Radical Head and Neck Surgery with Free-Flap Reconstruction
An estimated 55,000 Americans develop head and neck cancer mainly of the pharynx, larynx, and tongue annually and 12,000 die from the disease.1 Radical head and neck dissection with free-flap reconstruction has evolved over the past 4 decades with free-flap success rates in the 90% to 99% range.
Otolaryngology patients with tumors of the head and neck undergo radical dissection and free-flap microsurgery. These patients are admitted postoperatively to the ICU for neurologic, free flap, and airway monitoring. Complications include flap failure, infections, postoperative bleeding, acute lung injury (noncardiogenic pulmonary edema), and venous thromboembolism (VTE). In addition to comorbid conditions such as chronic obstructive pulmonary disease (COPD) and diabetes, these patients often have a history of alcohol dependence and smoking, hence close monitoring for encephalopathy and alcohol withdrawal syndrome is warranted.
Radical head and neck dissection involves en bloc removal of all nodal groups between the mandible and the clavicle with removal of the sternocleidomastoid muscle, internal jugular vein, and spinal accessory nerve.
The free-flap transfer, also called a free tissue transfer, is an autologous transplantation of vascularized tissues that incorporate a direct cutaneous artery and vein in the base. The free flap may contain skin, muscle, bone, or fascia. Free flaps have a higher complication rate than skin grafts. Indications for free flaps include complex defects of the head and neck regions following tumor resection or chemoradiation, reconstruction in patients failing local or regional flaps, or failure of a prior free flap. Common sites of harvest include the anterolateral thigh, fibula, and iliac crest. Fascio-cutaneous flaps are used to repair superficial lesions; and muscle and myocutaneous flaps are used to repair deeper lesions. To reconstruct mandible and floor of the mouth defects, fibular free flaps with overlying skin, or iliac crest bone flaps are usually used.
Diabetics are at increased risk of flap failure secondary to microangiopathy, increased risk of thrombosis and wound infection. Increasing age is an independent risk factor and smoking is associated with injury to flap vessels causing intimal fibrosis and potentially poorer outcomes.
Anatomical changes occur in head and neck cancers that may create a difficult airway. Close monitoring for postoperative airway compromise after head and neck surgery in the ICU is warranted due to complications including soft-tissue edema, bleeding, hematoma, and bilateral laryngeal nerve damage. Securing the airway in head and neck reconstructions may be difficult hence caution should be exercised before extubation. Postoperative soft-tissue edema peaks around the second or third postoperative day.
Once the patient tolerates a spontaneous breathing trial, a cuff-leak test should be considered before extubation. Two cuff-leak methods exist. The qualitative method is listening for air movement with a stethoscope after deflating the endotracheal tube balloon. In the quantitative method the endotracheal balloon is deflated while on volume control ventilation and the difference between the inspired versus expired tidal volume is measured. An at least 110 mL difference or a decrease of 12% to 24% in the expired versus inspired tidal volume supports adequate leakage around the tube and hence less airway edema.2 However, none of these tests are highly predictive of extubation success or failure and are only adjuncts to the overall clinical assessment.
In the event of respiratory distress or lack of an air-leak, extubation should be deferred for the next 24 hours. Inspection via laryngoscope may be performed to assess airway edema while the patient is still intubated. Steroids may be used in some cases for soft-tissue edema to facilitate extubation.
Intensivists should be prepared for emergent endotracheal intubation or an emergent surgical airway. Anesthesiologists and ENT surgeons should be present at bedside or in close proximity during extubation. Emergent airway equipment should remain at bedside with ENT surgeons on standby to perform emergent bedside tracheostomy if necessary.
In patients with extensive head and neck reconstructions a concomitant tracheostomy is usually done at the time of surgery to overcome airway compromise. Elective tracheostomy is recommended for patients with a high risk of airway obstruction. Multiple scoring systems for elective tracheostomy in head and neck cancer patients have been developed based on various parameters that include tumor size, tumor localization (involving larynx, base of the tongue, pharynx) TNM (tumor, nodes, metastases) staging, mandibulectomy, neck dissection, pathological chest X-ray findings, multiple comorbidities (mostly cardiopulmonary), and chronic alcohol dependence.3,4 However, there is no gold standard scoring system for determining elective tracheostomy and so clinical judgment is required.
