The Wisconsin Foot & Ankle Institute is dedicated to providing world-class patient care. We specialize in disorders of the lower extremity, ranging from the simple problems, to the most complex limb salvage reconstructions.


The Chicago Lower
Extremity Surgical Symposium

Sept. 19-22, 2011
www.clesf.org »

The Journal
of Lower
Extremity
Surgery
www.jles.org »

Research : Wisonsin Foot & Ankle Institute
Wound Healing Complications and Infection Following Surgery for Rheumatoid Arthritis

Tags: , , , , , , , ,

Authors: Christopher Bibbo, DO, DPM, FACS, FAAOS, FACFAS

The patient who has rheumatoid arthritis (RA) often presents to foot and ankle specialists for the management of various musculoskeletal pathologies, often with crippling deformities of the forefoot, hindfoot, and ankle. Advances in the medical management of RA have allowed many patients to achieve previously unattainable levels of activity into later decades of life. This consequence from the improved medical management of RA is that although joint pain is being controlled from an inflammatory standpoint, musculoskeletal mechanical dysfunction that has already been incurred from years of low-grade, controlled disease continues unchecked. A perfect example of this is in the forefoot, in which multiple intrinsic and extrinsic musculoskeletal units are required for precise mechanical balance of the forefoot; despite good pain control and radiographic improvements in maintenance of bone mineral density, once biomechanical imbalances occur it is difficult, often impossible, to regain without ablative arthroplasties. Fortunately these surgical techniques have been shown to stand the test of time and to provide a platform for patients to regain functional ambulation. The foot and ankle surgeon, however, may be consulted to manage patients who have significant anatomic and biomechanical deformities, often in elderly, frail patients who depend on a host of disease-modifying medications. Additionally, even today there are still many areas of the nation that are underrepresented by rheumatologists, or patients are unable to access rheumatology services. In all these instances, patients may develop severe deformities that impose more risk than usual for perioperative complications. The goal of the foot and ankle surgeon is to achieve balanced correction of deformity while minimizing the risk for complications. To achieve these goals, a large number of procedures may be required at a single or in a staged operative setting. In this article the evaluation of the patient who has RA to assess operative risks and the management of perioperative complications in the RA foot and ankle patient are presented.

Biologic considerations

Wound healing is multifactorial

Effect of age and gender
The aging process has been demonstrated clinically and experimentally to impair the healing process, with advancing age and male hormone status having a greater negative influence. Men and postmenopausal women demonstrate impaired collagen deposition and wound healing after surgical incisions [1,2]. Age-related elevated levels of thrombospondin 2, an inhibitor of angiogenesis, have also been experimentally linked to delayed wound healing [3]. The negative effect of age on wound healing is also attributable in part to a decrease in stem cell reserve and function associated with the aging process [4].

The preoperative evaluation of the patient who has rheumatoid arthritis

Cervical spine evaluation
Preoperative medical clearance is a necessity for patients who have RA undergoing major reconstructive foot and ankle surgery or in any patient undergoing general anesthesia, regardless of the extent of the intended operation. As part of the preoperative health screening process, all patients who have RA undergoing general anesthesia should be evaluated for cervical spine disease. An upper extremity and neurologic examination and screening C-spine radiograph (lateral flexion-extension films) uncover clinically significant C-spine rheumatoid disease. Positive findings should prompt neurosurgical evaluation and appropriate C-spine MRI studies. By this process the author has had the fortunate occurrence of twice in the past 5 years uncovering silent but alarming mechanical instability of the atlantoaxial junction. In both cases, successful neurosurgical management was followed by elective foot and ankle surgery.

Preoperative vascular evaluation
To minimize wound and infectious complications, before surgery the preoperative assessment of all patients must include a thorough vascular evaluation, arguably the most important evaluation as it relates to postoperative surgical site complications [5].

