Revisional Surgery of Forefoot and Midfoot Amputations

Revisional Surgery of Forefoot and Midfoot Amputations

Monica Schweinberger


The increasing incidence of diabetes has resulted in a rise in diabetic complications including neuropathy, peripheral arterial disease (PAD), and end-stage renal disease leading to higher numbers of foot ulceration and subsequent amputation.1,2 The global annual incidence of diabetic foot ulcers is 6.3%. Forefoot ulcers are the most frequent comprising 56% of all diabetic foot ulcers and are more likely to lead to major amputation than midfoot or hindfoot ulcers.3 Partial foot amputation is being utilized to avoid higher level transtibial amputation (TTA) or transfemoral amputation (TFA) in the hopes of maintaining mobility and improving quality of life.1,4 The incidence of toe amputation is 10 times greater than all other forms of partial foot amputation.5

Some studies have reported revision rates for partial foot amputation as double that seen with TTA.1 However, a 2019 study by Czerniecki et al showed similar rates of reamputation for transmetatarsal amputation (TMA) and TTA except in patients with diabetes, renal failure, or the combination of diabetes and ipsilateral revascularization within 1 year, in which TMA had higher revision rates. Current smoking, alcohol abuse, and rest pain or gangrene were also predictive of reamputation for both TMA and TTA groups.4 Humphrey et al published on 36 TMAs with primary wound closure and a mean follow-up of 2.3 years. Reoperation was required in 4 patients including 3 (8%) major amputations.6 A 2019 study by Miller et al identified an increased risk of TTA in patients who failed to heal their TMA within 3 months.7 Another retrospective study by Kaiser et al in 2019 looked at 105 TMAs with a 40% revision rate. Nine percent of patients underwent soft-tissue revision, 17% underwent minor amputation, and 14% underwent a major amputation. Most commonly, revision surgery was performed for delayed healing, new soft-tissue infection, persistent or new osteomyelitis, gangrene, or ulceration.8 In 2020, Jupiter et al compared complication rates of TMA and other minor amputations to major leg amputations. They found that the minor amputees had a 2.5 times greater risk of returning to surgery for irrigation and débridement, while the major amputees had a higher risk of postoperative urinary tract infection and need for blood transfusion. Major amputees tended to have more severe comorbid disease, were less likely to have a dirty or infected surgical site, and were less likely to be discharged home. They were also more likely to be functionally dependent than those patients undergoing minor amputation.9

Lisfranc and Chopart amputations are generally performed as a last resort for limb salvage. Lisfranc amputations are less desirable than TMAs due to a reduced level of function, increased reulceration risk, and decreased longevity of the stump.10 Faglia et al published on 83 patients who underwent Chopart amputation at their institution with a mean follow-up of 2.8 years. Only 56.6% of their patients healed in a mean interval of 5.4 months. Of these, 16 patients underwent primary closure with a mean healing interval of approximately 3 weeks and 31 patients healed by secondary intention within a mean interval of 7.5 months. Ulcer recurrence occurred in 31.9% of patients, all of which healed ­without additional surgery within a mean interval of 7 months. Of the total, 27.7% of the patients underwent major amputation and 15.7% died prior to healing of their Chopart amputation site.11 These studies suggest that choosing an amputation level in which primary or delayed primary closure can be achieved might improve healing rates and reduce the likelihood of more proximal amputation.

Chu et al retrospectively reviewed clinical outcomes of toe amputation over a 5-year period with an 82.4% healing rate. Reamputation was seen in 12.5% of patients in the first year, 22.3% of patients in the third year, and 47.1% of patients in the fifth year after the initial surgery. Independent predictors of reamputation included hemoglobin A1C (HgbA1C) > 9% and age older than 70 years.12 A study by Doyle et al looking at the effects of endovascular timing on the success of pedal amputation found that toe and partial ray amputations had greater rates of healing than TMA.13 Ahn et al looked at the association between end-stage renal disease and TMA failure. End-stage renal disease was associated with a 100% increased likelihood of TMA failure, a 128% increased likelihood of major amputation, and a 182% increased likelihood of 30-day mortality. There was no increase in failure of TMA with worsening kidney function prior to renal failure.14 Kantar et al found an 11.9% wound complication rate in a group of 2316 patients who underwent TMA from 2009 to 2015. Increased operative time and obesity were both identified as risk factors for wound complications.15 An understanding of how comorbidities could influence surgical outcomes in partial foot amputation should guide proper patient selection.

