Prevalence and Epidemiology

Cocaine abuse and addiction continue to be a problem that plagues the United States and many other countries. Data from the U.S. Drug Abuse Warning Network (DAWN) showed that cocaine accounted for 43% of the 2.1 million drug abuse emergency department visits that occurred during 2009. The National Survey on Drug Use and Health (NSDUH) estimates that 5 million Americans are regular users of cocaine, 6000 use the drug for the first time each day, and more than 30 million have tried cocaine at least once. On the basis of these data, practicing anesthesiologists will likely come across cocaine-abusing patients, regardless of the setting of their practices.

The classic profile of patients reported to experience cocaine-related myocardial ischemia is typically a young, nonwhite, male cigarette smoker with no other significant risk factors for atherosclerosis. However, this profile no longer holds true as the problem becomes more severe and is not confined to a particular race or gender. Cocaine abuse in parturients has been the focus of attention lately, and the reported incidence is between 11.8% and 20%.

Pharmacokinetics and Mechanism of Action

Cocaine produces prolonged adrenergic stimulation by blocking the presynaptic uptake of sympathomimetic neurotransmitters, including norepinephrine, serotonin, and dopamine. The euphoric effect of cocaine, the cocaine high, results from prolongation of dopamine activity in the limbic system and the cerebral cortex. Cocaine can be taken orally, intravenously, or intranasally. Smoking the free base (street name for the alkalinized form of cocaine) results in very effective transmucosal absorption and a high plasma concentration of cocaine. It is metabolized by plasma and liver cholinesterase to water-soluble metabolites (primarily benzoylecgonine and ecgonine methyl ester [EME]), which are excreted in urine. The serum half-life of cocaine is 45 to 90 minutes; only 1% of the parent drug can be recovered in the urine after it is ingested. Thus cocaine can be detected in blood or urine only several hours after its use. However, its metabolites can be detected in urine for up to 72 hours after ingestion, which provides a useful indicator for recent use. Hair analysis can detect use of cocaine in the preceding weeks or months. Table 7-1 summarizes the pharmacokinetics of cocaine with different routes of administration.

Cardiovascular Effects

Cardiovascular effects of cocaine are largely due to the sympathetic stimulation resulting from inhibition of the peripheral uptake of norepinephrine and other sympathomimetic neurotransmitters. Central sympathetic stimulation has been suggested as an alternative mechanism to explain the exaggerated sympathetic response. The resulting hypertension, tachycardia, and coronary artery vasospasm are responsible for the myocardial ischemia seen with cocaine toxicity. In addition, there is evidence that cocaine activates platelets, increases platelet aggregation, and promotes thrombus formation. Knowledge of the mechanism of myocardial ischemia in patients with cocaine abuse is key for effective treatment. Classically, beta-blockers are avoided because their use may lead to unopposed alpha-mediated coronary vasoconstriction. This concept has been recently challenged, and there is some evidence to support the use of beta-blockers in cocaine-related myocardial ischemia. Esmolol is used for treatment of cocaine-induced myocardial ischemia because of its short duration of action and the ability to titrate the dose to a target heart rate. Labetalol offers some advantage in that regard because of its combined alpha- and beta-receptor blocking effect. Alpha-blockers and nitroglycerin have been used effectively for symptomatic treatment.

A major concern in the anesthetic management of the cocaine-abusing patient is the occurrence of cardiac arrhythmias. These include ventricular tachycardia, frequent premature ventricular contractions, or torsades de pointes. Myocardial ischemia has been suggested as the underlying mechanism for these arrhythmias ; however, cocaine-induced sodium and potassium channel blockade is currently believed to be more important. This cation channel blockade results in QRS and QTc prolongation, which is considered to be the primary mechanism for induction of these cocaine-induced arrhythmias.

Aortic dissection and ruptured aortic aneurysm have been reported with short-term abuse. Peripheral vasoconstriction may mask the picture of hypovolemia in the setting of acute cocaine toxicity.

