A Realistic Approach to Charcot Foot
Author: Robert Frykberg, DPM, MPH
Carl T. Hayden VA Medical Center
Well, I have a very daunting task and that is within 20 minutes go over a subject that I could spend three hours talking on, a subject that's been the favorite topic of my career over the last 30 years, that's a Charcot foot. Now, you're going to have a lot of talks on surgical management, so I'm going to really cut that short and focus on what I think is the most important part, which is the underlying pathophysiology. You need to understand this condition for you to effectively manage it.
A little bit of history - Charcot first described this entity in 1868 when he was practicing in La Salpetriere Paris, France which was called the grand asylum of human misery. And he termed this the arthropathy of locomotor ataxia at that time while ataxic patients due to tertiary lues with significant arthropathies, bone and joint disorders. He was also actually the first one in 1883 to describe this affectation in the foot, which he termed the pied tabetique. And subsequent to that time there have been many, many, many reports. But, it was not until 1936 that William Riley Jordan, a former Joslin Clinic fellow had described the first reported case in a patient with diabetes. So it was in early 1936, up to that point it had been tertiary lues for the most part. And since 1936, of course, the literature is replete with mainly case reports, case series on the Charcot foot.
Now, I showed this slide earlier, but I think it deserves some mention once again, recognizing here that of all the lower extremity complications of diabetes the Charcot foot imparts a five year mortality of 41% based on the recent Van Baal study. So patients with a Charcot foot have a lower life expectancy than do people with breast cancer, Hodgkin's disease, and prostate cancer. So this is a significant complication for these people. It does impart earlier mortality. So, again, it highlights the significance of this disorder and serves to remind that these are not robust healthy patients. These are very, very complex sick people who might otherwise look fairly robust in most cases.
So let us talk about the pathogenesis and the natural history. Again, it's critical that you understand the underlying pathophysiology if you were to effectively manage this condition.
Basically, we are well aware of the neurotraumatic versus the neurovascular theories. These are the classic German and French theories that were combined in a unified theory by Steve Edelman in 1987. He was a Joslin Clinic fellow at that time as well. On this side we have the neurotraumatic theory which really evolves around trauma applied to an insensate foot. We have the sensory and motor neuropathy, decreased proprioception, pain, vibration with intrinsic muscle atrophies, etc. We have applied trauma to this system, we developed spontaneous dislocations, neuropathic fractures, and osteoarthropathy. Now these are actually separate entities, but in osteoarthropathy for the class of Charcot foot we have everything combined.
On the other side we have the neurovascular or French theory, which is really the result of inflammation and the autonomic neuropathy, underlying autonomic neuropathy, which is very important. Why? Because this leads to sympathetic failure and increase in bone blood clot. So we have sympathetic failure, increased blood flow due to A-V shunting leading to osteolysis, osteopenia, bone atrophy. Again, apply trauma to this system and you get spontaneous fractures, dislocations, and Charcot foot, continued weightbearing due to the insensitivity leads to this so called vicious cycle that leads to further and further joint destruction that we now recognize as being the classic Charcot foot.
We also need to know something - I do not want to do belabor this - but over the last decade or so it's become abundantly clear the role of nuclear transcription factor kappa beta (NF-kβ) and underlying bone physiology and aberrations in that physiology that is attendant to any type of bone loss, any condition with bone loss going from osteoporosis to periodontal disease including Charcot arthropathy. So I just want to briefly mention this because I'd be remiss if I didn't, and we really do need to understand this. It took me three hours of reading one paper to finally get this straight.
Basically, we know that osteoclastogenesis resulting in osteolysis or osteopenia is dependent upon the activation of nuclear transcription factor kappa beta (NF-kβ). That's well established within the literature. NF-kβ activation is dependent upon the increased expression of RANKL, which is the receptor activator of nuclear factor kappa beta ligand. So RANKL is the ligand that binds with osteoplastic progenitor cell based on who have expressed RANK which is just the receptor on the preosteoclast cells. So when the RANKL binds with RANK receptor activator of nuclear factor kappa beta residing on the precursor cells, we have an up regulation of this NF-kβ. And the NF-kβ leads to up regulation of these osteoclasts and osteoclastic proliferation, osteoclastic activity and that results in the severe osteopenia and osteolysis. And that is about as simple as I can make it - osteoclastogenesis is due to RANKL binding with RANK which up regulates NF-kβ which promotes the preosteoclasts to develop into active osteoclasts.