A systematic review in 2016 by Quigley and Dillon found insufficient evidence to conclude whether the quality of life of patients with partial foot amputation was significantly different from that of patients undergoing TTA.1 Dillon et al performed a subsequent systematic review in 2017 comparing outcomes of dysvascular partial foot amputation with TTA. They found a 25% reamputation rate for partial foot amputations within 1 year. Additionally, mobility for most people declined following TMA and TTA; however, in the first months following TMA, mobility may be better than TTA.17 Young et al compared complete TMA to preserving the first ray or first 2 rays and found significantly better ambulatory function in the ray group.18 Studies of partial first-ray resection have shown abnormal gait patterns and increased pressure under the second-fourth metatarsal heads,19,20 which could lead to transfer ulceration. A retrospective study out of Switzerland looking at 185 patients found a 38.4% revision rate after partial ray resection and an 11.9% subsequent major amputation rate. Transfer ulcerations occurred in 21.6% of patients, mainly in adjacent rays, indicating the need for preventative off-loading. The mean time to revision surgery was 1.4 years, and patients were able to walk almost normally after healing from the procedure.21 The pre- and expected postprocedure function of the patient should be considered during surgical planning, particularly as it affects long-term independence.

In 2019, Norvell et al reported the risk of mortality within the first year after amputation in patients with diabetes and/or PAD. Mortality for those undergoing TMA was 17.7%, TTA was 24.7%, and TFA was 34.5%.22 An editorial by Dillon et al in 2017 contended that the highest amputation mortality rates occur in older patients with more advanced systemic disease and multiple comorbidities. Studies that controlled for these confounding factors found no difference in mortality rates by amputation level.23 In 2016, Thorud et al published a systematic review on mortality rates after nontraumatic major amputation in patients with PAD and diabetes. They found a high 5-year mortality rate for TTA from 40% to 82% and for TFA from 40% to 90%. Mortality risk factors included renal disease, advanced age, proximal amputation, and PAD. They also noted the need for further studies to determine if amputation increased the risk of death or was just an indication of the severity of systemic disease.24 A meta-analysis by Stern et al looking at mortality following major amputation reported rates at 1 year of 47.9%, 2 years of 61.3%, 3 years of 70.6%, and 5 years of 62.2%. Factors associated with increased mortality rates included diabetes, PAD, coronary artery disease, cerebrovascular disease, renal disease, American Society of Anesthesiologists (ASA) class greater than or equal to 4, dementia, and nonambulatory status. Higher amputation level and staged surgery with initial guillotine amputation also increased mortality risk.25 A retrospective study looking at amputations below the ankle in Denmark from 1996 to 2013 identified a 43% 5-year mortality rate for patients with type 2 diabetes and a 52% mortality rate for those with type 1 diabetes. The average time from amputation to death was 2.5 years.26 Mortality may or may not be associated with amputation level, so decision-making should be focused on the level at which the individual patient can reasonably be expected to heal and that will have the least long-term impact on their functional status and lifestyle.

Indications and Contraindications

The indications for forefoot and midfoot amputations include osteomyelitis, dry or gas gangrene, severe deformity resulting in recurrent ulceration, and unreconstructible trauma. The level of amputation must be determined by the extent of infection, the vascular supply, available soft-tissue coverage, and the likely functional result. For example, if a patient has had 2 or more ray amputations and has developed osteomyelitis in a third metatarsal, a TMA may be appropriate if there is adequate skin coverage and circulation for healing at that level. Conversely, a single toe or ray amputation would be preferred when infection or gangrene is confined to this area in a patient without previous amputation. With TMA and Lisfranc amputation levels, consideration must also be given to the necessity for tendon balancing to improve the likelihood of long-term success.27,28 Additional incisions may be required to allow for proper tendon balancing, which will also need adequate circulation to heal. In addition, tendon balancing procedures could change the postoperative course, requiring non–weight bearing for a longer period, which may or may not be feasible depending on the functional status of the patient. Nursing home placement and/or extended healthcare facilities may be required for many of these individuals during their recovery.

