Principles underlying complications and errors

A complication is an undesirable, unexpected event occurring in the course of an intervention. It is necessary to differentiate between such events and the side-effects one can normally expect to encounter in clinical practice. Side-effects, in general, are predictable occurrences, and their prompt recognition and treatment can avoid more serious sequelae. Complications, however, may occur as a result of human factors on the part of the anesthesiologist, or may be attributable to environmental or equipment factors, or may occur secondary to ‘system’ factors.

Human factors can be defined as lapses in, or lack of, safe habit, or the occurrence of a vigilance decrement resulting from sleep deprivation, fatigue, recent alcohol or drug ingestion, or boredom. Inexperience on the part of the anesthesiologist is likely to contribute to poor decision-making or an error in judgment. An important component in the avoidance of complications arising from these factors is awareness of anesthesiologists as individuals and of their role in complication development. They should thus seek to establish safe working practices and individual self-discipline that may act to counterbalance these risks.^1–3

Environmental factors leading to complications may include lack of appropriate patient-monitoring systems and protocols, inadequate drug identification systems, or pressures originating from practice managers, surgeons, or financial concerns.

Patient selection and management are dealt with briefly in this chapter. Selection of an anesthetic technique that fits the patient, surgeon, and anesthesiologist at a particular point in time will form the basis of a successful intervention.

Systemic toxicity of local anesthetic drugs

Toxic reactions following local anesthetic drug administration can involve the CNS and/or the cardiovascular system. CNS toxicity is more common, occurs in association with lesser plasma drug concentrations, and responds more readily to treatment.

Central nervous system toxicity

The signs and symptoms of local anesthetic-induced CNS toxicity are shown in Figure 5.1. An initial phase of CNS excitability, as demonstrated by light-headedness, dizziness, visual and auditory disturbance, muscle twitching, and convulsions, is followed by CNS depression, with coma, then respiratory depression and arrest. This sequence of events occurs because of an initial inhibition, at lesser concentrations, of inhibitory pathways in the amygdala. At greater concentrations, both inhibitory and excitatory pathways are inhibited, resulting in generalized CNS depression.^5–8

Figure 5.1 Relations of signs and symptoms of local anesthetic toxicity to plasma concentrations of lidocaine.

(From Ref. 4, Covino BG, Wildsmith JAW. Clinical pharmacology of local anesthetic agents. In: Cousins MJ, Bridenbaugh PO (eds). Neural blockade in clinical anesthesia and management of pain, 3rd edn. Philadelphia: © Lippincott-Raven; 1998.)

The toxic potential of each anesthetic drug is related to its potency as an anesthetic agent (Table 5.1),⁹ and the rate at which it is injected or absorbed (Fig. 5.2). Hypercapnia and acidosis lower the convulsive threshold for local anesthetic drugs. This occurs in a number of ways. A high pCO2 will increase cerebral blood flow, resulting in greater rates of drug delivery; decreased intracellular pH facilitates the formation of the cationic form of drug, i.e. the active form; and hypercapnia and acidosis result in diminished protein-binding of drug, thereby making available a greater proportion of free drug.¹⁰

Figure 5.2 Arterial plasma concentrations following intravenous injection of 100 mg of lidocaine hydrochloride over 0.1 and 2 min to simulate concentrations of an inadvertant intravenous injection during a block procedure. Prolonging injection time reduces peak concentrations.

(From Ref. 4, Covino BG, Wildsmith JAW. Clinical pharmacology of local anesthetic agents. In: Cousins MJ, Bridenbaugh PO (eds). Neural blockade in clinical anesthesia and management of pain, 3rd edn. Philadelphia, © Lippincott-Raven; 1998.)

Local anesthetic-induced seizures are effectively terminated by administration of barbiturate or benzodiazepine drugs.^11,12 The doses required are small and one should remain mindful that their myocardial depressant effects are additive to those of local anesthetic drugs.

