Comparative analysis of uniplanar external fixator and retrograde intramedullary nailing for ankle arthrodesis in diabetic Charcot's neuroarthropathy

Abstract

Background:

Charcot's neuroarthropathy of ankle leads to instability, destruction of the joint with significant morbidity that may require an amputation. Aim of surgical treatment is to achieve painless stable plantigrade foot through arthrodesis. Achieving surgical arthrodesis in Charcot's neuroarthropathy has a high failure rate. This is a retrospective nonrandomized comparative study assessing the outcomes of tibio-talar arthrodesis for Charcot's neuroarthropathy treated by uniplanar external fixation assisted by external immobilization or retrograde intramedullary interlocked nailing.

Materials and Methods:

Records of the authors′ institution were reviewed to identify those patients who had undergone ankle fusion for diabetic neuroarthropathy from January 1998 to December 2008. A total of11 patients (six males and five females) with a mean age of 56 year and diabetes of a mean duration of 15.4 years with ankle tibio-talar arthrodesis using retrograde nailing or external fixator for Charcot's neuroarthropathy were enrolled for the analysis. Neuropathy was clinically diagnosed, documented and substantiated using the monofilament test. All procedures were performed in Eichenholz stage II/III.Six patients were treated with uniplanar external fixator, while the remaining five underwent retrograde intramedullary interlocking nail. The outcomes were measured for union radiologically, development of complications and clinical follow-up, according to digital archiving systems and old case notes.

Results:

All five (100%) patients treated by intramedullary nailing achieved radiological union on an average follow-up of 16 weeks. The external fixation group had significantly higher rate of complications with one amputation, four non unions (66.7%) and a delayed union which went on to full osseous union.

Conclusion:

The retrograde intramedullary nailing for tibio-talar arthrodesis in Charcot's neuroarthropathy yielded significantly better outcomes as compared to the use of uniplanar external fixator.

INTRODUCTION

Effects of neurological affliction of joints was first described by Charcot in 1868.¹ Diabetes mellitus is currently the most common cause of neuroarthropathy.² Abnormal nociception and proprioception leadto progressive destruction of the joint. Management of Charcot's arthropathy by definitive surgical treatment remains a challenge despite technical advances in treatment modalities.

In the diabetic population, ankle neuroarthropathy constitutes 3% to 10% of all cases. This leads to significant instability, fracture and collapse of ankle mortise and carries a high risk of amputation of the lower limb.^3–7

There has been change in the philosophy of surgical management of neuroarthropathy, from amputation to limb salvage. Sir John Charnley used uniplanar external fixator with compression clamps for ankle arthrodesis with variable success. Recent studies have shown good results with deformity correction and arthrodesis with internal fixation.⁸,⁹ The principal goal of surgical intervention is to achieve a stable plantigrade foot without any deformity and to prevent re-ulceration.¹⁰ There are numerous methods proposed to obtain a stable arthrodesis of the ankle joint.^11–13 External fixation or internal fixation have been used,however, there are still controversies over the most successful and reproducible method. This is a retrospective nonrandomized comparative study assessing the outcomes of tibio-talar arthrodesis for Charcot's neuroarthropathy treated by uniplanar external fixation assisted by external immobilization or retrograde intramedullary interlock nailing.

MATERIALS AND METHODS

The records of the author's institution were reviewed to identify those patients who had tibio-talar fusion for diabetic neuroarthropathy, from January 1998 to December 2008 [Table 1]. Hospital data and radiological images were used to follow all patients to clinical, radiographic union and to measure clinical results.

Table 1.

Clinical details of the patients

A total of 11 patients (six males and five females) with a mean age of 56 years with tibio-talar arthrodesis for Charcot's neuropathy were enrolledfor this analysis. All patients were assessed preoperatively for circulation withdorsalispedis and posterior tibial artery pulse palpation and saturation monitoring with pulse oximetry. All patients were found to have satisfactory distal circulation.Preoperatively, clinical examination and monofilament test were used to diagnose and substantiate neuroarthropathy (inability to identify three out of 10 points). We used a 5.07 Semmes-Weinstein monofilament test.¹⁴,¹⁵ All the patients were confirmed to have neuropathy by the above test.

The Eichenholz Radiographic Classification¹⁶ for Charcot's neuroarthropathy was used to determine the stage of the arthropathy. This classification is as follows:

Stage I: Hyper-vascular (acute) joint laxity, subluxation, osteochondral fragmentation, debris formation

Stage II: Coalescent (subacute) absorption of the debris, fusion of the larger fragments of bone, sclerotic bone margins

Stage III: Loss of sclerosis, further fusion of bone segments

Preoperatively, patients were screened radiologically and hematologically to rule out infection. Their blood sugar levels were monitored along with HbA1c values to ensure there was proper control of diabetes. Surgical procedure was carried out in the quiescent phase of the disease in stage II/III. Simon et al¹⁷ advocated arthrodesis in stage one of disease process. None of our patients were operated in stage I because they were relatively asymptomatic or they refused surgery at that point of treatment. Four patients with plantar ulcers were treated with offloading total contact cast preoperatively, and the ulcers were allowed to heal. The type of surgical procedure a patient underwent, whether retrograde nailing or uniplanarexternal fixator, was senior author's choice irrespective of the stage or condition of the bone.

M/DN (metaphyseal/diaphysis nailing) supracondylar nail (Zimmer) was used for retrograde nailing and Charnley's compression uniplanar fixator assisted with posterior slab was used in external fixation cases. All of the intramedullary nails used in this study had locking bolts in the calcaneum in a mediolateral direction. There are devices available which allow the surgeon to lock the calcaneal bolts anteroposteriorly or obliquely, allowing for better bone purchase and stability.

Exclusion criterion for patients treated with external fixator was ulceration over potential external fixator pin sites. Exclusion criteria for patients treated with retrograde nail include normal subtalar joint; significant tibial deformity with malunion, greater than 10 degrees in any plane; marked loss of calcaneal body height; and active infections of foot and ankle.

Six (54.54%) patients underwent open reduction, debridement, synovectomy, compression of cancellous tibio-talar bony surfaces and stabilization with an external fixator with or without retrograde calcaneotibial Steinmann pin. All the patients were bonegrafted with cortico-cancellous bone from iliac crest. They were further immobilized in posterior below-knee slab support in the postoperative period for twelve weeks.

Five (45.46%) patients underwent open reduction, debridement, synovectomy, compression of cancellous tibio-talar bony surfaces and internal fixation with an intramedullary nail for tibio-talar joint arthrodesis. All the patients were bonegrafted with cortico-cancellous bone from iliac crest.

Postoperatively, patients were analyzed with respect to wound healing, stability of construct, pin loosening, pin tract infection, screw loosening, backout and wound condition. The monitoring of blood sugar levels was done postoperatively to ensure good control. HbA1c levels were taken every three months during follow up to ensure good control of diabetes. All ankles were immobilized in a below-knee slab or bivalve total-contact cast with Velcro straps for 12 weeks. Patients were allowed to ambulaten on weight bearing with crutches. External fixation removal was done at an average of 12.3 weeks, and patients were immobilized in a below-knee cast. Every six weeks, plain radiographs were taken to evaluate consolidation of arthrodesis. Those who failed to show consolidation were further protected with a below-knee cast for a period of six to 12 weeks, and repeat X-rays were taken. Those who showed consolidation were allowed custom made partial weight-bearing shoes with special protection till they showed full bony union. Those with radiological union were allowed to walk with a brace. Delayed union was defined as failure to obtain union by 30 weeks but eventual achievement of union by 40 weeks.¹⁸ Non-union was defined as failure to obtain arthrodesis by 40 weeks post-surgery clinico-radiologically or by reoperation.¹⁸ Bony arthrodesis was defined as bony trabeculae traversing between two compressed bony surfaces.

The average duration of follow up was 3.2 years (range, 1-10 years). Grading of clinical results was determined according to scale of Johnson and Bosekar¹⁹ which is Excellent: Solid arthrodesis and no limp, Good: Successful fusion with mild limp, Fair: Constant limp with poor cosmetic appearance, Poor: Non-union or amputation.

RESULTS

Results were analyzed with respect to radiological union status, complications and clinical grading scale of Johnson and Bosekar.¹⁸

Amongst the internally fixed group, all the five (100%) patients who had intramedullary nailing progressed to clinico-radiological union in an average period of 16 weeks (range, 14-18 weeks). There was a patient with wound breakdown and superficial infection eachthat resolved with conservative treatment. Two patients had a back-out of distal interlocking bolt that was removed when there was bony union.

Amongst the external fixator group(n=6), the complication rate was significantly higher. Four (66.7%) patients progressed to non union; two of them were stable fibrous non-unions allowing the patient to walk with a brace. There was one (16.7%) patient with delayed union, one (16.7%) patient who required below-knee amputation due to fulminating infection. Four patients (66.7%) had pin tract loosening;three (50%) patients had pin tract infection; and a revision of external fixation was required in four (66.7%) patients, which included revision of loose pins and not the complete fixator. Operative site wound infection occurred in three (50%) patients, and two (33.3%) patients had wound breakdown. Thus the complication rates were higher in the external fixation group when compared with those in the intramedullary nailing group.

Grading according to the scale of Johnson and Bosekar¹⁹ was as follows for all the groups. For the intramedullary nailing group, all five (100%) patients had excellent results [Figure 1]. In the group arthrodesed using external fixation, five (83.3%) patients achieved poor results [Figure 2]. Only one (16.7%) patient achieved a good result, but this case was also complicated by delayed union.

