Compared To What? Examining The Evidence With Diabetic Ankle Fractures
In this column, we will attempt to broadly, but specifically and critically, evaluate the medical literature with respect to the treatment of diabetic ankle fractures and answer a seemingly simple clinical question: “Do ankle fractures in patients with diabetes mellitus (DM) have higher complication rates?”
Upon initial consideration and with a dose of common sense, it would seem that there is a relative preponderance of literature indicating that the treatment of ankle fractures in patients with DM is indeed associated with higher complication rates. But what exactly is this literature telling us? In order to address this from an evidence-based standpoint, it is important to consider what type of comparison is in question. “Compared to what?” is always a simple but incredibly important question to ask as a critical reader of medical literature.
The best answer to our initial question would theoretically be provided by a direct comparison of a large group of patients with ankle fractures, both with and without a diagnosis of diabetes. A retrospective comparison of these groups would probably be considered level 3 evidence (retrospective case control study) based on the broadly accepted levels of clinical evidence.
Perhaps somewhat surprisingly, one of the best study examples to address our question is a 2019 publication by Liu and colleagues.1 Utilizing a large U.S. database, the study authors retrospectively assessed 17,464 patients who underwent open reduction and internal fixation (ORIF) of ankle fractures, and subsequently divided these patients into 2,044 patients with DM and 15,420 patients without DM. Liu and coworkers found that patients with DM were at least three times as likely to have an unexpected hospital readmission within 30 days of the procedure, develop a surgical site infection, have a wound dehiscence, and/or undergo a repeat operation in comparison to patients without DM.
Contrast this comparison to a similar but fundamentally different 2016 study by Dodd and colleagues.2 Using a large U.S. database of ankle fractures, the authors once again found that patients with DM were at an increased risk (1.43 times as likely) of 30-day complications following ORIF. However, the specific comparison in this study was different and should influence our interpretation of the results. Dodd and team primarily compared ankle fractures with and without a complication. Diabetes mellitus was simply one of several variables that researchers assessed as potentially leading to a complication.
In other words, based on the direct comparison, the study by Liu and colleagues more effectively answers our clinical question than the study by Dodd and coworkers.1,2 While both studies provide potentially important information in a level 3 design, as critical readers, we should seek the most direct possible comparison. From an evidence-based standpoint, the study by Dodd and team provides secondary evidence that DM is a risk factor for the development of a complication.2 The investigators could then follow this up with another study attempting to primarily and directly analyze this potential risk factor. In fact, Liu and colleagues did exactly that, demonstrating good progression and development of a clinical hypothesis within our literature.1
So is this our final answer? Based on the results of the Liu and colleagues study, can we feel confident educating our patients with a diagnosis of DM and an ankle fracture that they are at least three times as likely to experience a complication because of their diagnosis? Well, not quite. After all, simply considering the “diagnosis of DM” as a single variable casts a fairly wide net, doesn’t it? In fact, we know from clinical experience that patients with diabetes have different levels of glycemic control and differing spectrums of microvascular complications for example. Accordingly, it would be prudent to further refine the inclusion criteria of our comparison in order to provide further clarification of our initial question.
What About The Impact Of Glycemic Control?
A good example of this is the 2013 work by Liu, Ludwig and Ebraheim, who assessed the impact of glycemic control (as measured by hemoglobin A1c (HbA1c)).3 Specifically, they compared diabetic ankle fractures with an HbA1c less than 6.5 percent to the same cohort with an HbA1c greater than 6.5 percent. The researchers also considered HbA1c as a continuous variable allowing for calculation of a correlation coefficient. The authors concluded that higher HbA1c values were associated with poorer radiographic and clinical outcomes, a finding that has been similarly corroborated in other investigations.4-7
While the study by Liu, Ludwig and Ebraheim is still level 3 evidence, changing the specific comparison again changes our interpretation of the results.3 By refining the inclusion criteria of the cohort, we can take a step past a conclusion that “the diagnosis of DM leads to higher complications” and also now conclude that “not all diabetic ankle fractures are the same, and that glycemic control is likely to have some effect on outcomes.” Articles such as these with a comparison specifically involving glycemic control represent a step toward developing an elusive “threshold” number in which the risks potentially outweigh the benefits of a given procedure.