Now, my friend William Jeffcoate in 2005 published a little bit different pathway than that the one I promoted earlier, but incorporates everything that we now know about the Charcot foot. This is your typical neurotraumatic theory where we have neuropathy due to diabetes in most cases or leprosy or alcoholic neuropathy, whereby we have fracture, abnormal loading, trauma, increased force, leads to dislocation, fracture, further abnormal loading, and you see this cycle develop. But now with what we know about the so called inflammatory cycle and inflammatory pathway, based on proinflammatory cytokines RANK and RANKL, we have an additional pathway that we need to merge with the more classical pathways.
So here we have the fracture/dislocation leading to an up-regulation of proinflammatory cytokines, TNF-α IL 1β, so this is the inflammation cycle. Now, the proinflammatory cytokines, especially TNF- α lead to the up-regulation or expression of this RANKL and the up-regulation of the NF-kβ, which we already mentioned, leading again to osteoclastogenesis, osteopenia, and further fracture/dislocation on that continuing cycle of destruction because of the insensitivity and the continued walking. So we need to realize how these vicious cycles really can occur leading to the severe bone and joint deformity that we see that are so characteristic of the Charcot foot.
Basically we are well aware of the Eichenholz staging system published by Sydney Eichenholz in his monograph in 1966 - primarily dealing with osteoarthropathy of tertiary lues, although he did include some cases of diabetic Charcot feet in that series. But in 1990 we know that Shibata had proposed in his series of patients dealing with ankle arthrodesis and patients with Charcot ankles due to leprosy, he proposed a stage 0, which is the prodromal stage where radiographs might be silent but there is bone stress injury and inflammation. So this is a radiographically silent precursor or Charcot in situ as I'll often say where there are no radiographic changes evident unless you look very carefully or obtain an MRI.
Then we have stage 1 or the acute stage, the stage of development, the classic Eichenholz acute stage where we see the gross edema of the fractures, the dislocation, and dissolution of the joints recurring. Once we establish a diagnosis and get that patient off their foot, then we can move into the healing stages or stage 2, which is coalescence, which is really marked by lessening of edema and absorptions of some fragments and may be a little cooling down, lessening of the inflammatory, but it's very difficult to relate say if this a coalescence stage or is it the final stage of reconstruction, which indicates complete lessening of edema, cooling off, and healing of these fracture fragments in whatever deformity was obtained at the high point or the acute stage.
So basically it is easier to consider the two, acute stage, the earlier stage. Once healing has started we move into the chronic stage and what we're left with, though, are the significant deformities. Now, just deformity in a plantigrade position is not bad, but we oftentimes see the severe deformities due to the rocker bottom effect, due to patients being misdiagnosed or diagnosed early in the course of this disease. So that inflammatory vicious cycle just continues on for way too long leading to the deformities at hand.
These are just some of the various images of different types of Charcot feet. Remember, generally all it takes is an x-ray. I rarely will get an MRI, especially if I see a foot like this because I know what I'm dealing with. However, when you have that stage 0 and you suspect unilateral swelling in a densely neuropathic patient with or without a history of injury, then you want to get the classic MRI because that will be diagnostic for that bone stress injury, so typical of the stage 0 foot.
This actually is the first Charcot foot that I ever saw back in my training, way back in the '70s before many of you were born, and this is what really caught my eye and developed my interest in the Charcot foot.
So, again, we see x-rays like this, we know what we're dealing with. We don't need to do anything further unless you want to do a CT with 3D reconstruction for possible surgical planning, and that's really when I will get the 3D CT.