Contraindications for partial foot amputation would include inadequate circulation for healing with an inability to be revascularized, inability of the patient to comply with postoperative instructions required for healing, and severe deformity that would result in a nonfunctional extremity. A relative contraindication for a particular amputation level would be insufficient soft-tissue coverage for wound closure. There is some evidence to suggest that closure by secondary intention may result in more amputation failures than primary or delayed primary closure or closure by flap or graft coverage.68,11 Therefore, if incomplete soft-tissue coverage is available at one amputation level, a higher level amputation may be preferred. The preoperative global health of the patient must be taken into consideration when choosing limb salvage versus primary TTA or TFA. If the patient has a high ASA classification, a smaller number of surgeries along with a shorter postoperative recovery at the amputation level with the highest likelihood of successful healing may be the best choice.

Preoperative Considerations

Diabetic patients have an increased risk of perioperative complications when compared with their nondiabetic counterparts. HgbA1C levels above 7.0% have been associated with an increased risk of surgical site infection, although no specific value has been identified as a cutoff for elective surgery. In situations where an amputation is required, there may not be time to optimize a patient’s HgbA1C prior to surgery; however, working with their primary physician and/or endocrinologist and a dietitian in the perioperative period could improve blood sugar control and aid healing over time. In addition, the patient should be counseled regarding the importance of good blood sugar control in reducing complications and improving postoperative outcomes.30,31

Peripheral neuropathy has also been identified as a risk factor for surgical site infection and complications with postoperative healing. Diabetic patients may have autonomic neuropathy that can affect multiple organ systems including the gastrointestinal system resulting in gastroparesis and the cardiovascular system causing cardiac volatility. This may increase the risk of aspiration and reduce the patients’ ability to increase heart rate, blood pressure, and cardiac output during anesthetic induction.31

In patients with a cardiac history, preoperative cardiology consultation may be indicated to determine perioperative risk of a cardiac event. Once the risk has been assessed, strategies to reduce perioperative risk may be implemented. Significant cardiovascular risk factors include age older than 70 years, myocardial infarction within 6 months, congestive heart failure, arrhythmia on preoperative electrocardiogram (ECG), and insulin dependent diabetes, among others.32

Chronic renal disease has detrimental effects on wound healing but can also increase the risk of postoperative cardiac complications and acute renal failure. Calculating the glomerular filtration rate is the most effective way of determining a patient’s renal risk. Platelet dysfunction can occur in this patient population increasing perioperative bleeding risk. Renal patients are also susceptible to fluid overload, requiring careful management of intravenous volume. Electrolyte imbalances particularly with sodium and potassium may be seen preoperatively and should be addressed prior to surgery and monitored perioperatively. Anemia is common in this population due to a decreased production of erythropoietin as kidney function declines. Erythropoietin-stimulating agents may sometimes be utilized, but transfusions may be necessary depending on expected procedural blood loss. Medications may require renal dosing, particularly with long courses of intravenous antibiotics that are often required in patients undergoing amputation.33 For patients on dialysis, procedures should be planned around their dialysis schedule. Consultation with a nephrologist may be indicated, depending on the extent of the patients’ renal dysfunction.

PAD has an obvious impact on wound healing of partial foot amputations. Preoperative assessment of the vascular supply is imperative for surgical planning. Initially, physical examination should identify the presence or absence of palpable pedal pulses. If pulses are absent, a Doppler examination in the office may provide an indication of the severity of arterial disease. Noninvasive arterial studies to include segmental pressures, arterial brachial indices (ABI), and toe pressures should be ordered when pulses cannot be palpated. Segmental pressures can identify a significant drop (>30 mm Hg) between the pressure at one level of the limb and another, indicating a blockage in that segment. An ABI of less than 0.9 is considered an indication of PAD but can be elevated with vascular calcification common in patients with diabetes and renal disease. Toe pressures are therefore deemed to be a more predictable determinant of healing due to their small vessels that are less affected by calcification. A toe pressure greater than or equal to 45 mm Hg has been associated with wound healing.31,34 Consultation with an interventional cardiologist or vascular surgeon is indicated in the presence of arterial disease. These specialists may recommend additional studies such as computed tomography (CT) angiography with runoff for more information regarding the extent of arterial disease. Care must be taken when administering contrast to patients with renal disease.33

Severity of infection may necessitate emergent amputation, in which case revascularization could be performed afterward, if required. If the amputation is not emergent, revascularization is often performed prior, which allows for a better determination of appropriate amputation level. Shi et al looked at 153 patients with PAD who underwent TMA and found that there was no significant difference in outcomes between patients who had revascularization before and after their amputation. They also could not identify a preferred time frame between revascularization and amputation. They did find a greater risk of major amputation in patients who underwent endovascular revascularization compared with open bypass.35 Amputation level will be determined by adequate arterial supply for healing along with available soft-tissue coverage and expected functionality of the residual limb.

