Introduction
Infrainguinal revascularization includes endarterectomy, bypass of the femoral artery or its branches, or both. Patients with peripheral vascular disease often have conditions associated with generalized vascular disease, such as diabetes, nicotine use, hypertension, or dyslipidemias. Some may have pre-existing endovascular stents at risk of perioperative thrombosis. Risk factors for or the presence of coronary artery disease has been associated with an increased risk of perioperative cardiac morbidity in numerous studies. Patients having infrainguinal revascularization surgery are at high risk of perioperative complications including graft failure, myocardial infarction, respiratory failure, and death. In a large cohort study, patients undergoing infrainguinal bypass had a 30-day mortality rate of 5.8% and a 1-year mortality rate of 16.3%. About half of all perioperative deaths in this population are caused by cardiac complications.
Neuraxial anesthesia has two primary postulated benefits for patients undergoing this surgery. First, patients may benefit with respect to outcomes related to concurrent diseases, for example, reduction in myocardial infarction rates or respiratory complications. Second, they may benefit from reduced complications related directly to their surgery, for example, a reduction in the rate of vascular graft failure that leads to infection, a second procedure, or even an amputation. Harm may also come to patients because of neuraxial anesthesia. The most obvious concern is about neurologic injury secondary to epidural or subdural hematoma, but another concern is about direct nerve root or spinal cord trauma. Evidence for and against these benefits and harms follows.
Therapeutic Options
Typical anesthetic options for patients having lower extremity vascular grafting include general anesthesia (GA), epidural anesthesia, spinal anesthesia, and combinations thereof. It is important to consider that clinical practices in any hospital or study may differ in basic choices that in turn may influence outcomes to a similar or perhaps greater degree than the variable studied. When studies designed to address anesthetic choice and infrainguinal revascularization outcomes are interpreted, the use of postoperative epidural infusion, invasive monitoring–guided hemodynamic optimization, and antithrombotic therapy are examples of “standardized” therapeutic choices that, in fact, vary between studies. Anesthesiologists must evaluate these choices in their own practices and clinical settings, as well as in the body of published evidence, to determine how best to serve their patients.
Evidence
Benefits
Mortality and Morbidity in Mixed Surgical Populations
Rodgers and colleagues performed a large meta-analysis of 141 randomized trials comparing neuraxial anesthesia with GA for all types of patients. Neuraxial anesthesia was associated with a significant (approximately 30%) reduction in the postoperative mortality rate. When odds of dying were examined by type of surgery, neuraxial blockade appeared salutary for orthopedic surgery more than for vascular, general, or urologic procedures. When odds of dying were examined by type of anesthesia, neuraxial blockade alone was superior to GA alone. Nonfatal operative morbidities including deep venous thrombosis, pulmonary embolism, perioperative transfusion, pneumonia, and respiratory depression were reduced for patients randomly assigned to neuraxial blockade. Myocardial infarction was possibly reduced (odds ratio [OR] 0.67; 95% confidence interval [CI], 0.45 to 1.00) in patients receiving neuraxial blockade.
The Multicentre Australian Study of Epidural Anesthesia (the MASTER Anesthesia Trial) included 888 patients with high-risk comorbidities undergoing major abdominal surgery or esophagectomy, randomly assigned to either GA with epidural anesthesia/analgesia or GA with postoperative intravenous opioids. Pain scores were lower at rest on the first postoperative day (POD) and with coughing on POD 1 to 3 in the epidural group. The respiratory failure rate was also reduced, but no significant differences in mortality rate or cardiovascular morbidity were demonstrated. The rate of death or at least one major complication was 57.1% in the epidural group and 60.7% in the GA group; to demonstrate a statistically significant 3.6% benefit of regional anesthesia/analgesia would require a study of roughly 6000 patients. Ultimately, it remains controversial whether a small but significant benefit of regional anesthesia exists for high-risk mixed surgical populations.
Bode and colleagues tested the hypothesis that regional anesthesia reduces operative cardiovascular morbidity and mortality rate associated with infrainguinal revascularization. A total of 423 patients were randomly assigned to receive general (138), epidural (149), or spinal (136) anesthesia for femoral-to-distal-artery bypass surgery. Epidural catheters were removed at the time of discharge from the postanesthesia care unit, but some patients received epidural morphine before catheter removal. All patients were monitored for at least 48 hours postoperatively with arterial lines and pulmonary artery catheters (but without standardized treatment protocol). Patients received subcutaneous heparin on POD 1 until ambulation, then 81 mg aspirin daily thereafter. There was no significant reduction of myocardial infarction, angina, congestive heart failure, or all-cause mortality rates between GA (16.7%), epidural (15.4%), or spinal anesthesia (21.3%). Because of the study design, the potential benefit of postoperative epidural infusion was not addressed. In sum, current evidence for significant reduction of mortality rate and cardiac risk by use of regional anesthesia during infrainguinal revascularization is limited. If favorable, the benefit of regional anesthesia is small.