However, when we do have that stage 0, when see radiographs are negative but we have a high degree of suspicion, which is really critical, then we'll get the MRI looking for this bone stress edema, and you can see the bone marrow edema right here, typical of that bone stress injury. I don't often have to do this, but on occasion I certainly will.
Now, how do we treat this problem? This is really a serious clinical conundrum for us all. Which patients do we operate on? Which patients do we not operate on? Is there any evidence to support operating on one patient versus another? And are those patients really suitable for operation? So, I think it's important again to always focus on the primary goal, which is in my mind to avoid surgery on these patients and to pay attention to the basics, as always.
And our primary goal is always to convert from an active to that quiescent stage, because we know when it's active there's further ongoing destruction, and when it is quiescent it is healing and we want to minimize deformity. So the quicker that we can convert from the active to the quiescent or chronic stage, the better off this patient will be. How do we do this? Well, I recommend nonweightbearing with bed rest, crutches or wheelchair. However, Pinzur in 2006 and de Souza in 2008 published papers in JBJS that indicated that people who were putting a total contact cast in the acute stage can ambulate with weightbearing and not have any detrimental effects. I tend to treat my patients non-weightbearing in a cast because I don't trust them. And if I tell them to non-weight bear they will bear weight but if I tell them that they can bear weight, they'll be walking all over creation, and I don't do that. So my preference is to treat them as non-weightbearing.
We also do immobilize patients, however, if they are in bed or in a wheelchair or very reliable, you can just use an Ace bandage of a Jones dressing. But we tend to rely on total contact cast or bi-valve cast, later on the cast walker or a CROW walker. But I think these latter two are really for more in the later stages after you've reduced the edema and have converted at least to a convalescence stage.
And then finally, we also have the adjunctive therapy, the pharmacological therapies with the bisphosphonates, calcitonin, etc. There are a number of studies that I don't have time to go into that indicate that these may be adjunctive. How? Because they interact in that RANKL, RANK, NF-kβ, osteoclastogenesis pathway. The bisphosphonates actually act later. They work on the activated osteoclasts while calcitonin will put a break on that whole system to prevent the osteoclasts from becoming activated in the first place by interfering with the RANKL, RANK system.
Also, bone stimulation, electrical stimulation, and ultrasonic stimulation with low-intensity pulse ultrasound have been reported. Very few good controlled trials to do this. Trials are present for bisphosphonates and calcitonin, but I view these as adjunctive to the basic tenants of care, which is the immobilization and the offloading, and there are plenty of references for you.
So, in the acute active stage surgery is generally not performed. There's very little evidence to support it, except in acute dislocations and these are primarily talonavicular with bone fragmentation, can be treated with early fusion. This goes back to 1979 in Newman's paper in British Medical Journal and Lesko's 1989 paper.
Simon was the first to publish any series with only 14 patients on arthrodesis of the TMT joint in patients with acute Charcot. So that's 14 patients in the established literature which have been operated on in the acute phase. So we need much more evidence for this before you can be assured that operating on these people in the acute stage is warranted, because these are sick people and obviously a surgery can create more problems and can reactivate a Charcot process as well. So you just need to be aware of this. We generally do not perform surgery on acute Charcot patients unless there's an acute dislocation that needs to be rapidly reduced.
Now, management of the chronic Charcot foot, however, really is predicated upon - do we have an ulcerated foot or a non-ulcerated foot. So in the non-ulcerated foot chronic Charcot we know that therapeutic footwear is the ultimate goal, and so we will use therapeutic footwear whenever possible. However, for some deformities in some unstable patients, they do require patellar tendon bearing braces, CROW walker or from Germany a Muenster brace and if these work, then we're fine, that's where we leave it and we follow our patients very carefully.
However, we do operate on these people with severe instability, especially for ankle joints, if there is no ulceration. If I can manage any deformity and keep them ulcer free, I'll just manage them conservatively because I think that's safer and, again, there are no studies that show any increased longevity or better lifestyle, surgical versus non-surgical.
