Long-Term Mortality After Charcot Reconstruction Free

This study included patients who had undergone Charcot reconstruction by a single board-certified surgeon (P.B.) at a single institution between 2007 and 2013. Patients were included if they had been seen and evaluated within the given time frame, were ≥18 years of age, had a diagnosis of Charcot neuroarthropathy, and underwent an operation for limb salvage or reconstruction. Limb salvage and reconstructive procedures included staged and primary arthrodesis, wedge midfoot or hindfoot osteotomies, and midfoot beaming and hindfoot intramedullary nailing. Patients were excluded if they did not receive management during the given time frame or did not have at least 3 months of follow-up post-procedure.

Patient data were recorded, de-identified, and documented on a Microsoft Excel spreadsheet. Patient characteristics were obtained from records of the initial operation and work-up and included age, sex, smoking status, and presence of end-stage renal disease (ESRD), peripheral vascular disease, diabetes, ulceration, and osteomyelitis.

The primary outcome was 10-year mortality rate. Additional mortality rates were assessed for 30 days, 60 days, 90 days, 1 year, and 5 years post-procedure. Secondary outcomes included fixation used, number of operations performed within the lifetime, progression to below-knee amputation (BKA), and relationship of BKA to time of death.

These data were sent to an independent statistician at the University of Pittsburgh for analysis. Patients’ mean age at initial operation was calculated, and a Wilcoxon rank sum test was performed for both 5- and 10-year mortality. χ² or Fisher exact tests were performed for all additional variables, with statistical significance defined as P = 0.05. Finally, a Kaplan-Meier curve was generated to assess survival probability after Charcot reconstruction. All statistical analyses were performed using SAS, v. 9.4, statistical software. Covariates were assessed to find correlations between associations and survival using a Cox proportional hazards model.

Our cohort included 33 patients, of whom 23 were male. The mean age of patients at the time of their procedure was 56.8 years (range 29–76 years).

No deaths were identified at the 30-, 60-, or 90-day marks. One death occurred within 1 year of the procedure, 11 deaths occurred after 1–5 years, and five deaths occurred after 5–10 years. The overall 10-year mortality rate was 52% (17 of 33). We were unable to show a significant difference in mortality based on patients’ age at initial operation; however, there appeared to be more deaths in older patients (Table 1). Similarly, there did not appear to be any significant variable related to mortality risk when reviewing smoking status, presence of an initial wound, or diagnosis of vascular disease or diabetes (Table 2).

The median survival time after surgery was 7 years, with survival probability of 35% (Figure 1). Mortality after surgery did not appear to be associated with any additional medical parameters assessed. Of the 17 total deaths, 10 had unknown causes, three were caused by acute myocardial infarction, two involved septic shock and multiorgan failure, one was the result of an abdominal wall infarction with bladder cancer, and one was the result of chronic lymphocytic leukemia.

Thirty-one surgical cases used internal fixation, with 19 crossing the ankle joint. Fifteen involved the isolated hindfoot, eight included the isolated midfoot, and eight included the mid- and hindfoot. Of the 31 cases, 19 were staged procedures with a mean of three documented surgeries over the lifetime, ranging from one to 11 per individual.

Fourteen patients required revisional, unplanned operations. Of those who underwent revisional surgery, two (14%) required it bilaterally. Of these 14 patients, two (14%) went on to have a BKA. Six patients died within 10 years, of whom three died within 5 years. Two of the deceased were those who had gone on to have a BKA, including one who died within 5 years and one who died within 10 years.

Six patients (18%) went on to have a major amputation. Fixation attempts for these patients included one external fixation alone, one midfoot internal fixation, and four internal fixations through the ankle/rearfoot. The time from initial operation to BKA for these six patients ranged from 7 months to 10 years, with a mean time of 4 years. Five of the six are now deceased, with three deaths occurring within 5 years and the other two within 10 years of the initial operation. The time from BKA to death ranged from 1 week to 3 years, with a mean of 1.7 years.

The sequelae of Charcot neuroarthropathy pose medical and surgical challenges. Gazis et al. (11) in 2004 reported on the mortality of patients with diabetic Charcot neuropathic osteoarthropathy managed by a single specialist and found a mortality rate of 44.7% and a mean time to death of <4 years. Additionally, they found that 23.4% of Charcot patients went on to have a major amputation. Most of the studies like this one have had a follow-up time of ∼5 years. We were able to demonstrate a substantially lower amputation rate for those undergoing Charcot reconstruction over a longer follow-up time (Figure 1).