Comorbid conditions such as underlying lung disease, cardiovascular disease, renal disease, alcohol abuse, and malnutrition are significant risk factors for failure to wean from mechanical ventilation.
Free-flap tissues can become ischemic and can be sensitive to ischemia/reperfusion injury with different responses to hypotension or vasopressors. Balanced volume resuscitation is very important in microsurgery, because volume overload is associated with flap edema and compromising flap microcirculation. Given the small caliber of blood vessels, hypotension should be avoided because even minimal vasoconstriction in the flap microcirculation leads to decreased blood flow. Low-molecular-weight (LMW) dextran (eg, Dextran 40) is used at some centers because of its rheological properties in reducing viscosity and providing antithrombotic effects thus potentially improving blood flow in the free flap. Increased risk of flap edema and acute lung injury/noncardiogenic pulmonary edema may occur more often in patients receiving higher volume crystalloid administration, for example, more than 130 mL/kg in 24 hours or more than 7 liters intraoperatively as reported in 1 study.5
Vasopressors have been used in clinical practice to increase the mean arterial pressure (MAP) and improve perfusion pressure across the free flap. In a recent study of patients undergoing free flaps receiving mainly phenylephrine or ephedrine, there was no significant relationship between intraoperative vasopressor and major free-flap complications.6 Prospective studies on the postoperative use of vasopressors have shown that norepinephrine and dobutamine can improve free-flap skin blood flow, with the greatest benefit from norepinephrine. Other pressors such as dopamine and epinephrine decreased flap flow.6,7,8 The use of albumin over synthetic colloids to replace intraoperative blood loss has not proven to be beneficial to date. However, the judicious use of pressors in an adequately volume resuscitated patient using norepinephrine is advised; there is no added benefit to use dobutamine which may lead to tachycardia, tachyarrhythmias, and hypotension.
Coronary ischemia and hemodynamic instability play an important role in flap outcomes. Long-term smoking and tobacco abuse not only increase the risk for cancer but also increase the risk of coronary artery disease (CAD) and myocardial infarction (MI). The risk of MI in the perioperative period is low, but associated with high mortality. High-risk patients with preexisting renal insufficiency, CAD, peripheral vascular disease (PVD), congestive heart failure (CHF), age of above 74, hypertension and prior chemoradiation are at increased risk for troponin elevation in the first 24 hours. Monitoring of troponin levels and the EKG is important in the immediate postoperative period. Elevated troponins post-ENT cancer surgery are associated with an 8 fold increased risk of death.9
Harvest sites need to be monitored frequently for signs of bleeding, hematoma, ischemia, or infection.
The reexploration rate after free-flap surgery due to circulatory compromise is about 5% to 25%. The success of the flap transfer is inversely proportional to its ischemic time and time to recognition of complications. Venous compromise occurs more often than arterial compromise and is usually seen in the first 2 or 3 postoperative days, while arterial compromise usually occurs later.10,11 Flaps complicated by postoperative hematomas require early evacuation to avoid vascular pedicle compression and flap engorgement. Good surgical technique, close flap monitoring, and early recognition of flap compromise are the keys to free-flap transfer success.
Data on the success rate of free-flap salvage within the first 1 to 4 hours following vascular compromise are close to 100%, but the success rate drops to 70% to 80% for ischemia times of more than 4 to 8 hours. If circulatory compromise is not addressed within 8 to 12 hours there may be development of no-reflow phenomenon12 where restoration of blood flow results in initial hyperemia with free-oxygen radicals and complement system activation followed reperfusion injury, a gradual decline in perfusion and cell death.
Flap color, temperature monitoring, color duplex, Doppler ultrasound, and pin-prick testing are a few of the monitoring modalities,13 but no ideal or gold standard exists. Clinical evaluation and color duplex Doppler has been proven to be more accurate in diagnosing vascular compromise of free-flap transfers.14 In the first 24 hours, patients are closely monitored in the ICU, kept nil per os (NPO) and are on strict bed rest to prevent potential complications in the event that reexploration is required. Flap checks are initially done half-hourly or hourly to assess flap viability. Avoidance of hypotension, hypoxia, and hypothermia are important. Venous congestion and necrosis in the skin paddle of a free flap might result in loss of the flap. Venous flow can be preserved by using systemic heparin therapy, LMW dextran or medicinal leech therapy.