Cardiovascular system toxicity

The depolarization phase of the action potential in cardiac tissue differs from nerve tissue in that the fast influx of Na⁺ is followed by a slow influx of Ca²⁺. This influx of Ca²⁺ is responsible for the spontaneous depolarization that is characteristic of cardiac tissue (Fig. 5.3, Table 5.2). Local anesthetic drugs depress the maximal depolarization rate of the cardiac action potential, Vmax, secondary to inhibition of Na⁺ conductance. With increasing concentrations of local anesthetics, prolongation of conduction times occurs, producing an increase in the P–R interval and QRS duration. At greater concentrations this is followed by sinus bradycardia, sinus arrest, and atrioventricular dissociation.^14,15 Local anesthetics also profoundly depress cardiac contractility, a phenomenon that may be related to the displacement of Ca²⁺ from the sarcolemma.^16–18

Figure 5.3 Cardiac action potential recorded from a ventricular contractile cell (A) or atrial pacemaker cell. (B) TP, threshold potential.

(From Ref. 13, Stoelting RK. Heart. In: Pharmacology and physiology in anesthetic practice. 2nd ed. Philadelphia, © JB Lippincott; 1991.)

Table 5.2 Ion movement during phases of the cardiac action potential

(From Ref. 13, Stoelting RK. Heart. In: Pharmacology and physiology in anesthetic practice. 2nd ed. Philadelphia: © JB Lippincott; 1991.)

The CC/CNS ratio is that of the dosage required for cardiovascular collapse (CC) to the dosage required to produce convulsions. It is approximately 7.1 for lidocaine and 3.7 for bupivacaine, suggesting a greater margin of safety in the use of lidocaine.¹⁹ The high lipid-solubility of bupivacaine results in a slow rate of dissociation from the tissues, and thus a persistent effect on Vmax. Cardiovascular collapse resulting from bupivacaine is therefore resistant to treatment. The potential for cardiac toxicity is enhanced in pregnancy, for reasons not fully understood, and also in the presence of hypoxia and hypercapnia.^20–22 These factors enhance the toxic potential of bupivacaine to a greater degree than they do lidocaine.

Treatment of Local Anesthetic Systemic Toxicity

Following the demonstration that lipid emulsion could reverse local anesthetic systemic toxicity in rat and canine models, and the publication of case reports demonstrating similar effects in humans, ASRA has published a practice advisory outlining the evidence and providing guidance.

Peripheral vasculature

Local anesthetic drugs have a biphasic action on vascular smooth muscle.²³ At low concentrations they produce vasoconstriction. As the concentration increases, the effect becomes one of vasodilatation. These observations have been explained as being due to stimulation of spontaneous myogenic activity at low concentrations and inhibition of the same at greater concentrations.

Following an inadvertent intravascular injection of an amide local anesthetic, should the plasma concentration reach levels sufficient to produce CNS toxicity, one may also observe an increase in blood pressure, heart rate, and cardiac output. As the plasma concentration increases, reversible cardiovascular depression ensues, associated with a decrease in cardiac output and systemic blood pressure. Finally, myocardial contractility becomes profoundly depressed, marked peripheral vasodilatation occurs, and cardiac arrest ensues.

Nerve injury

Nerve stimulation is one effective technique for locating a peripheral nerve. Prospective studies have demonstrated that a paresthesia technique can significantly increase the risk of postblock neuropathies (2.8%), while the transarterial approach to the brachial plexus is associated with paresthesia in as many as 40% of cases,^24,25 producing neuropathy in 0.8% (Table 5.3). In contrast, a nerve stimulation technique aims to avoid nerve contact and has been shown to produce important block-related neuropathies in only 0–0.3% of cases.²⁷

The risk of penetrating a nerve fascicle is reduced when a short-bevel (45°) needle is used, compared with a standard long-bevel (15°) needle, the reason being that nerve fascicles tend to roll away more readily from the advancing short-bevel needle tip.²⁸ Although the incidence of injury is less with short-bevel needles, when injury does occur it is more severe.