We did not find significant differences in mortality after reconstruction based on the presence of a wound, osteomyelitis, or smoking (Table 2). Sohn et al. (7,8) demonstrated that the presence of both diabetic foot ulceration and Charcot neuroarthropathy significantly increase the risks of mortality and amputation compared with diabetes alone. Our data did not reveal any association of mortality risk with prior wound; however, the lack of this finding may have been the result of our small study size. We do believe that deformity correction is crucial to provide lower-extremity stability and prevent wounds.

Yammine et al. (9) found the greatest mortality rate of 44% at 1 year, followed by 24.5% and 16% at 5 years and ≥7 years, respectively. Our study paralleled some of these findings, suggesting that the greatest risk of death falls within 5 years after the reconstruction procedure.

Although our study looked specifically at diabetes, osteomyelitis, vascular disease, and renal disease (Table 1), this patient population has numerous other possible comorbidities. Rosen et al. (12) found increased risk factors for mortality after lower-limb amputation to include dementia, nonambulatory status preoperatively, and heart failure. These findings both underscore the importance of maintaining limb salvage and ambulatory status when assessing the benefit of Charcot reconstruction and mortality versus BKA and point to other confounding factors that should be examined further when assessing mortality after reconstruction.

Certainly, these complex cases often involve complications and the need for subsequent hospitalizations. Even with reconstruction, Ha et al. (4) reported complications rates with Charcot foot reconstruction as high of 36%. Nonetheless, prevention of major amputation sequelae may still prove to be valuable and cost-effective. Driver et al. (13) reported that the United States generated $116 billion in direct costs, one-third of which were linked to diabetic foot ulceration. Our study showed an overall limb salvage rate of 86%. The ability to prevent ulceration, infection, and amputation in people with diabetes may prove valuable in reducing the health care burden of the disease.

The strengths of this study include the continuity of having a single surgeon provider and one care location. Patients were able to be seen perioperatively with continuity of care and management over a long period of time at a single center.

This study does have several limitations, including its retrospective design, inclusion of four bilateral cases, and small population size. Although our population size is smaller than that of other studies, we found it necessary to limit our population to focus on an exact 10-year review. Additionally, we included all operations that fell under Charcot reconstruction, which varied from patient to patient. Certainly, there may be differences in management between a more localized Charcot joint procedure versus that of a multiarticular Charcot procedure. Additionally, there are inherent limitations in examining only one surgeon’s experience, which is likely to grow and change with experience over a longer duration. However, it must also be noted that each Charcot case has its own unique features, in terms of deformity; presence of wound, osteomyelitis, and complex comorbidities; and social factors, and these variable features continue to pose different challenges even for an experienced surgeon. Additionally, operative techniques and hardware have evolved through the years from plate and screw to intramedullary nailing.

We retrospectively reviewed the cases of 33 patients who had undergone a Charcot reconstruction by a single physician over a 10-year follow-up period. To the authors’ knowledge, this is the first study of its kind to record 10-year mortality rates after Charcot reconstruction and limb salvage. This study found an overall mortality rate of 52% (17 of 33). There additionally appeared to be a peak in the rate of death in the 1- to 5-year post-surgery time frame, with 33% of deaths occurring during this time window. Additionally, the study found rates of 18% (6 of 33) for subsequent major amputation and 81% (27 or 33) for successful limb salvage. Inevitably, patient mortality was not related to lower-extremity amputation.

We did not find significant differences in outcomes based on the presence of diabetes, ESRD, or peripheral vascular disease, which have been documented previously; however, we believe our findings may have been the result, in part, of our limited sample size. We additionally saw greater mortality in our older cohort immediately after initial surgery. Although this difference did not approach statistical significance, it may suggest that earlier intervention in younger patients may be preferred.

Given expected future increases in the number of people with diabetes in the United States, physicians should be prepared to see more cases of Charcot neuroarthropathy in their practices. As these surgeries are performed more often and operative techniques and material science continue to evolve, we hope this study can serve as an example of the potential long-term success of performing these surgeries. More long-term and prospective studies are needed to appropriately assess the outcomes of Charcot deformity correction, limb salvage, and mortality.

The authors thank Clair Smith, MS, of the Department of Orthopedic Surgery, University of Pittsburgh School of Health and Rehabilitation Science, for assistance with the data analysis.

No potential conflicts of interest relevant to this article were reported.

C.M. and A.M. researched data, contributed to the discussion, and wrote, reviewed, and edited the manuscript. P.B. reviewed and edited the manuscript. All authors approved the final version of the manuscript for submission. C.M. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis.