta-analyses, overviews of reviews [2]) as these should usually be the basis for changing practice/policy, which then leads to improved patient and health system outcomes or impacts.

What else can a journal do? Using the results of the network meta-analysis, Dr. Meulenkamp is now developing a decision aid for patients to help them decide the best approach to managing their Achilles tendon rupture based on, what we now know to be, the best evidence on the benefits and harms of the different options. Journals could encourage authors of knowledge syntheses to also produce decision aids or other decision tools that could accompany their knowledge synthesis papers. Patients would likely want these to be open access, easy-to-find, and of high quality. Obviously, additional strategies other than publication would also be needed to disseminate these decision aids to patients and healthcare providers. Specifically thinking of knowledge syntheses, I would suggest asking your readers what would make these types of papers more useful to them. For example, would they like actionable messages and recommendations to be provided with articles? What about dissemination strategies of key messages using social media? Are there derivative products (infographics, 2-minute video lectures, apps) that should be created to make it easier for them to “use” these discoveries?

In terms of other entities playing a role to increase knowledge translation, I will shamelessly plug the Ottawa Hospital Research Institute Decision Centre [8], which has the largest inventory of quality assessed patient decision aids in the world. For anyone looking for a patient decision aid, this is the site to start with.

Dr. Leopold:Dr. Meulenkamp, in a sense, rerupture was “counted twice” in your analysis: Rerupture was its own endpoint, and it also was the diagnosis behind a large majority of the complications resulting in surgery, where it was counted again. That being so, I wondered if the differences favoring some of the surgical treatments in your paper were perhaps overstated a bit. Can you inject a dose of balance here? How much better is surgery, really, than functional rehabilitation, and based on your discoveries, in what clinical situations might we perhaps be doing “too much surgery,” such that functional rehab may be the better choice for patients with this injury?

Dr. Meulenkamp: This is exactly the balance that I (and I think many surgeons) find so challenging to strike with patients, and I don’t think we can answer that question with our—or any other—Achilles study to date. We attempted to measure discrete outcomes that were important and impactful to patients. Reruptures and other complications that result in subsequent surgery both are important events that derail a patient’s recovery and rehabilitation timeline, and these outcomes were not intended to be additive, but individually significant to patients.

My main disappointment with our study was that we were unable to perform a meaningful pooled analysis of clinical and functional outcomes (they are a bit all over the place). But while we were not able to pool the functional outcomes, when you look at strength outcomes, rehab protocols never outperform the repair options. Granted, several studies demonstrate no differences between arms, but several demonstrate that repair options outperform functional rehabilitation. I don’t think this would be surprising to many surgeons; after all, we don't treat many professional athletes with functional rehab. But how much these strength differences matter to our individual patients is what’s so difficult to translate.

The authorship collaborations on this paper were strategically designed to bring a team together that can hopefully help bridge this knowledge gap: Who will most benefit from surgery on their Achilles tendon rupture? We are now bringing patients aboard who have been treated with these various techniques to help translate how the patient experiences the outcomes we have quantified. After all, they are the real experts. We will be integrating the patient experience with the data synthesis from this paper to develop a patient decision tool, aligning what we know (and don’t know) about outcomes of interest with patient preferences. Our orthopaedic shared decision-making program is in its growth phase and we are excited to be able to help patients with Achilles tendon ruptures make this decision.

Are we doing too much surgery? I’m not sure I’m in a position to really say. Based on our data, I would expect that functional rehab should be (and is) acceptable for most patients. Avoiding surgery is usually a win for patients. But I do imagine the uncertainty surrounding functional outcomes and return of strength will drive a subset of patients toward surgery. I think the important thing is to empower patients to make the right decision for themselves.

Dr. Leopold:Back to you for the last word, Dr. Graham. From what I can tell, your work is intensely collaborative. On this paper, you’ve got—among others—surgeons, MD-PhDs, a librarian with an advanced degree, and a medical sociologist. Can you share with our younger readers who want to emulate your approach to answering big questions some of the lessons you’ve learned about how to create, sustain, and incent the best performance from multidisciplinary teams?

Dr. Graham: Just so no one feels left out, the team also includes a PhD biostatistician, PhD epidemiologist, and a PhD-prepared nurse, who is our shared decision-making and patient decision aid guru, to use your words.