When There is No Surgical Intervention For A Diabetic Ankle Fracture
However, critical readers should use caution here. The comparisons in the aforementioned studies have ignored a potentially large patient subset: those with DM and an ankle fracture who do not have surgical intervention. Additionally, at this point in our discussion, we have only considered literature evaluating surgical interventions. With this focus, it can be easy to make the mistake of assuming that non-surgical interventions are without risks of their own.8
Although there are many potential negative outcomes to consider here, including the development of post-traumatic arthritis, delayed and non-unions, functional outcomes, cast complications, etc., one consistent complication concern within the literature on diabetic ankle fractures is the progression from acute fracture to the development of Charcot neuroarthropathy. Unfortunately, we have very few comparative studies of operative versus non-operative intervention in the literature from which to draw definitive conclusions.8
Instead, most of the literature are level 4 retrospective case series that describe progression to Charcot in a single group of patients. For example, Jones and colleagues retrospectively looked at 42 patients with DM and an ankle fracture undergoing surgical intervention, and found that eight progressed to Charcot neuroarthropathy.9 Conversely, Holmes and Hill followed 20 lower extremity fractures in patients with DM treated non-operatively and found that eight progressed to Charcot neuroarthropathy.10 However, these individual studies do not allow for a direct comparison between the two different interventions because of the potential for selection bias.
In other words, it is not prudent from an evidence-based standpoint to assume that both studies started out with the same type of patients and it might be more likely to assume that there must have been differences between the groups that caused the physician to initially recommend operative versus non-operative treatment in the first place. Directly comparing results might be more likely to evoke differences in the presenting characteristics of the patients than actual differences between the two interventions.
This might be more basically understood as an indirect “apples to oranges” comparison. Although level 4 retrospective case series are very common, it is very difficult to answer the question of “Compared to what?” within this methodology.
Does The Literature Provide Definitive Guidance On Fixation Options?
There is a similar challenge when considering different surgical fixation options for ankle fractures in patients with DM. It is a reasonable hypothesis that stronger fixation constructs for this group of patients might reduce or even negate the observed higher rates of complication. It is relatively commonly accepted that one should “double up” the amount of fixation for patients with diabetic ankle fractures (see first photo above).11-18
However, little comparative information is available to help critically evaluate the efficacy of one specific technique versus another in patients with diabetic ankle fractures. Perhaps the best example of this is a 2011 study by Wukich and team, who primarily compared outcomes of ankle fracture ORIF between patients with uncomplicated and complicated diabetes.4 This retrospective review noted the use of several different fixation constructs including standard internal fixation, advanced internal fixation, internal fixation combined with external fixation and external fixation alone. Wukich and colleagues even performed an after-the-fact comparison of these techniques, but similar to our first discussion point, this was not the intended primary comparison of the investigation.
Although this study does provide some interesting information and a potential avenue for future investigations, it is also at risk for selection bias because of the retrospective nature of data collection. In other words, there must have been some difference between the patient and fracture characteristics that caused the surgeons to opt for one fixation option over another. Therefore, a direct comparison between techniques is a bit “apples to oranges.”
Concluding Thoughts
Although it is reasonable to conclude that ankle fractures in patients with DM are associated with a higher rate of complications, we still have much to learn regarding the specific risks of specific interventions for this challenging patient cohort. Critical readers should always consider not only what information a specific investigation is providing but also how the specific comparison that researchers are performing affects the interpretation of results.
Dr. Mateen is a second-year resident with the Temple University Hospital Podiatric Surgical Residency Program in Philadelphia.
Dr. Heineman is a first-year resident with the Temple University Hospital Podiatric Surgical Residency Program in Philadelphia.