However, for the ulcerated patient, as we could see here, it's a little different story. We try to get them back to the non-ulcerated state by applying our basics of care - offloading with TCC especially or cast walker or what have you. Once healed, if we can get them healed, then we'll consider surgery - exostectomy, realignment arthrodesis when necessary, especially for severe deformities, or debridement and treatment of osteomyelitis when they have a severe deformity probing to bone and there is concomitant osteomyelitis with the Charcot deformity. It's a very difficult clinical conundrum, so oftentimes we'll do a realignment arthrodesis or exostectomies and also treat for underlying osteomyelitis when it's proven to be present.
So, again, our goal is to get these patients from the ulceration state back to the non-ulcerated state and hopefully just be able to treat them with therapeutic footwear unless there's severe deformity or severe instability that does require surgery for long-term benefit.
So in the quiescent reparative or chronic stage then we know that surgery can be performed as necessary, and there is abundance of literature to support the role of surgery in selected patients with selected deformities or instability in this stage. And it's often a better alternative to amputation or sometimes amputation is a better alternative, i.e. Syme amputation might be better than a below-knee amputation for some people or vice-versa.
But, we have to remember that the complications can also be limb threatening. These people are at risk of losing their limb when they have the acute Charcot foot, they can lose their limb if we do operate on these feet, especially if they become infected, they can develop DVTs, pulmonary emboli. Remember these are sick people. Most of them are morbidly obese, they are very sick people. So you have to have a lot of preoperative counseling before you entertain surgery on these patients, but don't shy away from it if it's indeed necessary.
Again, we treat the operated foot as an active Charcot foot, so we tend to immobilize these people and keep them off their foot for quite a while, because we know the diabetic patients are going to take perhaps twice as long to heal or to fuse and these are the most complicated of our diabetic patients. So we tend to keep them non-weightbearing for as long as we can. But remember that this immobility can also result in other problems like DVTs and PEs, so most of our operated patients we're treating with anticoagulation routinely, because I don't want to deal with the pulmonary emboli, and we've seen that in several patients.
What are the common techniques? I'm really not going to get much involved in the discussion of surgery. I'll leave that up to Chuck who will be talking about this later. Common techniques that we often will see are exostectomy like plantar planing, very common, it's the most common type of surgery, midfoot or medial column arthrodesis with internal fixation, columnar screws, plates, Ilizarov circular frames, very, very commonly done, or spatial frames, rearfoot ankle arthrodesis using Ilizarov techniques, IM rods and screws, with or without tendo Achilles lengthening. I don't routinely do TALs unless the patient has equinus. I do not subscribe to the equinus causation of Charcot because it's never been proven. But when you have demonstrable equinus deformity, you must do a tendo Achilles lengthening, of course.
And as I said earlier, amputation sometimes needs to be considered, sometimes especially for the real sick patient, amputation like at the Syme level with prosthesis. You'll keep that patient down far less than trying to do a major fusion on these people that might take six months of recovery and immobility. And for severe instability, especially the ankle with intercurrent ulceration and osteomyelitis, I will sometimes advise perhaps a Syme or below-knee amputation. But there are many, many types, there's a lot more literature on all of these things. If I can get away with a simple procedure or a combination of procedures, I'll certain do that when possible.
There's a lot of new published studies going from 1990 to this year indicating that we can generally expect 80% to 100% limb salvage on these patients. Multiple surgeries are sometimes and often required. Sometimes we do multiple procedures at once or we stage the surgery or many times we have to revise these surgeries because non-unions are not uncommon.
Stated complications in all these papers - fibrous unions or non-unions, not always bad if you've maintained some degree of correction and stability. We can see recurrence at adjacent sites or have residual deformities that we are unable to correct or fail to correct. The worst problem will be infection, osteomyelitis, amputation or death during the observation period, and death is usually MI as a result of the surgery, I mean we are three- or four-hour surgeries or pulmonary embolus. So we do see these and we have to keep this in mind when we're considering our surgical intervention.