Long-term use of cocaine can cause left ventricular hypertrophy, systolic dysfunction, and dilated cardiomyopathy. Repetitive cocaine administration is associated with the development of early and progressive tolerance to systemic, left ventricular, and coronary vascular effects of cocaine. The mechanism of tolerance involves neither impaired myocardial nor coronary vascular responsiveness to adrenergic stimulation but rather attenuated catecholamine responses to repetitive cocaine administration.

Pulmonary Effects

Approximately 25% of individuals who smoke crack cocaine develop nonspecific respiratory complaints. Within 1 to 48 hours, the smoking of cocaine may produce a combination of diffuse alveolar infiltrates, eosinophilia, and fever that has been termed crack lung. Long-term cocaine exposure can produce diffuse alveolar damage, diffuse alveolar hemorrhage, noncardiogenic pulmonary edema, and pulmonary infarction.

Central Nervous System

Stimulation in acute toxicity can lead to euphoria, psychomotor agitation, violence, hyperthermia, and seizures. Cocaine-induced psychomotor agitation can cause hyperthermia when peripheral vasoconstriction prevents the body from dissipating the heat being generated from persistent agitation. The resulting fever has to be differentiated from other causes of hyperthermia in the setting of general anesthesia. Cocaine is associated with both focal neurologic deficits and coma. Possible causes include vasoconstriction (i.e., transient ischemic attack or ischemic stroke) and intracerebral hemorrhage. Minimum alveolar concentration (MAC) of halothane and other inhalational agents is increased with the long-term use of cocaine. Cocaine was found to delay gastric emptying via a central mechanism. This effect becomes more relevant in the setting of trauma and obstetrics. Cocaine and amphetamine–regulated transcript (CART) is a chemical that acts in the CNS to inhibit gastric acid secretion via brain corticotropin–releasing factor system.

Drug–Drug Interactions

Even though cocaine is a known inhibitor of the enzyme cytochrome P450 2D6, pharmacokinetic drug–drug interactions (DDIs) are generally unlikely to be clinically relevant. However, pharmacodynamic DDIs need to be taken into account in the perioperative period. Cocaine’s potent sympathomimetic effects may act synergistically with other drugs (e.g., stimulants, anticholinergic agents, and noradrenergic reuptake inhibitors) to produce an array of undesirable side effects (e.g., blurred vision, constipation, tachycardia, urinary retention, arrhythmias, and other effects). Synergistic pressor effects can produce vascular compromise that can precipitate cardiac ischemia or cerebrovascular accidents. Ketamine may exacerbate the sympathomimetic effect of cocaine. Halothane and xanthine derivatives sensitize the myocardium to the arrhythmogenic effect of epinephrine and should be avoided as well. Cocaine has been reported to alter the metabolism of succinylcholine because they both compete for metabolism by plasma cholinesterases. However, Birnbach found that succinylcholine can be used safely in standard doses. Cigarette smoking was found to enhance cocaine-induced coronary artery vasospasm in the atherosclerotic segments when compared with the vasoconstriction produced by cocaine alone. This effect was not evident in normal coronary arteries.

Evidence to Support Perioperative Risk of General Anesthesia with Acute Cocaine Toxicity

