BioMechanics Diabetes Supplement
August 1999

Charcot:
Early staging can improve functional outcomes

Today diabetes mellitus is the most common disease associated with Charcot joints. The differential diagnosis includes osteomyelitis, arthritis, and pes planus. Often the surgeon is asked to biopsy and culture the foot of a patient with diabetes, under the assumption that it represents osteomyelitis instead of Charcot neuroarthropathy. It is important, therefore, to be able to recognize this condition, stage it, and institute treatment accordingly. We undertook to describe such changes and to define a staging system that can be used clinically to intervene in a meaningful fashion to minimize deformity in the Charcot foot.

Our study population was derived from the patients who were referred to the foot and ankle clinic at our institution between 1990 and 1996.¹ Many patients were referred for ulcer care secondary to deformities. All patients had evidence of diabetic neuropathy by history and by examination, increased local skin temperature, decreased sensation to pin prick, decreased vibratory sense, or an abnormal Semmes-Weinstein monofilament examination.

Histories, physical examinations, radiographs, and charts were reviewed. Anteroposterior, lateral, and oblique radiographs were evaluated for specific changes. The joints involved were identified on initial presentation. Soft tissue swelling, localized osteopenia, cystic erosions, diastases, subluxations, dislocations, fractures, and fragmentation were recorded. The staging system was based on a series of progressive changes observed on successive radiographs taken from the time the diagnosis was first made, at our clinic or elsewhere. Radiographs were taken every two weeks initially if the process was acute (stages 1 and 2) before midfoot breakdown occurred. In the later stage (stage 3), radiographs were done every eight to 12 weeks until resolution of the process (stage 4), which usually took 24 months. All patients were staged at the time of presentation.

• Stage 0 was a clinical diagnosis made at the time of the first examination of a patient who presented with a warm, swollen foot and normal radiographs. Technetium-99m and gallium and indium scans had been obtained by the referring medical service and were used to differentiate between Charcot neuropathy and osteomyelitis.
• Stage 1 demonstrated radiographic cysts, erosions, localized osteopenia, and sometimes diastases between the first and second cuneiforms in addition to the clinical findings.
• Stage 2 was identified by joint subluxation.
• Stage 3 was marked by dislocations and arch collapse.
• Stage 4 represented the healed stage of the bony process.

Since the Charcot process may arrest at any stage with or without deformity, a foot that had stopped progression in the early stages and had evidence of sclerosis or fusion had become a stage 4, bypassing the destructive stage 3 phase.

The study

Results

Fourteen of the deformities began at the talonavicular joint (medial column). The talus crushed the navicular (Figure 2), then plantar flexed causing a plantar talar head prominence. The cuneiforms articulated with the dorsal aspect of the talus. The tarsometatarsal and the naviculocuneiform joints were involved in 37 feet. The metatarsals displaced dorsally on the cuneiforms. The cuneiforms fragmented and began to implode plantarly. A severe rocker-bottom foot developed as the integrity of the transverse tarsal and medial longitudinal arches was compromised.

Ten feet were seen at stage 0. All were swollen and warm. Most patients had an onset of new pain in a previously insensate foot. Several patients had observed the identical presentation many times without any resultant deformity. Tc-99m bone scans, when available, were markedly positive in the midfoot, indicating nonspecific increased bone flow and turnover. Indium and gallium scans were not indicative of infections, as they were negative or less active than the Tc-99m. We did not do scans in the absence of ulcerations, as the results may be misleading. Initial radiographs were normal except for soft tissue swelling. Following successive radiographs, three feet were noted to have stabilized at stage 1, two stabilized at stage 2, and three progressed to stage 3. Two feet were lost to follow-up.

Six feet were initially seen at stage 1. These patients had a warm and swollen foot. Radiographs showed erosions and cysts localized to the involved joints. The navicular and second cuneiform joints were most frequently involved. Diastasis was seen between the first and second cuneiforms and first and second metatarsals. Successive radiographs showed that one foot stabilized at stage 2, while three progressed to stage 3. Two were lost to follow-up.

Joint subluxations were the hallmark of stage 2. Sixteen feet were identified at this stage. Two stabilized without progression by fusing in situ. Eleven progressed to stage 3. Three were lost to follow-up.

Joint dislocations and arch collapse marked stage 3. Twelve feet were initially seen at this stage. Three open reductions and internal fixations were done early in the series. Two deformities recurred when the hardware became loose and was removed. Seven feet progressed to stage 4 with evidence of sclerosis and fusion on radiographs. Two feet required a transtibial amputation for chronic nonhealing ulcers with diffuse osteomyelitis and unstable joints.

Seven feet presented at stage 4. Severe collapse of the arches with sclerosis and fusion was seen on radiographs. The deformities remained stable over time. Six patients in this group sought medical attention because of chronic, nonhealing plantar ulcers. Three feet required transtibial amputation. One foot underwent surgery for a dislocated navicular and two for a dislocated first cuneiform. The medial column of the foot was then successfully fused.

