Operative versus non-operative treatment for closed, displaced, intra-articular fractures of the calcaneus: randomised controlled trial

Dear Sir,

We read with interest the article by Griffin et al, BMJ 2014;349 (24 July 2014) and would like to make the following points:

One of the aims of fixing calcaneal fractures is to minimise the late morbidity from malunion, subsequent arthritis and eventual fusion. The average patient in the group studied will likely live for around 30 years after fracture. A significant number of cases requiring fusion would present after the two year follow up of this study. Late reconstruction surgery rates may differ between the two groups. Did the authors consider longer follow up to allow evaluation of the late outcome?

The authors stated that CT scans showed intra-articular reduction up to a 2mm step. They also described less than perfect reductions in their cohort. Considering this, what would the results of surgery have been for fractures with a better reduction? We would be interested to see the long term results for the 40 patients who had less than a 2mm step after ORIF compared to the non-operatively managed group.

In this paper the definition of a specialist surgeon was not made clear. Including surgeons ‘recognised as specialists’ was in our opinion not sufficient. We would suggest that for this paper a specialist is a fellowship trained consultant foot and ankle surgeon who performs a high number of foot and ankle procedures each year and who has passed the learning curve for these operations. A median of two operations for each surgeon was surprisingly low. This may indicate that many included surgeons were not specialists as we define them.

We note that the selection criteria included only 502 of 2006 fractures encountered. It is difficult to know what the true selection criteria for operation were. Considering this and the above points we find it hard to agree with the sensationalist headline ‘surgery no longer justified for most intra-articular calcaneal fractures’. In our opinion this was a failure of the BMJ’s editorial process. In a journal widely read by non-specialists such a declaration was misleading. The true conclusions of the study were more nuanced: severe and open fractures require surgery. The important long term morbidity could not be determined from this study.

Finally, the trial was described as pragmatic. The word ‘pragmatic’ was mentioned 6 times in the text. We would welcome a definition of this term and how it affected the study.

Yours sincerely,

E Lindisfarne
D Wilson
B Rogers
D Ricketts

Mr L. Parker, Senior Foot and Ankle Fellow, North Bristol NHS Trust
Mr I. G. Winson, Consultant Orthopaedic Foot and Ankle Surgeon, North Bristol NHS Trust, Past President European Foot and Ankle Society, British Orthopaedic Association Vice President Elect

We read with interest the report by the UK Heel Fracture Trial, after noting the rather alarming headline on the cover of the BMJ “Calcaneal fractures: surgery provides no benefits”.(1) In a journal with a wide readership, including the general public and journalists from more mainstream media there is a need for responsible publishing, not for headlines which will influence the thoughts of commissioners yet bear no resemblance to the scientific content of the journal. The results of the UK Heel Fracture Trial have been hotly anticipated but probably do little to guide our treatment of closed displaced calcaneal fractures. The trial certainly does NOT conclude that “surgery provides no benefit” and actually reviews a very specific group of selected patients with fractures that one would expect to have had acceptable outcome regardless of the treatment.

The major problem with the work is the cases. As highlighted in the full on-line article only 502 of 2006 patients with calcaneal fractures were deemed eligible for randomization in this trial and 151 of those agreed to take part in the study- a total of 7.5% of all of the calcaneal fractures attending the centres involved in the study. The trials exclusion criteria removes the outcomes of a significant volume of what would be considered “run of the mill” calcaneal fractures, such as those with gross displacement and ipsilateral lower limb trauma. The small number of fracture cases remaining after the exclusion criteria and the participation rate of 30% are reflected in the fact that 27 surgeons in 22 different hospitals only operated on a median of 2 fractures for this study. In light of this, it is a complete mis-representation of the research to publicise the trial on the cover of the BMJ as though its conclusions apply to all calcaneal fractures.

The infection rate of 19% of the surgically treated patients is hugely worrying. This is by no means the standard wound complication rate and raises questions about the nature of the surgery performed. Furthermore, with the advent of arthroscopically-assisted surgery and minimal access reduction techniques we have advanced considerably in minimizing the impact of the soft tissue trauma and consequent wound complications and infections that have blighted this type of surgery in years past.

In contrast to previous observations that initial fracture severity and fracture severity classifications do correlate with clinical outcome (2) the authors were unable to power the trial to investigate this, instead estimating through regression analysis but managing to draw the conclusion that there was no evidence (p=0.697) that the effect of surgery was affected by fracture severity. It is completely inappropriate to conclude this based on the selecting out of the more severe fractures requiring surgery as this trial has done.

One of the main justifications for undertaking surgery on displaced intra-articular calcaneal fractures is that surgery may delay the onset of disabling subtalar joint osteoarthritis. CT scanning was undertaken to assess the adequacy of reduction post-operatively in 72% of patients who had undergone surgery. We question the threshold of “no more than 2 mm articular step” in the subtalar joint as being indicative of a satisfactory reduction- indeed the indication for surgery itself would include such joint displacement. Considering there is evidence that restoration of Bohler’s angle and articular surface congruency may provide superior outcome , this data requires further interrogation.