Graft Failure in Lower Extremity Revascularization
In two randomized studies, one of which (Christopherson and colleagues ) compared epidural with GA for patients having lower extremity grafts and the other of which (Tuman and colleagues ) compared epidural-supplemented with unsupplemented GA for patients having either aortic or lower extremity vascular surgery, vascular graft failure was reduced in patients with epidurals. Both these studies reported high rates of vascular graft failure, and both of them continued epidural analgesia into the postoperative period. In the study by Christopherson and colleagues, preoperative aspirin was withheld and heparin was continued into the postoperative period only when there was suspicion of graft failure. Few patients in that study were monitored with pulmonary artery catheters. In the study by Tuman and colleagues, intraoperative heparin was reversed with protamine at the end of surgery. High rates of graft failure in these two studies might have been reduced had different antithrombotic strategies been used. However, high rates of adverse outcomes made it possible for these two studies to show a significant reduction of graft failure in patients who received epidural anesthesia.
A focused retrospective chart review by Kashyap and colleagues also showed a possible benefit to regional anesthesia. This review examined graft survival after infrapopliteal revascularization with polytetrafluoroethylene graft material for critical ischemia. These criteria narrowed the results to 77 patients from 1500 lower extremity revascularization surgeries over the period of 1978-1998 and functionally selected for a study population with a high rate of graft failure, thus strengthening the ability to detect a small effect. GA accounted for 75% of these cases and regional anesthesia, mostly spinal anesthetics, accounted for 25% of the cases. There were 11 incidents of acute graft thrombosis, all in the GA group. The regional group had prolonged primary graft patency at 36 months (35%) when compared with the GA group (15%). The specific breakdown of which patients had neuraxial analgesia continuing into the postoperative period was not reported. Postoperative warfarin use was not statistically associated with an improvement in graft patency, but only some of the patients received warfarin in this retrospective, nonrandomized study.
A large chart review study using data from the Veterans Affairs National Surgical Quality Improvement Program (NSQIP) was done by Singh et al. Patients undergoing infrainguinal vascular bypass surgery during the period from 1995 to 2003 were identified by Current Procedural Terminology (CPT) codes and their charts retrospectively reviewed for type of anesthetic and its effect on 30-day graft failure, cardiac events, pneumonia, length of stay, and surgery-related return to the operating room. A total of 14,788 patients were identified: 9757 (66%) received general endotracheal anesthesia (GETA), 2848 (19%) were administered a spinal anesthetic (SA), and 2183 (12%) had an epidural anesthetic (EA). The study showed the odds of graft failure were 43% higher with GETA versus SA and EA, which represented a 40% increase in the need to return to the operating room versus SA and a 17% increase versus EA The study also showed a significantly greater number of cardiac events and double the rate of postoperative pneumonia within 30 days of the procedure. However, the inherent limitations of a nonrandomized, retrospective study apply. Differences in the specifics of operative complexity (e.g., redo surgery, spliced or arm vein, longer operative times, and urgency of surgery) were not reliably captured in the database and may have been associated with bias in the selection of the type of anesthesia. Of note, the authors projected that a well-controlled randomized study would require more than 20,000 patients to demonstrate a statistically significant outcome effect of anesthetic choice on the rate of graft failure using the rate of failure found in this chart review.
In contrast to the already mentioned studies, a retrospective chart review by Schunn and colleagues examined 294 primary femoral–popliteal–tibial bypass surgeries occurring between 1989 and 1994 and found no significant difference of early graft thrombosis rates between GA alone (9.4%) and epidural alone (14%). It is unclear whether epidural analgesia was always continued into the postoperative period or continued selectively in certain cases, and, as a chart review, there was no randomization between the two groups. In two prospective randomized trials, one study of 101 patients comparing spinal to GA (Cook and colleagues ) and one of 264 patients (Pierce and colleagues ) in which patients were randomly assigned to SA, EA, or GA but without neuraxial analgesia in the postoperative period, there was no graft patency benefit associated with regional anesthesia. Rates of graft failure were very low overall; in fact, rates were so low in the study by Pierce and colleagues that the study was underpowered to find a difference in rates of graft failure. In the study by Pierce and colleagues no difference was found in the rate of postoperative amputation. All patients received aspirin and either subcutaneous heparin or oral warfarin. Additionally, all patients were monitored with arterial lines and pulmonary artery catheters for 24 to 48 hours after surgery. It has been shown that patients undergoing lower extremity vascular surgery under GA had improved vascular graft survival if they were monitored and treated appropriately with the use of pulmonary artery catheters.