There have been some conflicting studies regarding tobacco use and amputation healing. Czerniecki et al found that that smoking increased the risk of reamputation in TTA and TFA4; however, a 2017 systematic review regarding the effects of smoking and tobacco use in foot and ankle surgery found that partial foot amputations did not appear to have negative outcomes associated with smoking.36 Despite these findings, smoking cessation should be recommended to patients undergoing amputation, particularly those with PAD.

The physical examination for a patient requiring amputation should include a general evaluation for symptoms of nausea, vomiting, or altered mental status that could be signs of sepsis. A set of vitals should be taken to determine the possibility of systemic infection and assess the stability of the patient. High fever, low blood pressure, and rapid pulse could indicate a severe systemic infection and possible need for emergent surgical drainage. Laboratory testing should be performed for preoperative assessment of the patient and to help determine the severity of infection. A complete blood count (CBC) will identify an elevated white blood count (WBC) and/or left shift that could indicate a more severe infection. A significant increase in WBC is not always seen in diabetic patients, even in the presence of a severe infection, so the physical examination of the foot may better estimate the need for emergent surgery, hospitalization, and intravenous antibiotics in this patient population. The CBC will also identify anemia and give a platelet count to help assess bleeding risk. If a patient is anemic preoperatively or has a severely decreased platelet count, a blood or platelet transfusion may be indicated depending on the expected blood loss from the procedure. A basic metabolic panel may be ordered to evaluate kidney function and electrolyte balance. If kidney function is elevated, medication dosing may need to be altered. Electrolyte imbalances should also be addressed to reduce the risk of perioperative myocardial infarction. A lactic acid test may be done to rule out sepsis, which would result in lactic acidosis. In diabetic patients, a HgbA1C may be used to assess glucose control.

Albumin and prealbumin studies can be used to assess the nutritional status of the patient, since malnutrition may impair healing. Many patients with chronic wounds suffer from malnutrition, and healing may be improved with nutritional supplementation that could be recommended by a dietitian. Erythrocyte sedimentation rate and C-reactive protein level can identify inflammation that could be associated with osteomyelitis. These tests can be performed prior to surgery and periodically during the postoperative course to monitor the patients’ response to treatment. They should decrease as the infection resolves.

Prothrombin time, partial thromboplastin time (PTT), and international normalized ratio (INR) testing may be ordered prior to surgery to assess bleeding risk. If a patient is on a blood thinner and the surgery is not emergent, they may require discontinuation of the blood thinner a few days prior to surgery and possible bridging with an injectable anticoagulant. This should be discussed with the anticoagulation team at your hospital or the patient’s primary care physician.

A preoperative ECG would be indicated in patients with a history of cardiovascular disease or age older than 70 years. Presence of a pacemaker will require bipolar cautery use intraoperatively. If the patient has a history of pulmonary disease, a preoperative chest radiograph should be performed. When general anesthesia is planned, patients with rheumatological conditions might also require a preoperative cervical spine radiograph to rule out significant arthritis that could limit extension of the neck. Preoperative guidelines may differ depending on the institution and should be based on the individual patient’s history and clinical findings.29

Examination of the foot should include an evaluation of the skin, noting the size, depth, and location of any open wounds including signs of infection that may be present such as erythema, edema, calor, purulence, malodor, and probing to bone, which could indicate osteomyelitis.37 Any areas of fluctuance may indicate a deep space abscess that might require urgent surgical drainage. Crepitus on palpation could indicate gas gangrene, which would be a surgical emergency. The dorsalis pedis (DP) and posterior tibial (PT) pulses should be palpated. If nonpalpable, further evaluation and possible referral to a vascular surgeon or interventional cardiologist would be indicated, as discussed above. In cases requiring emergent surgery, such as gas gangrene, the initial surgery may need to be performed prior to additional vascular testing or specialist evaluation. A sensory examination should be performed with a 5.07 Semmes-Weinstein monofilament to determine if neuropathy is present. A musculoskeletal examination should be completed, evaluating any deformity that could increase the likelihood of reulceration after amputation, including equinus, varus, or valgus positioning of the foot; hammertoe deformity; and other deformities. Previous surgery near the intended amputation site must be identified, particularly if hardware is present as it may require removal. A previous ankle fusion could seriously affect the longevity of a TMA or Lisfranc amputation due to lack of ankle dorsiflexion that would result in increased pressure at the distal stump with a high likelihood of reulceration and/or amputation.