Arterial evaluation
Special attention is given to evaluation of the patient's arterial inflow status. The physical examination should include documentation of dorsalis pedis and posterior tibial pulses, capillary filling, and signs of venous insufficiency. Nonpalpable pedal pulses are an indication for noninvasive arterial studies, namely ankle-brachial indices (ABI). The author generally also reviews segmental limb pressures, because this information can key the clinician in to the anatomic location of arterial disease. A rule of thumb is that from one contiguous arterial segment to the next, there should not be a drop-off of more than 15 to 20 mm Hg pressure. When pulses are readily palpable further vascular evaluations are generally not required. A notable exception to this general rule is when the patient is a smoker and long-standing digital deformities are present. In these instances, when forefoot procedures are planned, it must be realized that distal to the arcuate artery, the caliber of the terminal arterial branches taper rapidly. Deformed toes fixed in place may now also have digital vessels fixed in position. Intimal changes, tunica media calcifications, and low-flow states place these toes at considerable risk. Surgical correction of the toe and simple manipulation of the toe at surgery may stretch the vessels past their tolerance, resulting in vascular embarrassment of the digit. Patients must be counseled as to the potential for digital vascular embarrassment after correction of long-standing toe deformities, which may result in simple tip necrosis to a frankly gangrenous toe. Preoperative evaluation by way of toe pressures and digital waveforms (pulse volume recordings) are helpful in establishing a baseline knowledge of toe vascularity and the potential for vascular embarrassment or healing complications. Toe pressures should be greater than 40 mm Hg for adequate healing, with pulsatile waveforms. Transcutaneous oximetry (TcPO2) (transcutaneous oxygen measurements [TCOMs]) may also provide valuable information as to surgical wound healing potential. The author uses TcPO2 when noninvasive arterial studies have proven suboptimal inflow and vascular reconstruction is not feasible. TcPO2 values should generally be at greater than an absolute value of 40 mm Hg or a regional perfusion index (RPI) of 0.6 to ensure a high probability of healing. TcPO2 values of 20 to 40 mm Hg or an RPI of 0.4 to 0.6 indicate an intermediate probability of healing; TcPO2 values less than 20 mm Hg or an RPI less than 0.4 indicate a low probability of healing. Smokers are advised to quit smoking. Patients who have vasospastic disease are instructed to avoid all caffeinated beverages and foods in the perioperative period. In at-risk individuals, including patients who have marginal noninvasive studies, vasospastic disease, a subcutaneously tunneled arterial bypass, recent in situ lower extremity bypass (8 weeks), recent prosthetic graft leg bypass (8 weeks), or heavy smokers, a tourniquet during surgery is best avoided.

Venous evaluation
The venous examination, although less sophisticated, is helpful in determining which patients may develop delayed wound healing and persistent postoperative foot edema. The author simply searches for dependant rubor and signs of chronic venous stasis (history of ulceration, chronic skin pigmentation, and brawny induration) and performs a standing venous filling examination. Preoperatively edema control is achieved with Unna boots, compression stockings, or custom sequential compression devices. General skin conditioning is commenced with gentle emollients; treatment of localized dermatoses is used starting 3 to 4 weeks before surgery, up to the day of surgery, and postoperatively as needed. Psoriasis patients may harbor higher numbers of flora or even resistant flora; for these patients the author initiates daily washing of the part with chlorhexidine solution for 1 week before surgery. Patients who have severe psoriatic lesions are referred to dermatology before surgery.

Perioperative rheumatoid medications and complications

Steroids, methotrexate, D-penicillamine, cyclophosphamide, azathioprine, cyclosporine, leflunomide, nonsteroidal anti-inflammatory agents, and in particular, tumor necrosis-alpha (TNF-a) antagonists have all been implicated as causal agents for perioperative wound healing and infectious complications [6-9]. Because of the ever increasing role of the TNF-a antagonist in the medical management of RA, the impact of these agents on perioperative complications is explored in detail.

Effect of TNF-a inhibitors: basic science
TNF-a inhibitors have become more commonplace for the treatment of recalcitrant disease and the early management of RA. TNF-a is a ubiquitous 56-kD protein cytokine produced by endothelial cells, macrophages, and Tcells that is encoded on chromosome 6. TNF-a is an important mediator in the normal inflammatory cascade and is believed to be required for normal tissue healing and immune surveillance [7,10,11]. Several cell signaling pathways are believed to be involved with the TNF-a pathway. Recently the neutrophil activation antigen CD-69 has been found to be expressed at baseline in patients who have RA but reduced after TNF-a inhibitor therapy [12]. TNF-a inhibitor therapy has also been demonstrated to result in apoptosis of macrophage/monocyte lineage cells, which has also been suggested as a potential mechanism for inhibition of inflamed joint synovium in patients who have RA [13]. TNF-a antagonists have also been demonstrated to down-regulate the OPG/RANKL system of already TNF-primed osteocytes and endothelial cells [14]. These findings and others point to a complex array of functions that involve TNF-a. and the immune system. There is, however, no compelling basic science data demonstrating a certain and critical suppression of immune system function as it relates to the suppression of vital function of the neutrophil/macrophage system that would lead to an immune system left incompetent to survey and combat bacterial infections.

