The incidence of serious, permanent neurological complications is exceedingly rare and data comprises a handful of case reports. The incidence of spinal hematoma has classically been thought to be 1:150,000 in the presence of normal coagulation, though recent studies have suggested it may be more common. The incidence of epidural hematoma formation in a patient who is anticoagulated is not known, but it is estimated at 1:3,000 in patients receiving therapeutic low molecular weight heparin. A full discussion of epidurals and anticoagulation is beyond the scope of this chapter; the most recent guidelines can be found on the American Society of Regional Anesthesia web site (www.asra.com). Epidural abscess incidence is not known but considered rare (at most 0.05%). Risk factors for abscess formation may be longer times of having the epidural in place (possibly >6 days) as well as use in immunocompromised patients. Intrathecal or intravascular migration ofan epidural catheter is estimated at approximately 0.2%, though it is likely much less frequent than this. The incidence of premature catheter dislodgement is estimated at about 6%.
The major side effects due to local anesthetics in epidurals are hypotension, CNS toxicity, and motor block. The probability of hypotension from an epidural does not rise above 1-2% until the "spread" of the epidural is beyond 14 sympathetic dermatomes. Remember that the sympathetic blockade may be another possible six dermatomes beyond the amount of sensory block. The major mechanism by which an epidural might cause hypotension is through decreased venous return (pre-load) from reduced venous capacitance, therefore the most appropriate treatment is increasing pre-load with volume. Hence, hypotension is rarely seen in the supine normovolemic patient. Thus it is often the fluid status of the patient that is more of a causative factor than the epidural itself. This is evidenced by the fact that stopping the epidural infusion in a patient often does not improve blood pressure unless the pain becomes so great that a large sympathetic response occurs.
CNS toxicity (seizures) from systemic accumulation of local anesthetics is obviously based on the total amount of the drug used overtime and has an estimated incidence of 0.01-0.1%. Motor block of the lower extremities is estimated at less than 3% when lower concentrations of local anesthetic are used and upward of 25% with higher concentrations. In addition, the level of epidural insertion has much to do with the actual incidence in that lumbar epidurals are more likely to cause motor block than thoracic.
The incidence of nausea/vomiting with opioid-containing PCEA therapy is estimated at between 4 and 15%, though higher incidences have also been reported. The incidence of severe pruritus is between 2 and 17% with the incidence of any degree of pruritus likely being well over 50%. Sedation is reported to occur in about 15% of patients. Respiratory depression is estimated to occur in 0.1-0.4% of patients. However, a higher incidence (approaching 1.5%) of respiratory depression is seen in some studies using morphine.
The respiratory depression seen with epidural opioids is biphasic. Early respiratory depression (within an hour of epidural initiation) is thought to be due to systemic absorption of the opioids via epidural veins and is thus more likely with the lipophilic opioids. Delayed respiratory depression (occurring 4-8 h after epidural initiation) is seen especially when using the hydrophilic opioids and is thought to be due to the cephalad spread of the opioids in the cerebrospinal fluid (Wheatley 2001).
Was this article helpful?