Figure 1A.

(a) Preoperative anteroposterior X-ray of leg with ankle joint showing tibio-talar varus deformity. (b) Preoperative lateral X-ray of same patient showig with tibo talar joint space collapse

Figure 2A.

(a) Preoperative X-ray anteroposterior and lateral view showing tibio talar destruction. Anteroposterior (b) and lateral (c) view of same patient showing external fixation in situ

Figure 1B.

Immediate post operative X-rays anetroposterior and lateral views of same patient (a) showing fixation with interlocked intramedullary nailing. (b) 8 weeks postoperative X-rays showing consolidation. (c) 16 weeks postoperative X-rays showing complete union

Figure 2B.

X-ray anteroposterior and lateral views of leg bones with ankle of external fixator group showing non union

When the results are analyzed together for bony union rate, all the five patients that were arthrodesed with intramedullary nailing had solid arthrodesis, making the fusion rate 100%. In the external fixator group, bony arthrodesis rate was 16.7% as only one out of six patients went on to achieve bony fusion.

DISCUSSION

Ankle arthrodesis for Charcot's neuroarthropathy represents the only surgical approach that can obtain plantigrade stabilization of the ankle joint allowing patients to bear weight. Neuroarthropathy has been staged by Eichenholz.¹⁶,¹⁹ Arthrodesis was believed to be the most difficult to achieve at stage III because of inadequate fixation purchase in poor bone stock. However, there is a difference between various viewpointsas to which stage of neuroarthropathy is most suitable for arthrodesis.²⁰ Simon et al.¹⁷ advocated arthrodesis in stage I, with excellent results. All our patients were at stage II/III for the reasons mentioned earlier.

The higher risk of nonunion of arthrodesis after fixation has been well documented²¹ and is due to poor bone quality and difficulty in obtaining stable compression at the arthrodesis site. Our results of external fixation have been less than optimal. Out of six cases, four (66.6%) progressed to non-union. Only one (16.7%) united at 32 weeks (i.e., delayed union) and had a good result. One (16.7%) required a below-knee amputation for fulminating infection.

Failure to achieve fusion has been reported to range from 0% to 67%.⁹,²² Papa et al.⁹ reported a rate of non-fusion of 34% in patients with diabetic neuropathic bone disease with a protocol that kept the patients non weight bearing for two months. Shibata et al.¹⁹ found a rate of non-fusion of 27% in ankles with neuropathic bone disease secondary to leprosy. Lynch et al.²³ in 1998 noted that the Charnley compression arthrodesis was associated with the highest complications.The present study supports this conclusion that the external fixator is associated with higher complication rates.Thesynovial pseudo-arthrosis with hypermobile joint, gross osteopenia, gross bone resorption or loss, severe deformity, poor control of diabetes, immune-compromised statusare associated with poor result.

Transfixation pins in the distal tibia and talus are poorly tolerated because of thin soft tissue envelope and poor bone quality. This leads to a high rate of pin tract infection which in severe cases can lead to further operative intervention. Also, because of the biomechanics of the frame, uniform compression is not created and this leads to increased rate of delayed union and of non-union.

There is a disagreement amongst various reports in the literature as to which is the most successful method to achieve tibio-talar fusion. Several studies have proposed methods to achieve arthrodesis through internal fixation without disadvantages of external fixation.^24–28 However, few studies comparing two methods used within a single institution, as in this study, have been carried out.

The methods of internal fixation achieve a higher rate of fusion that is combined with the advantages of decreased periods of immobilization and non weight bearing. The internal fixation system provides stabilization and compression across the arthrodesis while avoiding the problems of external pin tracts and prolonged periods of immobilization.

The results of retrograde intramedullary nailing in our series were clearly superior to those of external fixator, with all five (100%) patients achieving bony union within an average duration of 16weeks; and out of these, four (80%) patients had excellent results and 1 (20%) had good results.

The high rate of a stable union obtained in our series compares favorably to data from previous studies^29–33 that also used an intramedullary device to obtain a stable ankle joint. One of the reasons for poor outcomes among external fixator group was the use of uniplanar Charnley external/AO type fixation with its limitations for providing a stable construct. Ilizarov ring fixator produces a more stable construct, leading to a lower complication rate and higher union rate.³⁴ But these external fixators are cumbersome and not ‘patient friendly.’

The limitation ofthis study is its retrospective/ nonrandomized nature and small sample size in each group. The small sample size is due to lower prevalence of diabetic neuroarthropathy.

CONCLUSION

The retrograde intramedullary nailing for tibio-talar arthrodesis in Charcot's neuroarthropathy yielded significantly better outcomes as compared to the use of uniplanar external fixator. The external fixator has higher morbidity and complication rates with low tibio-talar arthrodesis success.

A high-powered randomized control trial is required for further substantiation of the results of this study.