Numerous basic science investigations have demonstrated that the direct application to healing tissue [15-17] and the systemic administration of TNF-a during wound healing [18] result in a decrease in collagen production and wound strength by way of down-regulation of collagen gene expression [19]. These counterintuitive results regarding the inhibitory effect of TNFa on wound collagen production are explained by experimental data demonstrating that the inhibitory effect of TNF-a is attenuated by the presence of an intact IL-1 counter-regulatory system [20].

Additionally, experimental data have shown that the binding of excess TNF-a by specific antibodies ameliorates the negative influence of TNF-a on wound healing [18,21].

Clinical studies
Examples of TNF-a antagonist medications used to treat patients who have RA include etanercept (Enbrel) and infliximab (Remicade). Etanercept is a recombinant soluble human TNF receptor that is covalently linked to a human IgG Fc fragment (p75 sTNFR:Fc fusion protein), which is administered subcutaneously twice a week. Infliximab is a chimeric human/mouse anti-TNF monoclonal antibody administered intravenously every 4 to 8 weeks. Both agents exert their main action by blocking the binding of synovial TNF-a to immunocompetent target cells, preventing the release of inflammatory mediators.

Clinically several reports warn of an increase in serious infections in patients receiving TNF-a antagonists [22,23]. Most notably a recent large British national prospective observational study of more than 7000 patients receiving TNF-a antagonist were compared with control patients who had RA in relation to the development of serious infections, defined as infection requiring hospitalization, intravenous antibiotics, or death. The investigators found that patients who had RA receiving TNF-a antagonists had a higher rate of serious infection; however, the overwhelming leading type of infection was lower respiratory. Causal organisms were mostly Mycobacterium spp., Legionella pneumophila, Listeria monocytogenes, and Salmonella. Bone and soft-tissue infections were 67% less common than respiratory infections. Urinary tract infections assumed a similar role as bone and soft tissue. Additionally, data specifically addressing the orthopedic surgery status among study patients was not provided [24]. Although these data underscore the potential severe nature of infection and TNF-a inhibition treatment, the types of infections seem to be predominantly intracellular lung infections, those of which that are not typical of a postoperative orthopedic infection after clean elective surgery.

To date only a few studies have specifically addressed complications in patients who have RA and the relationship of RA medications to complication rates. In the general orthopedic arthroplasty literature it has been clearly shown that methotrexate does not need to be discontinued before elective orthopedic surgery; in fact some investigators have noted lower infection rates in patients who have maintained their methotrexate regimen in the perioperative period [25,26]. Other investigations have not been able to discern an independent risk for perioperative joint arthroplasty wound and infectious complications based on RA medications [27]. Reports examining the affect of leflunomide on perioperative elective orthopedic surgery seem to be equivocal [28,29].

The literature specific to complications of RA foot and ankle surgery is limited to only a few studies. In 2003 Bibbo and colleagues [30] reported on complications in 104 patients who had RA who underwent clean, elective foot and ankle surgery. The investigators' study population underwent 725 procedures with an overall 32% complication rate. Wound healing problems were the most common complication, followed by superficial infections and delayed/nonunions; all complications were minor. Logistic regression analysis failed to reveal any statistical association between RA medications, age greater than 55 years, gender, number of procedures per patient, or presence of rheumatoid nodules, and the occurrence of healing or infectious complications. In this study, NSAIDs, steroids, methotrexate, hydroxychloroquine, and gold were the primary RA medications analyzed. In a follow- up study that specifically examined the effect of TNF-a antagonists on wound healing and infectious complications after foot and ankle surgery, Bibbo and Goldberg [10] prospectively followed patients undergoing clean, elective foot and ankle surgery for the development of complications in the postoperative period. Although the patient group who continued their TNF-a antagonists in the perioperative period had more smokers than the non-TNF-a group, fewer complications were seen in the TNF-a group (P ¼ .033). Additionally this study did not find an increased risk for complications in those patients taking leflunomide. Although the study number was small (N ¼ 33), these findings are important to consider in patients whose RA symptoms are disabling when their TNF-a antagonist medications are withheld. As a continuation of that database, the author currently has more than tripled the original study number; preliminary analysis of the expanded database reveals nearly identical results.

Other RA foot and ankle surgery studies have provided data on general complication rates (not specifically examining the influence of medications). Reize and colleagues [31] reported a 14% infection rate (equally distributed among superficial and deep) after metatarsal head resections. Anderson and colleagues [32] report a 4% nonunion and a 12% deep infection rate in patients who have RA undergoing tibiotalocalcaneal fusion with retrograde IM nailing. The same investigators reported only a 26% primary union rate for ankle fusions performed with compression screws [33]. These data are in stark contrast to those of Nagashima and colleagues [34], who reported a 100% union rate, a 0% infection rate, and a 24% wound complication rate for patients who had RA who underwent ankle fusion with a modified IM nail technique.