Intraneural needle position is associated with painful paresthesias on injection, and intraneural injection causes nerve damage and cell death by mechanical disruption, disruption of the blood–nerve barrier, high endoneural pressure (above capillary perfusion pressure) (Fig. 5.4), and direct neurotoxicity of local anesthetic agent. This situation is further aggravated if the solution contains epinephrine.^29,30 Therefore it is important to maintain verbal contact with the patient, avoid paresthesias, administer small incremental doses of drug, and reposition the needle if paresthesias are elicited.

Figure 5.4 Recordings of intraneural pressure during and after injection of 100 µL in the sciatic nerve of a rabbit using an injection pump. Note the slow pressure decrease after intrafascicular injection. Green line, intrafascicular injection; blue line, epineural injection; arrows, start and end of injection; orange line, estimated endoneural capillary perfusion pressure.

(From Ref. 26, Selander D. Peripheral nerve injury after regional anesthesia. In: Finucane, BT (ed). Complications of regional anesthesia. Philadelphia, Churchill Livingstone, 1999, with permission from the American Society of Regional Anesthesia and Pain Medicine.)

In attempting to establish the etiology of nerve lesions in the postoperative period, the differential diagnosis must initially take into account patient positioning, tourniquet use, surgical trauma, and the presence of tight casts or dressings.^31–33 Follow-up of the patient in the immediate postoperative period will help to avoid inaccurate labeling of the deficit as ‘anesthesia-related’.

Allergic reactions

Allergic reactions to local anesthetics occur rarely.³⁴ Indeed, most ‘allergic reactions’ to local anesthetics are in fact adverse reactions. Nevertheless, para-aminobenzoic acid is a product of the hydrolysis of ester local anesthetics and is a known allergen.^35–37 Allergy to amide local anesthetics is still rarer. However, some preparations contain methylparaben (an allergen), because of its excellent bacteriostatic and fungistatic properties.³⁸ After a case of allergy to a local anesthetic agent, intradermal testing of the full range of anesthetic agents is worthwhile, because allergy to one agent does not necessarily imply allergy to another.^37,39

Infection

The presence of infection at the site of puncture is generally accepted as being a contraindication to regional anesthesia. The paucity of reports detailing infective complications of peripheral nerve block suggests that local and generalized infections following nerve blocks are rare. Disastrous infective complications continue to be reported following central neuraxial block, however, and the increasing use of peripheral nerve catheters suggests some elementary precautions be taken in this regard.⁴⁰ Unfortunately, no recommendations exist as to aseptic technique for spinal, epidural, or peripheral nerve block.⁴¹ A review of the literature serves to highlight the following points:⁴⁰

Neural toxicity of local anesthetics

All clinically used local anesthetic agents are potentially toxic at high concentrations. Under normal conditions, the drug is rapidly diluted and absorbed. However, if the nerve is ischemic or the drug is injected intraneurally, the nerve is exposed to greater than normal concentrations of local anesthetic and for a longer than expected period of time. This situation is exacerbated with epinephrine-containing solutions.⁵⁵ Lidocaine was shown to have a greater neurotoxic potential in this regard than the other clinically used agents.

Myotoxicity of local anesthetics

Injection of local anesthetic into muscle results in focal necrosis; the more potent the agent, the greater the degree of injury that results. This effect is localized and regeneration has been shown to be complete within 2 weeks. The changes are of a subclinical nature and do not appear to contribute to the peri-operative morbidity of regional anesthesia.⁵⁶ Some concern has been raised, however, regarding the role of local anesthetic drugs in the development of diplopia following cataract surgery performed under regional anesthesia. A 0.25% incidence of diplopia related to anesthetic factors has been reported.⁵⁷ It appears to be more common following peribulbar block than retrobulbar block and does not occur following topical or general anesthesia. The inferior rectus muscle is typically involved following infraorbital injection. Possible mechanisms underlying this complication are direct muscle injury from the block needle, vascular compromise secondary to elevated local pressures, and myotoxicity of the local anesthetic.