Clinical observation of flap color, swelling, temperature, and capillary refill remains a vital mode of monitoring. A healthy and a well-perfused flap appears pink, minimally swollen, and warm to touch; a congested flap appears bluish in color, swollen, warm to touch with a short capillary refill time of less than 2 seconds. If the flap is ischemic it appears pale, cold to touch with a delayed capillary refill time of more than 3 seconds. Upon puncturing a viable free flap with a small gauge needle (pin-prick test) bright red blood flows immediately, whereas in a congested flap the blood flow is dark suggesting compromise. This test is usually performed on flaps where clinical monitoring by visualization is difficult. In pharyngoesophageal reconstruction using small intestine, the sentinel loop that is left outside the neck is used to monitor the viability. A healthy intestinal graft is pink and warm with minimal swelling and visible peristalsis; loss of peristalsis suggests graft compromise.
Compression of the artery will lead to loss of the Doppler signal within the flap, whereas venous compression will lead to augmentation of the venous signal due to increased venous return. Implantable Doppler probes may be used to monitor the flow in buried flaps. A decrease in tissue oxygen partial pressure (ptiO2) monitoring via an implantable oxygen-monitoring probe (eg, Licox) may suggest free-flap compromise. A pitfall of these monitors is that the probes are localized and do not assess viability outside the small sample volume that is monitored.
Microvascular Flap Anticoagulation
Aspirin and LMW heparin alone or in combination are commonly used for postoperative anticoagulation after free tissue transfer. Other strategies to prevent thrombosis include decreasing blood viscosity by LMW dextran infusions and/or hemodilution with IV fluids and avoiding blood-product transfusions unless absolutely necessary. LMW dextran antithrombotic properties are not fully understood but are thought to be secondary to platelet function impairment, destabilizing fibrin, and prolonging bleeding time. Studies report variable efficacy of these anticoagulants in preventing free-flap thrombosis.
Complications of LMW Dextran—Dextran has significant dose-dependent adverse affects. These include anaphylaxis, a rare but potentially life-threatening complication,15,16 noncardiogenic pulmonary edema, renal insufficiency, and rarely cardiac arrest.
Dextran infusion syndrome should be suspected in patients on LMW dextran that develop noncardiogenic pulmonary edema. They may also develop hypotension, bronchospasm, or coagulopathy. The incidence increases with higher infusion rates and the longer the infusion continues. The treatment is prompt discontinuation of the dextran infusion and initiating symptomatic treatment.
Hirudotherapy—Medicinal leech therapy, called hirudotherapy is commonly used in the setting of free-flap venous congestion or thrombosis to temporarily establish venous outflow until neovascularization of the graft is established. Leeches produce an anticoagulant called hirudin that is a selective thrombin inhibitor; it also secretes vasodilators and hyaluronidase that allow the leech to ingest blood. The regimen is to use about 2 to 4 leeches depending on the size of the flap, attached to the tissue for approximately 20 minutes. Following the leech therapy, blood oozes for the next 24 hours. Complications from this therapy include bleeding, anemia, leech migration to a different body location, and rarely local or systemic infection with the gram-negative bacillus Aeromonas hydrophila.17
There are no clear transfusion thresholds in free-flap surgery; however, it appears that a restrictive transfusion strategy using a postoperative transfusion trigger of hematocrit of less than 25 is not inferior to a higher transfusion threshold of hematocrit of less than 30 in patients undergoing free-flap surgery, with no increase in flap-related complications in the lower transfusion trigger group.18
Wound infection is not uncommon in tissue transfer surgery involving head and neck tumors. Risk factors for infection include preoperative radiotherapy, blood transfusion, smoking, and the duration of surgery.
The overall flap survival rate with alcohol withdrawal syndrome (AWS) is about 83%. Patients with postoperative AWS have a higher chance of developing non-flap-related complications, especially respiratory problems (prolonged ventilator dependency), encephalopathy, seizures, and hemodynamic instability. The onset of alcohol withdrawal symptoms is usually 1 to 2 days after abstinence (delirium tremens may occur at 2-5 days). Once AWS signs are recognized, prompt treatment should commence with an initial regimen of parenteral thiamine, multivitamins, and folic acid supplements. Effective treatment for AWS involves treatment with benzodiazepines and occasionally antipsychotics.19 As more data become available, dexmedetomidine appears promising in the treatment of AWS.