The authors do highlight that their results may change over the duration of patient follow-up. In a similar muticentre randomized controlled trial in Sweden, clinical scores between operatively and non-operatively treated calcaneal fractures were similar at one year, however in the same cohort of patients there was a tendency toward improved Visual Analogue pain scores and the physical component of the SF-36 score at 8-12 years follow-up in addition to reduced incidence of radiographically detectable subtalar joint osteoarthritis in the operated group (a relative risk reduction of 41%).(3) A similar post hoc analysis of the Swedish study group has shown that there is a statistically significant correlation between the best results and surgery at 10 years.(4) Clearly therefore based on the two year results of the UK Heel fracture trial it is inappropriate to conclude that surgery provides no benefit to the treatment of calcaneal fractures.

References:

1: Operative versus non-operative treatment for closed, displaced, intra-articular fractures of the calcaneus: randomised controlled trial. BMJ 2014; 349 doi: http://dx.doi.org/10.1136/bmj.g4483 (Published 24 July 2014). BMJ 2014;349:g4483

2: Rammelt S, Zwipp H, Schneiders W. et al. Severity of injury predicts subsequent function in surgically treated displaced intraarticular calcaneal fractures. Clin Orthop Relat Res. 471(9):2885-98, 2013

3: Agren PH, Wretenberg P, Sayed-Noor AS. Operative versus non-operative treatment of displaced intra-articular calcaneal fractures- a prospective, randomized, controlled multicenter trial. J Bone Joint Surg Am 95: 1351-7, 2013

4: Agren PH, Mukka S, Tullberg T, Wretenberg P, Sayed-Noor AS. Factors Affecting Long-Term Treatment Results of Displaced Intra-Articular Calcaneal Fractures. A Post-hoc Analysis of a Prospective, Randomized, Controlled Multicenter Trial. J Orthop Trauma. 2014 May 12. [Epub ahead of print]

I was involved in discussions about the format of this trial and we stated that the question which needed answering was whether the well reduced calcaneal fracture did better than the displaced, intra-articular fracture treated non operatively. Part of the protocol was to have a post operative CT scan to ascertain this. It is disappointing that the trial has not answered this important question. These fractures are difficult to fix well and there is a learning curve. Unless we know that the well reduced calcaneal fracture does no better than non operative treatment we cannot condemn operative treatment. Professor Griffin would argue that the study shows the results in the hands of the average Orthopaedic Surgeon interested in these fractures. That is not a scientific answer to whether the problem is the Surgeon or the surgery. Interest or experience is not a guarantee of a good reduction; we need a scientific answer. Regrettably the trial still does not answer the fundamental question.

In this study, what the authors have nicely demonstrated is that calcaneal fracture surgery by surgeons who do not perform high volumes of this complex surgery does not yield outcomes that are significantly better than non-operative treatment, and with higher complications. The methodology of the current study, allows it to only address a specific segment of the discussion, that being the advisability of surgery by non-specialist surgeons (defined as high-volume calcaneal fracture surgeons) to operate on calcaneal fractures. In this subset, their data strongly suggests that such surgery is not advantageous. However, it needs to be made clear in the manuscript that this result cannot be extended to the include surgery by high-volume specialist surgeons.

The reported distribution of treatment centers means that for at least half of the centers, the annual number of calcaneal fractures treated annually was 1.7per center. If there were more than a single surgeon participating at each center, then this number would be even lower when expressed as "annual rate per surgeon". It is hard to argue that surgeons who are so infrequently operating on calcaneal fractures should be considered "experts", since the literature has clearly shown a learning curve that extends well beyond a surgeon's first 100 cases.

The distribution of cases by quartile is interesting. The lowest half of the centers (referred to above) accounts for 55 cases. The next quartile accounts for approximately 42 cases (assuming an average of 7 cases per center - which means 2.3 cases/year/center). The top quartile therefore accounts for 54 cases, which average out to 12 cases/center, or 4 cases/center/year. In the most extreme configuration, where most of the cases came from one canter (8,8,8,8, and 30 cases per center), only one center would have what might be considered a high volume of cases (10 per year). Since precise numbers are not provided by the authors, some of the calculations here may be a bit off, but the conclusion doesn't change regarding the large number of centers that have very low numbers of cases. This makes the proposition that the care rendered is representative of the highest quality in "expert surgical treatment" very difficult to make. If the authors want to suggest that calcaneal fracture surgery by non-expert surgeons is problematic, that would be supported by their data.

The data about the surgeons paints a similar picture regarding the level of expertise of the surgeons. In such a complicated surgery, it is well known that higher volumes of surgery on annual basis lead to better outcomes (in the field of Orthopaedics this has been best studied for total joint replacement). In the present study, no surgeon operated on more than 6 patients over the course of 3 years. That does not constitute the level of experience that would characterize an expert in calcaneal fracture surgery.

The comment that this study reflects the expected outcome of surgery in "the real world" actually makes the point that these surgeries should not be done by surgeons who do not do these surgeries in high volume. It says nothing about the outcomes one would see if the surgeries were performed by truly expert surgeons with a high ongoing experience treating such fractures surgically.