Dr. Meyr is a Clinical Professor in the Department of Podiatric Surgery at the Temple University School of Podiatric Medicine. He is the Program Director of the Podiatric Surgical Residency at Temple University Hospital in Philadelphia.
1. Liu JW, Ahn J, Raspovic KM, et al. Increased rates of readmission, reoperation, and mortality following open reduction and internal fixation of ankle fractures are associated with diabetes mellitus. J Foot Ankle Surg. 2019;58(3):470-474.
2. Dodd AC, Lakomkin N, Attum B, et al. Predictors of adverse effects for ankle fractures: an analysis of 6800 patients. J Foot Ankle Surg. 2016;55(4):762-766.
3. Liu J, Ludwig T, Ebraheim NA. Effect of the blood HbA1c level of surgical treatment outcomes of diabetics with ankle fractures. Orthop Surg. 2013;5(3):2033-2038.
4. Wukich DK, Joseph A, Ryan M, Ramirez C, Irrgang JJ.Outcomes of ankle fractures in patients with complicated versus uncomplicated diabetes. Foot Ankle Int. 2011;32(2):120-130.
5. Shibuya N, Humphers JM, Fluhman BL, Jupiter DC. Factors associated with nonunion, delayed union, and malunion in foot and ankle surgery in diabetic patients. J Foot Ankle Surg. 2013;52(2):207–211.
6. Lanzetti RM, Lupariello D, Venditto T, et al. The role of diabetes mellitus and BMI in the surgical treatment of ankle fractures. Diabetes Metab Res Rev. 2018; 34(2).
7. Pincus D, Velijkovic A, Zochowski T, Mahomed N, Ogilivie-Harris D, Wasserstein D. Rate of and risk factors for intermediate-term reoperation after ankle fracture fixation: a population-based cohort study. J Orthop Trauma. 2017;31(10):e315–320.
8. Lovy AJ, Dowdell J, Keswani A, et al. Nonoperative versus operative treatment of displaced ankle fractures in diabetics. Foot Ankle Int. 2017;38(3):255-260.
9. Jones KB, Maiers-Yelden KA, Marsh JL, Zimmerman MB, Estin M, Saltzman CL. Ankle fractures in patients with diabetes mellitus. J Bone Joint Surg Br. 2005;87(4):489-495.
10. Holmes GB Jr, Hill N. Fractures and dislocations of the foot and ankle in diabetic associated with Charcot joint changes. Foot Ankle Int. 1994;15(4):182-185.
11. Bazarov I, Peace RA, Lagaay PM, Patel SB, Lyon LL, Schuberth JM. Early protected weightbearing after ankle fractures in patients with diabetes mellitus. J Foot Ankle Surg. 2017;56(1):30–33.
12. Lillmars SA, Meister BA. Acute trauma to the diabetic foot and ankle. Curr Opin Orthop. 2001;12:100–105.
13. Schon LC, Marks RM. The management of neuroarthropathic fracture- dislocations in the diabetic patient. Orthop Clin North Am. 1995;26:375-392.
14. Wukich DK, Kline AJ. The management of ankle fractures in patients with diabetes. J Bone Joint Surg Br. 2008;90:1570–1578.
15. Manway JM, Blazek CD, Burns PR. Special considerations in the management of diabetic ankle fractures. Curr Rev Musculoskelet Med. 2018;11(3):445–455.
16. Perry MD, Taranow WS, Car JBr. Salvage of failed neuropathic ankle fractures: use of large-fragment fibular plating and multiple syndesmotic screws. J Surg Orthop Adv. 2005;14(2):85-91.
17. Jani MM, Ricci WM, Borrelli J, Barrett SE, Johnson JE. A protocol for treatment of unstable ankle fractures using transarticular fixation in patients with diabetes mellitus and loss of protective sensibility. Foot Ankle Int. 2003;24(11):838-844.
18. King CM, Cobb M, Collman DR, Lagaay PM, Pollard, JD. Bicortical fixation of medial malleolar fractures: a review of 23 cases at risk for complicated bone healing. J Foot Ankle Surg. 2012;51:39-44.