This is from Micky Pinzur's 2007 paper, a nice algorithm here that I try to use to say which patients do I need to operate on, which do I not. If we have a plantigrade foot with a collinear talar first metatarsal angle, we'll just treat that patient with a total contact cast, weightbearing and non-weightbearing, get them into a pneumatic walker brace and then custom shoes. If we have a non-plantigrade deformity where things are not collinear but we have a good host, good quality bone, then we can entertain surgery with internal fixation followed by total contact cast, walker brace, and then shoes. Always therapeutic shoes have to be done.
However, as we often see, we have a poor host, very sick host, poor quality bone, a lot of osteopenia, then we tend to go with percutaneous ring fixators, Ilizarov techniques with or without internal fixation followed by total contact cast for about four weeks if we get the frame off, diabetic pneumatic walker boot or cast walker and then shoes. This is a nice algorithm to follow because it gives you some idea of which patients may or may not need surgery, although we don't have a lot of good data, comparative data, to support that.
So where is the evidence on how we manage the Charcot foot? Always a critical question - where is the evidence? Well there's little-to-none. There are few randomized blinded control trials, those are mainly with the bisphosphonates and pharmacological treatments. Most of our data comes from case series, single case reports, cohort studies, both retrospective and prospective or non-randomized studies, and these aren't enough evidence to really support the routine use of any one agent or modality. And I say this because you need to know what is the evidence to support your treatment program, and you need that because you have to think of what happens if I have one of these complications, the patient loses a leg or their life due to my interventions. You want to base your treatment on some kind of evidence because it's the way to practice your medicine.
So we do need trials comparing conservative with surgical treatments. We need these so that we can ascertain who needs surgery, who doesn't need surgery and at which joint - is an ankle joint best treated by surgery or best treated by conservative care, and we need to know this and we need to know do patients have any operative intervention, have the higher quality of life, do they live longer than people who are not operated on or not. So we have a lot of questions to ask.
Remember that clinical outcomes are the most important outcomes - does the patient have a better quality of life, can they independently ambulate or are they are going to be restricted to a wheelchair or a motorized wheelchair and never get out of the chair? So many questions still needing answers at this point.
So, based on those questions I say what Charcot had said in his 1868 paper, "sera continue" - to be continued. We've learned a lot about Charcot foot, the pathophysiology and the management, but we have much, much more to learn about.
So, with that I want to thank you for your attention. Again, I hope I didn't go too far over. Again, a new textbook was published this last year, if anybody is interested on this subject.
Male: Just a couple questions, Bob. I guess the first one would be Symes or avoid amputation in some of these morbidly obese diabetics. Is that really worth it given the chance that a lot of them are morbidly obese and could define huge ulcerations under that very small surface area? What's been your experience and the literature on that?
Bob: Well, those are exactly the patients I might consider this on. We're talking 350-400 pounders. Do we want to put them into a frame for six months and keep them off their feet or do a Syme amputation and get them up in a prothesis within three months and walking? I say Syme amputation because with the Syme we can them in a prosthesis, they can walk without much physical therapies. It's far easier than something else. No I don't do that often, but sometimes, especially when you have osteomyelitis or severe midfoot ulcerations I'll talk about that.
Male: One other thing that I noticed in the classification scheme, I know you probably see this very commonly in your practice, is if you have that non-infected Charcot patient that has an open ulceration and you're planning on a surgical reconstruction, do you stage that, do you manage the ulcer first and try to get it healed and then go in and reconstruct it, or do you do it all in one stage?
Bob: I will try to get it healed but if they have Charcot cysts, synovial cysts under there, we're often not going to get it healed, so I'll try to make sure that it's not infected. In that situation I'll usually go to external fixation and when we surgically decompress the deformity, as you will know, surgically decompress that deformity and create an arch where there was a bulge before, those ulcers will heal very, very rapidly. So I won't shy away from that, but will always get bone cultures and pathology underneath that ulceration.
Male: Excellent. Thank you. We are lucky to have Dr. Frykberg here, he is one of the world experts on Charcot. Thank you so much.
Bob: Thank you.