Transcutaneous oxygen tensions (PO₂s) were obtained on all patients before surgery. When deemed necessary, a vascular consult was sought. Three patients required a vascular bypass before surgery.

Of the 16 feet seen at stages 0 and 1 (nondeformity stages), three stabilized at stage 1 and three stabilized at stage 2 and never developed significant deformity. Two feet that presented at stage 2 also stabilized at that stage and did not progress. Eight feet, therefore, that were diagnosed and treated early, healed without deformity.

Discussion

Cumulative microtrauma, when superimposed on the bony and ligamentous changes described above, leads to diastasis, subluxation, and dislocation of the unprotected, insensate joints. The pattern of early breakdown observed in this study followed the lines of force described by Harris.¹⁷ The patterns were crushing of the navicular by the talus, plantar presentation of the talar head, disruption of the navicular second cuneiform joint, diastasis, and disruption of the first and second cuneiforms with separation of the first and second metatarsals. The middle (second and third) and lateral (fourth and fifth) columns of the foot were similarly affected as the feet continued to deform. Sanders⁷ described five different patterns of Charcot neuroarthropathy based on site from forefoot to hindfoot. Brodsky¹⁸ described Charcot arthropathy as occurring in the midfoot (type 1) including the naviculocuneiform and tarsometatarsal joints; the hindfoot (type 2) comprising the talonavicular and subtalar joints; and the ankle (type 3). Schon’s¹⁹ classification describes four different patterns; i.e., Lisfranc, naviculocuneiform, perinavicular, and transverse tarsal.

We found that the Charcot changes in the talonavicular joint behaved similarly to those of the naviculocuneiform and cuneifom metatarsal joints. In this series the earliest Charcot changes were observed between the first and second cuneiforms and the bases of the first and second metatarsals; i.e., the medial column. They then spread to involve the middle and lateral columns. We noted in our diabetic population that involvement of the subtalar and ankle joints produced different deformities, with more instability and slower bone consolidation.

Eichenholtz in 1966 developed a staging system that described the natural progression of the neuropathic foot in tabes dorsalis.²⁰ His stage 1 is the stage of fragmentation and dislocation, which corresponds to our stage 3. His stage 2 is characterized by coalescence (our early stage 4). Stage 3 is the consolidated or healed stage. We thought that Eichenholtz staging missed an important phase of the Charcot process; i.e., the early stages when the disease is minimal to mild but still recognizable clinically and radiographically. Brodsky’s classification¹⁸ is an anatomical one. It localizes the process, but does not help us to diagnose it early. Our staging emphasizes early detection and treatment to prevent deformity.¹ In our series, eight patients (50%) who were diagnosed and treated at stages 0, 1, and 2 stabilized early without developing deformities.

Although the preferred treatment continues to be nonweight-bearing, compliance was a problem in our clinic patient population. Our patients freely admitted to frequent weight-bearing on the affected foot, wearing nonprescribed shoes, and forgetting their orthoses. Clohisy,¹⁵ when treating patients with juvenile diabetes and bilateral foot fractures, found that when the initially involved leg was protected, there was an earlier onset of Charcot neuroarthropathy in the other leg than when both legs were unprotected. Initially, our patients in stages 1 and 2 received a total contact cast^21-23 and were kept partially weight-bearing for six weeks. Such modalities as total contact casts and patellar tendon bearing walkers are being used with increasing frequency to assist in healing neuropathic ulcers while allowing the patients to remain ambulatory.^21,24 This methodology could be transferred to the treatment of Charcot neuroarthropathy. We used the Charcot Restraint Orthotic Walker (CROW).²⁷ The CROW stabilizes the foot, prevents progression, allows patients to remain ambulatory, and is well tolerated.²⁵ Other, less expensive, diabetic-type walkers are now commercially available.

Three patients underwent surgery at stage 3. One fused and two progressed to greater deformity. We have since learned that reconstructive surgery during the destructive phase of this process (stage 3), while the disease is still evolving, may lead to failure because of bone breakdown around the fixation device. Others have encountered a high rate of complications with surgery.^8,23 Schon¹⁹ has successfully advocated a more surgically aggressive approach. Although we did not intervene surgically at stage 2, this seems an ideal time to operate on selected patients, before gross deformity develops and while the deformities are still reducible. When the Charcot arthropathy has become chronic and the process is no longer progressing, exostectomies can be performed to decompress nonhealing ulcers. We performed preoperative transcutaneous oxygen tension (TcPO₂) on patients undergoing surgery. If the TcPO₂ was marginal, an arteriogram was ordered to see if revascularization was possible. Three patients required vascular bypass before surgery on their feet. Although unusual in Charcot, peripheral arterial obstructive disease can occur simultaneously in diabetics. We had no recurrence of Charcot in the midfoot after healing, but one patient suffered a new onset of Charcot in the ipsilateral ankle.

