August 21, 2001

Reduction of postoperative mortality and morbidity with epidural or spinal anesthesia: results from an overview of randomized trials

Anthony Rodgers, Natalie Walker, A Mckee, H Kehlet, A van Zundert, D Sage, M Futter, G Saville, T. Clark, S MacMahon. British Medical Journal 2000;324 (1), 193-197

An issue that remains controversial in clinical anesthesiology is whether neuraxial blockade, that is, epidural or spinal anesthesia, is associated with less perioperative mortality than general anesthesia (inhalational or intravenous anesthetics) in patients undergoing major surgical procedures. Neuraxial blockade has several physiological effects that may be expected to improve postoperative outcome. These include reduction in sympathetic and neuroendocrine responses to tissue trauma, improved venous blood flow, and the initiation of pre-emptive analgesia. Although solid evidence exists that neuraxial blockade reduces the incidence of deep venous thrombosis and one month mortality in patients recovering from hip fracture, insufficient data existed for patients recovering from other procedures. In an effort to more reliably estimate the benefits of neuraxial blockade on overall post surgical morbidity/mortality, the authors conducted an extensive and systematic review of all relevant clinical trials. Several electronic databases were searched, including Current Contents (1995-1996), Medline (1966-1996), and the Cochrane Library (1998) for all controlled trials in which patients were randomized to receive either neuraxial (epidural or spinal anesthesia) or general anesthesia for major operative procedures.

One hundred forty-one randomized trials that included 9,559 patients were identified. A total of 247 deaths within 30 days following surgery were recorded in 35 trials that tracked short-term outcome. Overall mortality was approximately one-third lower in patients who received neuraxial blockade. The observed improvements in survival were primarily related to trends towards reductions in pulmonary emboli, cardiac events, stroke, and pulmonary complications. Reductions in mortality did not differ by surgical group, type of block (spinal or epidural), or when general anesthesia was augmented by neuraxial block.

With regard to vascular morbidity, a total of 365 deep venous thromboses were reported in 18 trials. Neuraxial blockade reduced the risk of DVT by 50%. A total of 96 pulmonary emboli were reported, of which 21 were fatal. Again, the incidence of pulmonary emboli was reduced by 50% in the neuraxial blockade groups. A total of 104 myocardial infarctions were reported in 30 trials, and the incidence was 33% lower in the neuraxial blockade groups.

With regard to perioperative bleeding, 473 patients from 16 clinical trials documenting blood loss required 2 or more unit’s transfusion. The number of patients requiring 2 or more units was reduced 50% in the neuraxial blockade group. It should be pointed out however that many of these trials were performed at a time when the benefits of postoperative anticoagulation were not widely recognized, and low dose heparinoids and warfarin were not routinely administered

With regard to perioperative infection rate, 387 cases of postoperative pneumonia were reported in 28 clinical trials, and 62 wound infections were reported in 11 trials. The number of patients suffering either form of infection was significantly lower in the neuraxial blockade group.

This overview provided compelling evidence that neuraxial blockade reduces perioperative mortality and major postoperative complications in a wide range of patients and surgical procedures. These improvements in outcome might reflect inherent benefits associated with neuraxial blockade (suppression of intraoperative stress responses, improved blood flow, normalization of coagulation, and preemptive analgesic effects) or may be due to the avoidance of adverse effects associated with general anesthetics. Based on subgroup analyses (and evaluation of groups treated with both neuraxial block plus general anesthesia), the authors conclude that improvements in perioperative outcome are gained primarily from protective benefits associated with neuraxial blockade.

One drawback associated with this investigation is that the authors could not differentiate whether intra-operative benefits, or post surgical improvements in pain control or perhaps both factors were responsible for reductions in morbidity and mortality in the neuraxial group. Neither could they separate benefits provided by neuraxial local anesthetics or opioids, administered alone or in combination with each other. Anesthesiologists often place epidural catheters for post-surgical pain management, while employing a general anesthetic for the surgical procedure. Intraoperative neuraxial blockade is either avoided or restricted because of the fear of sympathectomy and additive cardiovascular depression. These patients also benefit from superior post surgical pain control and suppression of stress responses, which in several studies has resulted in superior post-surgical outcome.

Reduction of postoperative mortality and morbidity with epidural or spinal anesthesia: results from an overview of randomized trials

AUTHORS:
Anthony Rodgers, Natalie Walker, A Mckee, H Kehlet, A van Zundert, D Sage, M Futter, G Saville, T. Clark, S MacMahon

SOURCE:
British Medical Journal 2000;324 (1), 193-197

ABSTRACT:
No abstract available