An important consideration in any surgical candidate is bacterial skin colonization. One unique study examined the bacterial carriage rate of patients who had RA receiving TNF-a antagonist therapy compared with control patients. The investigators demonstrated that the carriage rate of Staphylococcus aureus in patients who had RA on TNF-a inhibitors was not increased over control subjects [35]. This finding is of particular interest in light of the fact that the most common organism causing infection in patients who have RA undergoing orthpedic surgery is S. aureus [36].

Based on available data, the author's opinion at this time is thus that although there exists theoretic risks for perioperative complications associated with specific RA medications, the major risk for developing a postoperative wound or infectious complication seems to be more related to the inherent risk of the procedure, host fragility (inherent to RA?), and the genomic make-up of the host for healing, rather than the influence of a specific medication on a specific host's healing capability. When dealing with the issue of the perioperative management of RA medication in patients undergoing clean, elective foot and ankle surgery, the author uses a simple algorithm for the management of perioperative medications that is based on the available basic science of wound healing, the orthopedic literature, and the author's own current database on complications in RA foot and ankle patients.

For those patients undergoing clean, elective RA foot and ankle surgery (especially for those whose disease symptoms are overwhelming without their RA medications), the author does not require discontinuation of RA medications in the perioperative period. Notable exceptions include:

1. Patients who have a documented history of poor wound healing (in particular, methotrexate is held); these patients should have their medications held based on each individual drug half-life.

2. Patients who develop a postoperative infection or a wound-healing complication (in particular, methotrexate, steroids, TNF-a inhibitors, and IL-1 inhibitors are held); medications may be resumed after wounds are healed or infections cleared.

3. When operating on patients referred for infection, these medications are withheld immediately and resumed as outlined (Table 1).

A special mention should be made as to those patients who have RA who pose a triple threat for complications: patients who have RA who smoke, drink, and have poorly controlled type I diabetes (as measured by HbA1c). This subset of patients who have RA is a challenge, and a conservative, individualized approach must be taken to their perioperative medication plan. Additionally, more frequent postoperative follow-up than usual is recommended (every 1-2 weeks until healed).

Acute deformity correction and impact on complications

Patients presenting with long-standing deformities present the risk for complications stemming from lack of tissue pliability from being in a chronically deformed state. This is particularly important when considering surgical correction of toe deformities. Over time, hammer and claw toes become ever more rigid; metatarsophalangeal joint subluxation proceeds to dorsal dislocation. With this process, the digital vessels match the deformity and assume a shortened position. Acute toe correction with straightening temporarily either stretches or buckles the vessel, narrowing the lumen and potentially injuring the intima. Intimal injury may lead to vessel thrombosis, placing the toes at risk for relative irreversible critical ischemia or venous impairment. Power saw vibrations may also set the stage for acute vessel spasm and vascular embarrassment. Surgeons who perform a large volume of RA cases in their career have almost uniformly experienced digital ischemia or venous engorgement after RA toe surgery. To prevent permanent ischemic changes, it is incumbent on the surgeon to quickly implement measures to reverse the problem (Figs. 1 and 2).

Another major consequence of acutely correcting a chronic deformity in the patient who has RA includes difficulties in wound closure and healing. This becomes important when straight incisions are placed over concave surfaces (contracted skin) or when angular corrections (lengthening) are performed acutely within a chronically shortened segment. This may lead to an inability to close incisions, excessive skin tension and subsequent necrosis of incisions, and de novo skin breakdown on areas in which incisions have not been placed. In these instances, careful planning of incision placement (eg, plantar approach to metatarsal heads) and the availability of back-up plans for wound closure (eg, VAC, local tissue rotations) are paramount. In instances in which acute correction places the skin at risk, it is often best to provide relief of length (ie, shorten instead of lengthening) to maintain the integrity of the soft-tissue envelop.