We have determined that Charcot deformity can be seen at five defined stages, based on clinical examination and plain radiographs (Table 1). Each stage has been combined with certain treatment modalities designed to allow the patient limited ambulation during the one to two years required to heal. Early diagnosis is important, since 50% of our patients diagnosed and treated early did not develop deformity. We believe that the staging of the Charcot process at presentation has some predictive value and helps formulate a plan of treatment. Close follow-up, preferably in a center devoted to foot care, is recommended. Although complete nonweight-bearing, at least through the acute stages, is desirable, the level of compliance in our clinic population was such that we had to adopt a more permissive attitude. Nevertheless, our results were satisfactory. Our patients found that the CROW was a good and acceptable compromise.

Enzo J. Sella, MD, is associate clinical professor in the department of orthopedics and rehabilitation at Yale University and a member of the Connecticut Orthopedic Specialists, in Hamden. Carol Barrette, MD, is in the department of surgery at Fallon Clinic in Worcester, MA.

References

1. Sella EJ, Barette C. Staging of Charcot neuroarthropathy along the medial column of the foot in the diabetic patient. J Foot Ankle Surg 1999;38(1):34-40.
2. Birke JA, Sims DS. Plantar sensory threshold in the ulcerated foot. Lepr Rev 1986;57(9):261-267.
3. Laughlin RT, Calhoun JH, Maden JT. The diabetic foot. J Am Acad Orthop Surg 1995;3(4):218-225.
4. Armstrong DG, Todd WF, Lavery LA, et al. The natural history of acute Charcot arthropathy in the diabetic foot specialty clinic. Diabet Med 1997;24:357.
5. Kozak GP, Hoar CS. Management of diabetic foot problems. Philadelphia: W.B. Saunders 1984:103-112.
6. Parkhouse N, LeQuesne PM. Impaired neurogenic vascular response in patients with diabetes and neuropathic foot lesions. N Engl J Med 1988;318(20):1306-1309.
7. Sanders LJ, Frykberg RG. Charcot joint. In: Levin ME, O’Neal LW, Bowker JH, eds. The diabetic foot, 5th ed. St. Louis: Mosby Year Book, 1993.
8. Edmonds ME, Roberts VC, Watkins PJ. Blood flow in the diabetic neuropathic foot. Diabetologia 1982;22(1):9-15.
9. Bevan RD, Tsuru H. Long-term denervation of vascular smooth muscle causes not only functional but also structural changes. Blood Vessels 1979;16(2):109-112.
10. Boulton AJM, Scarpello JHB. Venous oxygenation in the diabetic neuropathic foot. Evidence of arteriovenous shunting. Diabetologia 1980;22(1):6-8.
11. Watkins PJ. Diabetic autonomic neuropathy. N Engl J Med 1990;322(15):1078-1079.
12. Watkins PJ, Edmonds ME. Sympathetic nerve failure in diabetes. Diabetologia 1983;25(2):73-77.
13. Delbridge L. The etiology of diabetic neuropathic ulcerations of the foot. Br J Surg 1985;72(1):1-6.
14. Ward JD. The diabetic leg. Diabetologia 1982;22(3):141-147.
15. Clohisy DB, Thompson RC. Fractures associated with neuropathic arthropathy in adults who have juvenile onset diabetes. J Bone Joint Surg 1988;70A(8):1192-1200.
16. Pinzur MS, Sage R, Stuck R, et al. Transcutaneous oxygen as predictor of wound healing in amputations of the foot and ankle. Foot Ankle 1992;13:271-272.
17. Harris JR, Brand PW. Patterns of disintegration of the tarsus in the anesthetic foot. J Bone Joint Surg 1966;48B(1):4-16.
18. Brodsky JW. Outpatient diagnosis and care of diabetic feet. Instructional Course Lectures, AAOS. 1993;42:121-139.
19. Schon LC, Easley ME, Weinfield SB. Charcot neuroarthropathy of the foot and ankle. Clin Orthop 1998;349(April):116-131.
20. Eichenholtz SN. Charcot joints. Springfield, IL: Charles C. Thomas, 1966;3-10.
21. Helm PA, Walker SC. Total contact casting in diabetic patients with neuropathic foot ulcerations. Arch Phys Med Rehab 1984;65(1):691-693.
22. Mueller MJ, Diamond JE, et al. Total contact casting in the treatment of diabetic plantar ulcers. Diabetes Care 1989;12(6):384-388.
23. Papa J, Myerson M, Girard P. Salvage with arthrodesis, in intractable diabetic neuropathic arthropathy of the foot and ankle. J Bone Joint Surg 1993;75A(7):1056-1066.
24. Gristina AG , Thompson WAL, et al. Treatment of neuropathic conditions of the foot and ankle with a patellar-tendon-bearing brace. Arch Phys Med Rehab 1973;54(December):562-565.
25. Morgan JM, Biehl WB III, Wagner FW Jr. Management of neuropathic arthropathy with the Charcot Restraint Orthotic Walker. Clin Orthop 1973;296:58-63.