The risk of acute myocardial infarction (MI) is increased by a factor of 24 in the 60 minutes after the use of cocaine in persons who otherwise are at relatively low risk of myocardial ischemia. A meta-analysis, done in 1992, reported a total of 92 cases of cocaine-related MI. Two thirds of patients had their MI within 3 hours of the use of cocaine (with a range of 1 minute to 4 days). Data from the third National Health and Nutrition Examination Survey (NHANES III) found that 1 of every 20 persons ages 18 to 45 years reported regular use of cocaine. This survey demonstrated that the regular use of cocaine was associated with an increased likelihood of nonfatal MI. One of every four nonfatal MI in young patients was attributable to the frequent use of cocaine in this survey. No increased risk of nonfatal stroke was seen in this population associated with frequent or infrequent use of cocaine. The focus of research in this area is to determine risk factors for developing MI in cocaine-abusing patients. A recent study suggested that age, pre-existing coronary artery disease (CAD), hyperlipidemia, and smoking are associated with the diagnosis of MI among patients hospitalized with cocaine-associated chest pain. Cocaine-induced myocardial ischemia can occur regardless of whether CAD was pre-existing. However, it has been shown that coronary artery vasospasm tends to be more severe in the diseased segments of the coronary vessels when compared with the normal coronary arteries in response to intranasal cocaine in a dose of 2 mg/kg of body weight.

Most of the cases of cocaine-related myocardial ischemia are reported in the emergency medicine and internal medicine literature after recreational use of cocaine. Seven case reports of cocaine-induced myocardial ischemia were in the setting of the use of cocaine for topical anesthesia for ear, nose, and throat (ENT) procedures. In some of these cases, the patients were under general anesthesia. Two more cases of myocardial ischemia were reported with patients under general anesthesia after recreational use of cocaine. Other cardiac events reported with patients under general anesthesia with short-term use of cocaine include prolonged QT interval, ventricular fibrillation, and acute pulmonary edema. One case report described a patient coming to the operating room after a motor vehicle accident with a white foreign body in the back of the oropharynx that proved to be crack cocaine. This case goes on to report wide swings of blood pressure, patient agitation, and hypotension resistant to treatment with ephedrine.

One of the few studies that demonstrated the interaction between cocaine and general anesthesia was that by Boylan and colleagues. They found that increasing the depth of anesthesia with isoflurane from 0.75 MAC to 1.5 MAC in their swine model was not associated with reversal of, or decrease in, the hemodynamic responses to cocaine infusion. The observed responses were increase in systemic vascular resistance, ventricular arrhythmias, diastolic hypertension, and reversal of the endocardial/epicardial blood flow. Immediate administration of cocaine at a dose equivalent to doses abused by cocaine abusers decreased cerebral blood flow (CBF), cerebral blood volume (CBV), and tissue hemoglobin oxygenation StO ₂ in rats anesthetized with isoflurane ; cocaine-induced changes in CBF followed the peak uptake of cocaine in the brain.

Airway management may require special attention in acute cocaine toxicity. Supraglottic edema has been reported in this setting.

The half-life of cocaine ranges from 60 to 90 minutes. A reasonable assumption would be that most of the cocaine-related cardiac events in the perioperative period will happen at a time when the level of the metabolites, not the parent drug, is high in the circulation. The questions now are, “How active are the metabolites of cocaine, and can they affect the coronary vessels to the same extent as cocaine itself?” Brogan and colleagues randomly assigned 18 patients undergoing coronary artery catheterization for evaluation of chest pain to receive either intranasal cocaine or normal saline. They estimated the diameter of the coronary arteries and measured different hemodynamic variables at 30, 60, and 90 minutes. They found that coronary vasospasm happened twice, once at 30 minutes and the second at 90 minutes. The initial coronary artery vasospasm correlated with peak levels of cocaine in the blood. The recurrent vasospasm occurred at 90 minutes, when cocaine was hardly detected in the blood. The levels of the main metabolites of cocaine (benzoylecgonine and EME) were at their peak at this point. Although this study was able to document a temporal relation between recurrent coronary vasospasm and peak levels of cocaine metabolites, it did not prove that these metabolites were the cause of the vasoconstriction. Such proof will come only from assessment of coronary vasoreactivity after direct administration of each metabolite.

Recent studies have suggested that various metabolites of cocaine may exert a substantial influence on a variety of tissues, including the heart, brain, and arterial smooth muscle. In rats, norcocaine, another pharmacologically active metabolite of cocaine, was found to be equipotent to cocaine in inhibiting norepinephrine uptake and in causing tachycardia, convulsions, and death. In feline cerebral arteries in vitro, benzoylecgonine is a more potent vasoconstrictor than cocaine.