Intraoperative measures to prevent complications

Routine preoperative prophylactic antibiotics are used, typically a firstgeneration cephalosporin such as cefazolin (modified from [5]). When patients are penicillin/cephalosporin allergic, clindamycin (900 mg) or levofloxacin (500 mg) are used in the perioperative period. Vancomycin is used for patients who have a history of MRSA colonization (and a documented lack of eradication of such colonization). Patients who have RA and who are on steroids often possess thin, fragile skin, with only a thin layer of subcutaneous tissue. Delicate, atraumatic operative techniques must be used. Tourniquet use should be limited in these at-risk patients, but when a tourniquet must be used, the total tourniquet time should never exceed 120 minutes. The lowest pressure setting to provide visualization is used; a pressure of 125 points above the systolic provides adequate hemostasis for visualization of the operative field. Often in thin-skinned patients and elderly patients a tourniquet should not be used. When moving to a second operative site during surgery (eg, harvesting iliac crest bone graft), the limb tourniquet should be released and the foot/ankle wounds packed and wrapped with gentle compression (sterile Coban or an Esmarch works well). When returning to the main operative field, the surgeon often finds that bleeding is well controlled and tourniquet reinflation is not required.

In those patients on anticoagulants and antiplatelet agents, steps should be taken to minimize hematoma formation. The author does not require that patients discontinue Coumadin or Plavix (clopidogrel) before elective surgery; often the risks of discontinuing therapeutic anticoagulation outweigh the risks of the effect on hemostasis. The tourniquet should be released before subcutaneous closure and bleeding assessed. When general anesthesia is used, relative hypotension is common. The reduced limb perfusion pressure may give the false impression of controlled hemostasis. A systolic pressure of approximately 95 mm Hg is required for a femoral pulse, with incrementally higher inflow pressures required to adequately fill more distal arterial segments. In these instances the anesthesia staff should be instructed to bring the blood pressure to greater than 100 to 110 mm Hg systolic, which can be done by lightening-up the patients (vasopressors are not needed) and hemostasis reassessed. Packing the wound with Gelfoam and applying a pressure wrap for 5 to 10 minutes controls most bleeding. When surgical bleeding from a broad surface does not respond to cautery/ suture or pressure, a slurry of Gelfoam and thrombin (1 g powdered Gelfoam þ thrombin 1000 U in 10 ml sterile saline for injection) applied into the operative field with an overlying pressure bandage is helpful to control bleeding. Persistent brisk bleeding beyond this may require checking the aPTT/PT-INR and infusion of fresh frozen plasma.

This same approach can be helpful in instances in which digital perfusion is questionable after placement of retrograde pins. When digital inflow is in question, the first step is to monitor the response to an elevation in blood pressure. Warming the digits with warm saline-soaked gauze on the toes and at the ankle may assist in breaking an episode of vasospasm. Creating a relative shortening of the digit by sliding the toe up the pin can take off any excess stretch of digital vessels. The local application of nitroglycerin ointment or intra-arterial lidocaine is rarely needed. Ultimately the author has found that an unresponsive toe requires pin removal. Once a toe pin is removed for this reason it should not be replaced, because the original problem most often recurs. Local application of lidocaine to the operative field may be helpful if vasospasm or venous engorgement persists. Postoperative venous engorgement of a toe is treated best with elevation and observation. If cyanosis persists, the pin is removed. Postoperatively the limb is rested at heart level and ice is avoided. Smoking, nicotine patches, and caffeine are restricted. In most instances, careful observation results in a good outcome (Table 1). Simple digital tip necrosis usually results in an acceptable cosmetic and functional result. The surgeons must counsel the patients before surgery of these potential complications.

Immediate postoperative course

Antibiotics are used for 24 to 36 hours postoperatively. Drains are generally removed in 24 hours (modified from [5]). Edema control is balanced with positioning of the foot for optimal perfusion. Elevation is generally acceptable at two-pillow height. It may be best to avoid the application of ice after digital cases, especially when concern exists for the vascular status of the part. Caffeine and all nicotine products are restricted. Supplemental oxygen therapy is considered in at-risk toes, when flaps are created, and when susceptible incisions are created, such as incisions used for the approach for total ankle arthroplasty. Nasal cannula oxygen may be uncomfortable for the patients, thus humidified oxygen may provide better patient compliance (eg, 50%-100% humidified face mask O2 for 24 h). Ambulation with assistance/ physical therapy is begun as soon as possible. Postoperative nausea may be a significant hindrance to patient recuperation after general anesthesia, especially in elderly patients and those who have gastrointestinal dysmotility. Antiemetics are effective, especially when coupled with promotility agents.