Preoperative imaging is critical in diagnosis and identification of the extent of osteomyelitis, ruling out deep space abscess, determining whether gas is present in the tissues, and evaluating deformity that could affect long-term amputation success. Standard radiographs should be ordered initially and, in combination with a thorough physical examination, may be all that is required for surgical planning in some cases. If possible, radiographs should be weight bearing for better assessment of foot structure and deformity. Radiographic changes associated with osteomyelitis may not be seen for up to 2 weeks after the onset of an infection. Findings positive for osteomyelitis could include periosteal reaction, bone erosion, focal rarefaction, or decreased bone density (Figure 31.1). Gas may be present in the tissue but could represent either a communication to open air, if only seen adjacent to an ulcer site, or a gas-forming organism, if more widespread. The soft tissue cannot be evaluated on a radiograph, and therefore a suspected deep space abscess would not be identified with these images alone.38

Figure 31.1 Significant destruction of the hallux interphalangeal joint consistent with osteomyelitis.

CT provides similar information to radiographs, although it does have reconstruction capabilities. Intravenous contrast may enhance soft tissues and possibly identify rim-enhancing fluid collections consistent with abscess formation. CT is limited in determining the extent of infection in bone and soft tissue.38 CT may identify changes associated with chronic osteomyelitis including sequestrum, involucrum, and cloaca.39

Magnetic resonance imaging (MRI) is highly sensitive for osteomyelitis and depicts detailed images of the soft tissue, allowing identification of ulcers, sinus tracts, and abscesses. MRI may identify osteomyelitis prior to radiographic imaging. It provides more information about the extent of disease than other imaging modalities3840; however, it may also overestimate the extent of bone infection (Figure 31.2A-C). MRI is contraindicated in patients with a pacemaker, recently implanted hardware, or ocular foreign body.

Preoperative antibiotics should be limited to situations in which no infection is present, for example, elective amputations being performed due to a painful deformity in the absence of an open wound. When amputations are performed to eradicate infection, patients brought to the operating room from the emergency room should have antibiotics delayed until after deep cultures are taken intraoperatively. For those patients admitted for infection who are later taken to the operating room, antibiotics will likely have been administered on the ward and no additional antibiotics are required preoperatively. In those cases, cultures may be taken on admission to the hospital after bedside débridement to inform intravenous antibiotic management.

Deep vein thrombosis (DVT) is a possible complication of partial foot amputation particularly when patients will be kept hospitalized, immobilized, and/or non–weight bearing for long periods postoperatively. These patients and those with preexisting risk factors such as a previous history of DVT may be considered for DVT prophylaxis perioperatively.

If wound closure will be delayed or split-thickness skin graft coverage is planned, a negative-pressure wound therapy device may need to be applied in the operating room and should be ordered preoperatively. In general, selecting an amputation level where primary or delayed primary closure can be accomplished may lead to more predictable and successful wound healing. If this is not possible, flap coverage including local random flaps, muscle flaps, pedicle flaps, perforator flaps, and free flaps are considerations along the reconstructive ladder for wound closure. Adequate circulation to the proposed flap is required for healing.41,42 Fluoroscopy is needed when performing TMA, to assess the metatarsal parabola, and when performing minimal-incision partial ray resection, to identify the level of the osteotomy prior to skin incision and assess for adequate bone resection.

It is important to understand the patient’s current functional status, living situation, and assistance at home when determining the proper amputation level. Substance abuse should be identified, and a plan for managing that, at least in the perioperative period, must be made. The surgeon must identify any factors that will impact healing and long-term function of the patient. If the patient requires non–weight bearing on the operated extremity postoperative, they may require training by physical therapy, modifications to their home to allow safe access and mobility, and assistance for daily tasks. If the patient is not physically able to safely remain non–weight bearing, they may require placement in an extended healthcare facility for some period after surgery. These issues need to be discussed in detail with the patient and plans must be made with hospital social workers for proper and safe discharge from the hospital.