Evidence to Support the Relative Safety of General Anesthesia in Cocaine-Abusing Patients

The interaction between cocaine and general anesthesia is not well studied. Most of the information is derived from clinical case reports or animal studies. The few studies that looked into this interaction demonstrated that general anesthesia is probably safe in cocaine-abusing patients if certain conditions are met, especially in the absence of clinical signs of toxicity. Barash and colleagues studied 18 patients undergoing coronary artery surgery to examine whether cocaine in a clinically used dose exerts sympathomimetic effects during general anesthesia. Eleven patients received cocaine hydrochloride as a 10% solution (1.5 mg/kg) applied topically to the nasal mucosa. The other group received a placebo treatment. There were no important differences in cardiovascular function between groups. The rise in plasma cocaine concentration bore no relationship to any changes in cardiovascular function. Administration of topical cocaine did not exert any clinically significant sympathomimetic effect and appeared to be well tolerated in anesthetized patients with CAD. The results of this study should be interpreted cautiously because the doses used for recreational use may well exceed the doses used during this study.

A more recent study by Hill and colleagues studied 40 American Society of Anesthesiologists (ASA) physical status I and II patients between 18 and 55 years of age and demonstrated that individuals undergoing elective surgery requiring general anesthesia who test urine positive for cocaine but who do not show clinical toxicity are at no greater risk than drug-free patients of the same ASA physical status. The authors of this study caution that these results may not be applicable to the cocaine-abusing patient with a QT interval of 500 ms or more on a preoperative electrocardiogram or to those patients whose vital signs indicate acute cocaine toxicity. Another study looked into maternal morbidity in cocaine-abusing parturients undergoing cesarean section with general or regional anesthesia. Cocaine-abusing parturients were at higher risk of peripartum events such as hypertension, hypotension, and wheezing episodes. However, when the analysis was done in a multivariate model, cocaine abuse was not an independent risk factor. There was no increase in the rates of maternal morbidity or death in the cocaine-abusing group. Patients in the two referenced studies were relatively young and healthy. Based on the results of these two studies alone, it would be difficult to predict how anesthesia would interact with cocaine in the presence of multiple comorbidities.

Some authors proposed that patients who test positive for cocaine in their urine may undergo necessary surgical and anesthetic care, after an 8-hour period without cocaine, if they are hemodynamically stable and show no clinical signs of acute toxicity. This proposal was based on a survey of oral surgery and anesthesiology training programs in the United States. In the trauma setting, mortality rates and neurologic and cardiac complications during the first 24 hours after admission were not increased among patients testing positive after having a urine cocaine drug screen. Another study did not show a difference in mortality or length of intensive care unit stay between patients with cocaine-positive results and patients with cocaine-negative test results.

Regional Anesthesia and Cocaine-Abusing Patients

Any advantage of regional anesthesia over general anesthesia is controversial. The argument in favor of regional anesthesia, when possible, includes having an awake patient who will be able to communicate chest pain as a sign of myocardial ischemia. If regional anesthesia is selected, potential complications include combative behavior, altered pain perception, cocaine-induced thrombocytopenia, and ephedrine-resistant hypotension. Abnormal endorphin levels and changes in the mu and kappa receptors in the spinal cord may be responsible for pain sensation despite an adequate sensory level with regional anesthesia. The duration of action of spinal narcotics (sufentanil) in labor is shorter in cocaine-abusing parturients relative to control subjects. Many theories have been proposed to explain cocaine-induced thrombocytopenia. These include bone marrow suppression, platelet activation, and an autoimmune response with induction of platelet-specific antibodies. Gershon and colleagues challenged this concept. They concluded that obtaining a routine platelet count before epidural or spinal analgesia in cocaine-abusing parturients is not necessary.