Pain control

Pain control is paramount for patient comfort and to promote progression in physical therapy (modified from [5]). Regional anesthesia provides excellent postoperative pain relief during the immediate postoperative period and may be continued after discharge through the use of outpatient indwelling regional nerve sheath catheters. In addition to regional anesthesia modalities, patient-controlled analgesia (narcotics) should be implemented for postoperative pain control after large reconstructions. NSAIDs (eg, parenteral Toradol) are administered as a scheduled supplement for the first 24 to 36 hours. The author's observation has been that the short-term use of NSAIDs does not adversely affect soft-tissue or bony healing. Long-acting oral opioids with rescue dosing (eg, OxyContin, 10-20 mg orally every 12 h plus OxyIR, 5 mg orally every 4-6 h as needed for break-though pain) may be started on postoperative day number 1. The author continues this regimen for up to 2 weeks and then transitions to shorter duration agents (eg, Vicodin) thereafter. In most instances, with early aggressive pain control, narcotic use may be discontinued after 3 weeks. Intramuscular medications should be avoided in patients being actively anticoagulated.

Deep vein thrombosis prophylaxis

Although RA does not impart a higher risk than the general population for the development of a deep venous thrombosis (DVT), these patients may quickly become sedentary after lower extremity orthopedic surgery and thus may be considered at-risk for DVT from a functional standpoint (modified from [5]). Rheumatic diseases may coexist in patients, and other rheumatic disorders, such as systemic lupus, carry a higher risk for DVT. All patients have compression stockings on the unaffected limb, preferably commencing in the operating room. It is the author's practice to place all patients who have RA who will be non-weightbearing on unfractionated heparin (eg, Lovenox, 30 mg SQ every 12 h or 40 mg SQ daily) within 24 hours postoperatively while in the hospital. Patients continue daily dosing (eg, Lovenox, 40 mg SQ daily) for 2 to 3 weeks on discharge from the hospital. Outpatients are generally undergoing procedures that allow full or partial weightbearing, and for patients who have no prior history of a DVT a daily baby ASA for 2 to 3 weeks has proven adequate in this setting. Any patient who is actively anticoagulated or is undergoing antiplatelet therapy (ASA, clopidogrel) has their medication resumed immediately postoperatively. Patients on warfarin with therapeutic target INRs of R3 are started on unfractionated heparin, which is resumed immediately postoperatively to a therapeutic aPTT while their warfarin is resumed until their target INR is reached.

Any patient who presents in the postoperative period with clinical signs and symptoms of a DVT should undergo lower extremity duplex scanning. Scans positive for a DVT in the deep femoral-popliteal system are begun on immediate anticoagulation with either full-dose fractionated heparin or unfractionated heparin with transition to warfarin. Heparin therapy for at least 36 hours has been believed to assist in decreasing vessel wall inflammation. Oral anticoagulation should be continued for 3 to 6 months. Acute thrombosis in the calf veins that are within 2 cm of the popliteal system or in the saphenous system within 2 cm of the fossa ovalis should be considered for therapeutic anticoagulation. The author has observed that these thromboses tend to propagate more readily into the deep venous system. Otherwise calf vein involvement may be reasonably treated with ASA with repeat scanning at 2-week intervals to assess for clot propagation. Propagation of calf thrombi should be managed as a deep thrombosis, especially in prothrombotic populations. Symptomatic treatment of minor thrombosis includes elevation, warm moist heat, and oral NSAIDs.

Subacute postoperative considerations

Suture removal
Postoperative edema on an extremity with multiple skin incisions paints a clear picture that illustrates a need for ample time for wound healing, especially in postmenopausal women and in men [1,2] (modified from [5]). The author therefore does not rush to remove sutures; sutures or staples may need to be left in place for up to 4 to 6 weeks or more until edema and wounds are adequately healed. To minimize the risk for pin-track infections, however, percutaneous pins should be removed as soon as possible, usually at 4 to 6 weeks. Pins that have backed out should never be reinserted. Patients should be instructed to tape in place in the extruded position any pin that has backed out and immediately notify the surgeon, especially if pain, erythema, and swelling occur.

Edema and hematoma management
During the postoperative course, edema is universal. TCOMs on edematous feet show decreases in oxygen tensions (modified from [5]). Excess interstitial fluid finds portals for exit and tends to perpetuate a situation of chronic seepage of fluid through pin tracts and incisions, a set-up for bacterial colonization and infection. Edema control enhances wound healing and relieves discomfort from excess interstitial fluid. In certain patients effective edema control may be difficult. Strict elevation, and when wounds have epithelialized, edema wraps (Unna boot, elastic wraps), however, are usually effective in controlling most localized edema. If edema fluid is found to be leaking through unhealed incisions and pin tracts, antibiotics may be added for a short course. Elderly patients who demonstrate new bilateral edema postoperatively should be suspect for fluid retention of renal/cardiac origin and referred to their primary care provider. In these instances, appropriate diuresis may assist in edema control. When surgical corrections are stable, compression garments used for 1 to 2 months postoperatively are of great benefit. Compression garments should impart 20 to 30 mm Hg of pressure for typical postoperative edema.