When consenting a patient for partial foot amputation, the risks of reulceration, infection, and possible need for more proximal amputation need to be emphasized. These patients are generally high-risk surgical candidates with multiple comorbidities that would not normally be offered elective surgery. These risk factors, in addition to the likely presence of infection, make any surgery more complicated and therefore require strict attention to detail in all aspects of care. Revision surgery generally imparts even more risk, making it imperative to consider both social and host factors in determining the most appropriate level for definitive amputation to achieve a successful result.

Detailed Surgical Description

In amputations where infection is present, a staged approach may be appropriate. This, as with most partial foot amputations, requires inpatient management. A tourniquet is avoided in most of these procedures. The initial surgery involves removal of all infected bone and soft tissue, pulse lavage irrigation with normal saline (2-6 L depending on the size of the surgical site), and taking aerobic and anaerobic cultures to determine if any bacteria remain, followed by at least 2 L of additional pulse lavage with or without 50,000 units of bacitracin. An x-ray cassette bag can be placed over the foot and a hole cut in the end to allow the irrigator to be inserted into the bag, avoiding splatter, using suction to remove the fluid. After completion of irrigation, the surgeon’s and assistant’s outer gloves should be changed and a new towel placed under the patient. New instruments should be used from this point forward to avoid recontamination of the surgical site. All bleeding should be controlled with cautery or absorbable suture ties.

Antibiotic beads attached to 2-0 nylon suture and containing a mixture of vancomycin and gentamicin, if the patient has no allergy to these medications, should be placed into the open wound, which then can be covered by a nonadherent dressing followed by a dry, sterile dressing. Other antibiotics may be utilized in the antibiotic beads if preoperative cultures indicate resistance to those listed above. Removed bone and soft tissue are sent to pathology for evaluation and determination of clean bone margins. Radiographs should be taken in the postanesthesia care unit confirming the bone removed and the location of antibiotic bead placement. Antibiotic beads are generally left in place until 3 days after surgery when the initial dressing change is performed, and the tissues are evaluated for viability and any continued signs of infection. When they are removed at bedside, the coagulum should be left in place, as it still contains antibiotics. The coagulum is removed with a curette at the subsequent irrigation and débridement procedure.

A rise in the WBC is sometimes seen the day after surgery but should decrease in the subsequent postoperative period. If the WBC remains elevated and the patient remains febrile, other systemic signs of infection are present, or if there is excessive bleeding through the dressing, the surgical site may need to be evaluated sooner. If the surgical cultures are positive and/or bone margins are infected, or if the patient has continued clinical or systemic signs of infection, they may require repeat irrigation and débridement with antibiotic bead placement until cultures are negative, bone margins are clean, and signs of infection have resolved. Once that occurs, the patient can return to the operating room for a final irrigation and débridement with new aerobic and anaerobic cultures taken prior to wound closure. It is imperative that wound closure be delayed until the infection has been completely resolved. Ideally, patients should remain on intravenous antibiotics until approximately 3 days after final wound closure, at which point the postoperative dressing is changed and the surgical site is assessed. If there continues to be no signs of infection at this point, transition to oral antibiotics may be indicated for another 10 to 14 days and the patient may be ready for discharge to home or a skilled nursing facility.

In cases where infection is isolated to a particular toe, for example, the head of the proximal phalanx, where oral antibiotics have resolved clinical signs of infection, a single-stage toe amputation with immediate wound closure may be indicated. In this situation, outpatient surgery might be appropriate, even under local anesthetic, if the patient has adequate blood flow and is functionally able to follow postoperative instructions at home. Pulse lavage irrigation is still performed in these cases along with aerobic and anaerobic cultures to determine if any residual bacteria are present prior to wound closure. Oral antibiotics are generally continued 10 to 14 days postoperative, as a precaution, to avoid recurrent infection.

Oct 22, 2022 | Posted by in ORTHOPEDIC | Comments Off on Revisional Surgery of Forefoot and Midfoot Amputations
Premium Wordpress Themes by UFO Themes