As with most complex questions, interpretation of available research is equally complex and presents a number of contradictions. As such, it is important to carefully weigh the quality of each study. In this context, it is evident that the best designed studies—those using adequate blinding and randomization—show little outcome differences among the choices of anesthetics but are limited by small sample size. It will continue to be difficult to be guided by the available literature in choice of anesthetic techniques. These decisions will need to continue being made on a patient-by-patient and practice-by-practice basis.
Risks
The rapid evolution of antiplatelet and anticoagulant therapies may have a greater effect on outcome than anesthetic choice. Furthermore, these therapies affect anesthetic choice because of an elevated risk of neuraxial bleeding that may be associated with SA or EA techniques. Antithrombosis therapy is important in the maintenance of vascular graft patency. In some institutions aspirin is routinely given before surgery. Intravenous heparin is almost always given intraoperatively before clamping of the arteries to be grafted. Thus a spinal or an epidural needle might be placed into a patient whose platelet function is impaired from aspirin, and subsequent to placing of an epidural catheter, an anticoagulant is almost always given. Furthermore, intravenous heparin may be continued into the postoperative period, or low-molecular-weight heparin (LMWH) may be given subcutaneously. Because of concomitant diseases, vascular surgery patients may be taking warfarin or antiplatelet therapy.
The American Society of Regional Anesthesia and Pain Medicine (ASRA) has recently reviewed the evidence of risk of epidural hematoma for patients receiving neuraxial blockade while undergoing anticoagulation. Pertinent recommendations related to heparin and antiplatelet agents are summarized as follows. For more details or for evidence-based management of neuraxial anesthesia for patients taking other anticoagulants, the reader is referred to the ASRA consensus document, available at www.asra.com (accessed June 11, 2012).
- 1.
Unfractionated heparin: patients undergoing vascular surgery who will receive heparin intraoperatively should not receive neuraxial anesthesia if they have other coagulopathies. If there is difficult or bloody needle placement, they may be at increased risk of neuraxial hematoma; there should be a discussion with the surgeon as to whether the case should proceed or be canceled. In general, heparin should not be given until at least 1 hour after needle placement. In the postoperative period, there should be careful monitoring of neurologic status, and concentrations of local anesthetics should be limited to those that allow assessment of motor strength. Epidural catheters should be removed at least 2 to 4 hours after a heparin dose. Patients receiving heparin for 4 days or longer are at risk of heparin-induced thrombocytopenia; therefore a platelet count should be obtained before neuraxial block is performed.
- 2.
LMWH: patients receiving preoperative LMWH should be assumed to have impaired coagulation. The safest timing and type of anesthesia is likely a single-injection SA given at least 10 to 12 hours after the last thromboprophylaxis-dosed LMWH; patients receiving higher (treatment) doses of LMWH should not receive neuraxial anesthesia for at least 24 hours. If LMWH is to be started postoperatively, dosing and epidural catheter removal must be timed. Additional care and consideration of the risk and benefits of regional techniques should be considered when the patient is being treated with other drugs that may act synergistically with LMWH.
- 3.
Antiplatelet medications: nonsteroidal antiinflammatory drug therapy alone is not a contraindication to a regional technique. Before neuraxial regional anesthesia, an interval of 14 days is suggested for ticlopidine and 7 days for clopidogrel. The family of platelet glycoprotein (GP) IIb/IIIa inhibitors deserves special mention. Platelet aggregation is impaired for 24 to 48 hours after administration of abciximab and for 4 to 8 hours after eptifibatide and tirofiban.
Related posts:
Is a Preoperative Screening Clinic Cost-Effective?
Which Patient Should Have a Preoperative Cardiac Evaluation (Stress Test)?
What Is the Optimal Airway Management in Patients Undergoing Gastrointestinal Endoscopy?
What Works in a Patient with Acute Respiratory Distress Syndrome?
Which Are the Best Techniques for Reducing the Incidence of Postoperative Deep Vein Thrombosis?
Aspiration: Is There an Optimal Management Strategy?
Full access? Get Clinical Tree