Patients who are anticoagulated may develop postoperative subcutaneous hematomas. If large, these may be aspirated early if incisions or the overlying skin is in jeopardy, or these may be aspirated late (during liquefaction before resorption). Patients who have RA are prone to infectious complications, whether on steroids or not, with increased mortality from infection. The author therefore does not hesitate to administer a short course (7-10 days) of prophylactic antibiotics (eg, Keflex, 500 mg orally once daily) in patients who have a hematoma or continued serous drainage, especially if a low-grade fever, pain, or erythema is present. Frankly cellulitic areas overlying a suspected hematoma warrant consideration for urgent irrigation, debridement, cultures, and sensitivities. At this stage debridement may be minimal, with preservation of soft tissues. Left ignored, a simple complication may result in a large abscess area capable of compressing and rapidly thrombosing small vessels and producing widespread deep necrosis. In patients on active anticoagulation presenting with an unusually large hematoma, surgical evacuation of the hematoma and evaluation of the patient's coagulation profiles (aPTT/PTT-INR) should ensue.

References:
  1. Engeland CG, Bosch JA, Cacioppo JT, et al. Mucosal wound healing: the roles of age and sex. Arch Surg 2006;141(12):1193–7.
  2. Lenhardt R, Hopf HW, Marker E, et al. Perioperative collagen deposition in elderly and young men and women. Arch Surg 2000;135(1):71–4.
  3. Agah A, Kyriakides TR, Letrondo N, et al. Thrombospondin 2 levels are increased in aged mice: consequences for cutaneous wound healing and angiogenesis. Matrix Biol 2004;22(7): 539–47.
  4. Roh C, Lyle S. Cutaneous stem cells and wound healing. Pediatr Res 2006;59(4 Pt 2): 100R–3R.
  5. Bibbo C. The assessment and perioperative management of patients with rheumatoid arthritis. Techniques in Foot & Ankle Surgery 2004;3:126–35.
  6. Gorman JD, Sack KE, Davis JC Jr. Treatment of ankylosing spondylitis by inhibition of tumor necrosis factor alpha. N Engl J Med 2002;346:1349–56.
  7. Moreland LW, Bucy RP, Weinblatt ME, et al. Immune function in patients with rheumatoid arthritis treated with etanercept. Clin Immunol 2002;103:13–21.
  8. Weinblatt ME, Kremer JM, Bankhurst AD, et al.Atrial of etanercept, a recombinant tumor necrosis factor receptor:Fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. N Engl J Med 1999;340:253–9.
  9. Yazici Y, Erkan D, Lockshin MD. A preliminary study of etanercept in the treatment of severe resistant psoriatic arthritis. Clin Exp Rheumatol 2000;18:732–4.
  10. Bibbo C, Goldberg JW. Infectious and healing complications after elective orthopaedic foot and ankle surgery during tumor necrosis factor-alpha inhibition therapy. Foot Ankle Int 2004;25(5):331–5.
  11. Alldred A. Etanercept in rheumatoid arthritis. Expert Opin Pharmacother 2001;7:1137–48.
  12. Capsoni F, Sarzi-Puttini P, Atzeni F, et al. Effect of adalimumab on neutrophil function in patients with rheumatoid arthritis. Arthritis Res Ther 2005;7:R250–5.
  13. Catrina AI, Trollmo C, af Klint E, et al. Evidence that anti-tumor necrosis factor therapy with both etanercept and infliximab induces apoptosis in macrophages, but not lymphocytes, in rheumatoid arthritis joints: extended report. Arthritis Rheum 2005;52(1):61–72.
  14. Catrina AI, af Klint E, Ernestam S, et al. Anti-tumor necrosis factor therapy increases synovial osteoprotegerin expression in rheumatoid arthritis. Arthritis Rheum 2006;54(1):76–81.
  15. Rapala K. The effect of tumor necrosis factor-alpha on wound healing. An experimental study. Ann Chir Gynaecol Suppl 1996;211:1–53.
  16. Rapala K, Laato M, Niinikoski J, et al. Tumor necrosis factor alpha inhibits wound healing in the rat. Eur Surg Res 1991;23(5–6):261–8.
  17. Rapala KT, Vaha-Kreula MO, Heino JJ, et al. Tumor necrosis factor-alpha inhibits collagen synthesis in human and rat granulation tissue fibroblasts. Experientia 1996;52(1):70–4.