Patient selection

Appropriate patient selection for a regional anesthetic technique involves consideration of patient factors, the medical history, specific investigations, psychological preparation, the planned surgical intervention, and the expertise of the anesthesiologist.

When general anesthesia poses serious risks – for example in the patient with a full stomach, difficult airway, or poor general medical condition – regional anesthesia is quite often the anesthetic technique of choice. However, there are various medical conditions where the choice of anesthetic technique remains controversial. These include degenerative neurologic disease, diabetes, and severe cardiovascular and respiratory disease. These conditions are dealt with in more detail in Chapter 4. As for all patients, a thorough pre-anesthetic medical and laboratory evaluation is indicated in order to perform an appropriate risk–benefit analysis.

Morbid obesity, physical deformities, arthritis, fractures, local infection, or locally enlarged lymph glands may serve to hinder the administration of an adequate block and so should be noted prior to the formulation of an anesthetic plan. The disoriented or psychologically deranged patient may not only make intra-operative management difficult, but render the safe performance of a block impossible. A history of concurrent medication, such as anticoagulants (Box 5.1) or vasoactive drugs, should also be taken into account because of the implications for the performance and administration of regional anesthesia (see Ch. 4).

The evaluation of the patient must also include systemic disease, current medications, previous anesthetics, allergies, state of dentition, and family history. Occasionally it may be necessary to convert a regional to a general anesthetic technique. Therefore all the information relevant to provision of general anesthesia should be obtained.

The patient interview allows one to obtain the relevant information outlined above; it is also an opportunity to prepare the patient psychologically for the procedure and the peri-operative experience in general and to obtain informed consent. Informed consent implies that the material risks and benefits of the proposed procedure have been explained, as well as those of the available alternatives. When one explains the reasons for choosing one technique over another, and when appropriate assurances as to the standard of care are given, most patients will accept regional anesthesia. Assurances as to the availability of block supplementation and sedation up to the level of general anesthesia should be given, as well as a detailed description of the performance of the block(s). Patients may perceive sensations of an unusual or bizarre nature under regional anesthesia; therefore, explanation, reassurance, and appropriate sedation should be provided. Most patients who have benefited from a well-executed regional anesthetic will choose the same technique for future interventions when possible.⁵⁹

Pre-operative fasting

Traditionally, adult patients scheduled for a surgical intervention were required to abstain from both oral solids and fluids for a minimum of 6h, and, more often, from midnight the night before. However, unrestricted clear fluids (water and apple juice) up to 2 h before surgery have been shown to result in no significant difference in mean residual gastric volume or pH.⁶⁰ As a consequence, a number of organizations have amended their fasting guidelines (Table 5.4).⁶¹

Table 5.4 Example of fasting protocol for sedation and analgesia for elective procedures*

* Gastric emptying may be influenced by many factors including anxiety, pain, abnormal autonomic function (e.g. diabetes), pregnancy, and mechanical obstruction. Therefore the suggestions above do not guarantee that complete gastric emptying has occurred. Unless contraindicated, children should be offered clear liquids until 2–3 h before sedation to minimize the risk of dehydration.

^† This includes milk, formula, and breast milk. (High fat content may delay gastric emptying.)

^‡ There are no data to establish whether a 6–8-h fast is equivalent to an overnight fast prior to sedation or analgesia.

(From the American Society of Anesthesiologists,⁶¹ with permission of ASA.)

Equipment

In 1986, the Department of Anesthesia of Harvard Medical School published detailed, mandatory standards for minimal patient monitoring during anesthesia.⁶² For the safe conduct of regional anesthesia, in addition to the presence of an anesthesiologist or nurse anesthetist, the following equipment should be available and used:

The equipment included in Box 5.2 should also be available for immediate use, and a checklist performed prior to anesthesia. In addition, the nerve stimulator to be used should be checked according to the manufacturer’s instructions prior to each use.