Setting up teams specifically to address the research questions of those who would use the study discoveries creates the foundation for success. Knowledge users might include patients, healthcare providers, or managers of healthcare systems. Those knowledge users need to be part of the research team. There is growing acceptance that this approach, which is referred to by such terms as research co-production, integrated knowledge translation, participatory research, patient engagement, or patient and public involvement, is more likely to produce discoveries that are useful, useable, and utilized by knowledge users, and this translates into impact more quickly [5].

I have also learned to move from the concept of a multidisciplinary team (people from different disciplines working together, each drawing on their disciplinary knowledge) to an interdisciplinary team (integrating knowledge and methods from different disciplines, using a real synthesis of approaches) [4]. This is a critical shift in mindset because we tend to be socialized within our disciplines to think our own discipline knows best rather than thinking about how integrating perspectives across disciplines may lead to novel discoveries.

Here are a few more things that I have learned about how to improve interdisciplinary teamwork:

  • Health research is a team sport. This means continuously supporting, coaching, and nurturing the team. This work lasts the life of the project and must be grounded in good communication practices.
  • When pulling teams together, think about diversity and the needs of the project. Include patients and their caregivers who can make valuable contributions because of their lived experiences. Avoid working with people who are unduly competitive, difficult, unpleasant, untrustworthy, or uncharitable. Life is too short.
  • Create an environment where each team member is respected for what they bring to the team (their knowledge, skills, contributions). Treat team members as equals rather than seeing them as simply providing necessary services.
  • Demonstrate scientific and cultural humility. Being humble means it is okay to not know everything—this is why you have all the people you have on the team. It is also okay to admit you don’t know everything, which can be hard for some disciplines.
  • Meaningfully engage team members. This means treating them as research partners and creating safe places for discussion. Listen to each other. Find ways to integrate sometimes differing perspectives. Give the team some time to gel.
  • Avoid the use of disciplinary jargon. Instead, breakdown traditional disciplinary power and authority hierarchies and share decision-making power within the team. This does not mean that all decisions must be consensus or unanimous decisions, but it does mean all viewpoints are discussed and considered before decisions are made.
  • Include trainees on the team so that they can see how productive collaborative teams work and what is required to make them work. Training programs seldom include anything on team formation or functioning, so they need to learn via role modeling.
  • Set out to make the work as enjoyable as possible. Share your sense of humor. People usually prefer working with people who are fun to work with.

If you are successful in creating and nurturing these conditions, the team will thrive and be motivated to work together. And, the research will generate answers to those big questions that may be applied in practice.

References

1. Banner D, Bains M, Carroll S, et al. Patient and public engagement in integrated knowledge translation research: are we there yet? Res Involv Engagem. 2019;5:8.
2. Canadian Institutes of Health Research. A guide to knowledge synthesis. Available at: https://cihr-irsc.gc.ca/e/41382.html. Accessed August 5, 2021.
3. Horst PK, Choo K, Bharucha N, Vail TP. Graduates of orthopaedic residency training are increasingly subspecialized: a review of the American Board of Orthopaedic Surgery part II database. J Bone Joint Surg Am. 2015;97:869-875.
4. Jensenius AR. Disciplinarities: intra, cross, multi, inter, trans. Available at: https://www.arj.no/2012/03/12/disciplinarities-2/#:∼:text=Multidisciplinary%3A%20people%20from%20different%20disciplines,a%20real%20synthesis%20of%20approaches. Accessed August 5, 2021.
5. Kothari A, McCutcheon C, Graham ID. Exploring the frontiers of research co-production: the integrated knowledge translation research network concept papers. Available at: https://www.biomedcentral.com/collections/IKT.. Accessed August 5, 2021.
6. Meulenkamp B, Woolnough T, Cheng W, et al. What is the best evidence to guide management of acute Achilles tendon ruptures? A systematic review and network meta-analysis of randomized controlled trials. Clin Orthop Relat Res. 2021;479:2119-2131.
7. Nguyen T, Graham ID, Mrklas KJ, et al. How does integrated knowledge translation (IKT) compare to other collaborative research approaches to generating and translating knowledge? Learning from experts in the field. Health Res Policy Syst. 2020;18:35.
8. The Ottawa Hospital. Patient decision aids. Available at: https://decisionaid.ohri.ca/. Accessed August 5, 2021.
© 2021 by the Association of Bone and Joint Surgeons