  18. Lee RH, Efron DT, Tantry U, et al. Inhibition of tumor necrosis factor-alpha attenuates wound breaking strength in rats. Wound Repair Regen 2000;8(6):547–53.
  19. Rapala K, Peltonen J, Heino J, et al. Tumour necrosis factor-alpha selectivity modulates expression of collagen genes in rat granulation tissue. Eur J Surg 1997;163(3):207–14.
  20. MaishGO3rd, Shumate ML, Ehrlich HP, et al. Interleukin-1 receptor antagonist attenuates tumor necrosis factor-induced alterations in wound breaking strength. J Trauma 1999;47(3): 533–7.
  21. Regan MC, Kirk SJ, Hurson M, et al. Tumor necrosis factor-alpha inhibits in vivo collagen synthesis. Surgery 1993;113(2):173–7.
  22. Bongartz T, Sutton AJ, Sweeting MJ, et al. Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta-analysis of rare harmful effects in randomized controlled trials. JAMA 2006;295(19):2275–85.
  23. Listing J, Strangfeld A, Kary S, et al. Infections in patients with rheumatoid arthritis treated with biologic agents. Arthritis Rheum 2005;52(11):3403–12.
  24. Dixon WG,Watson K, Lunt M, et al. British Society for Rheumatology Biologics Register. Rates of serious infection, including site-specific and bacterial intracellular infection, in rheumatoid arthritis patients receiving anti-tumor necrosis factor therapy: results from the British Society for Rheumatology Biologics Register. Arthritis Rheum 2006;54(8):2368–76.
  25. Grennan DM, Gray J, Loudon J, et al. Methotrexate and early postoperative complications in patients with rheumatoid arthritis undergoing elective orthopaedic surgery. Ann Rheum Dis 2001;60(3):214–7.
  26. Perhala RS, Wilke WS, Clough JD, et al. Local infectious complications following large joint replacement in rheumatoid arthritis patients treated with methotrexate versus those not treated with methotrexate. Arthritis Rheum 1991;34:146–52.
  27. Escalante A, Beardmore TD. Risk factors for early wound complications after orthopedic surgery for rheumatoid arthritis. J Rheumatol 1995;22:1844–51.
  28. Tanaka N, Sakahashi H, Sato E, et al. Examination of the risk of continuous leflunomide treatment on the incidence of infectious complications after joint arthroplasty in patients with rheumatoid arthritis. J Clin Rheumatol 2003;9:115–8.
  29. Fuerst M, Mohl H, Baumgartel K, et al. Leflunomide increases the risk of early healing complications in patients with rheumatoid arthritis undergoing elective orthopedic surgery. Rheumatol Int 2006;26:1138–42.
  30. Bibbo C, Anderson RB, Davis WH, et al. The influence of rheumatoid chemotherapy, age, and presence of rheumatoid nodules on postoperative complications in rheumatoid foot and ankle surgery: analysis of 725 procedures in 104 patients. Foot Ankle Int 2003;24(1):40–4.
  31. Reize P, Ina Leichtle C, Leichtle UG, et al. Long-term results after metatarsal head resection in the treatment of rheumatoid arthritis. Foot Ankle Int 2006;27:586–90.
  32. Anderson T, Linder L, Rydholm U, et al. Tibio-talocalcaneal arthrodesis as a primary procedure using a retrograde intramedullary nail: a retrospective study of 26 patients with rheumatoid arthritis. Acta Orthop 2005;76:580–7.
  33. Anderson T, Maxander P, Rydholm U, et al. Ankle arthrodesis by compression screws in rheumatoid arthritis: primary union in 9/35 patients. Acta Orthop 2005;76:884–90.
  34. Nagashima M, Tachihara A, Matsuzaki T, et al. Follow-up of ankle arthrodesis in severe hind foot deformity in patients with rheumatoid arthritis using an intramedullary nail with fins. Mod Rheumatol 2005;15:269–74.
  35. Bassettei S, Wasmer S, Hasler P, et al. Staphylococcus aureus in patients with rheumatoid arthritis under conventional and anti-tumor necrosis factor-alpha treatment. J Rheumatol 2005;32:2125–9.
  36. Barbari EF, Osmon DR, Duffy MC, et al. Outcome of prosthetic joint infection in patients with rheumatoid arthritis: the impact of medical and surgical therapy in 200 patients. Clin Infect Dis 2006;42:216–23.