Clinical Toxicology

Loralie Langman, Ph.D., F.C.A.C.B., D.A.B.C.C.(C.C.,M.B., T.C.), D.A.B.F.T., Laura Bechtel, Ph.D., D.A.B.C.C. and Christopher P. Holstege, M.D.

Toxicology is a broad, multidisciplinary science whose goal is to determine the effects of chemical agents on living systems. Innumerable potential toxins can inflict harm, including pharmaceuticals, herbals, household products, environmental agents, occupational chemicals, drugs of abuse, and chemical terrorism threats. Each year millions of human exposure cases are reported to the American Association of Poison Control Centers.⁹⁶ The Centers for Disease Control has reported that poisoning (both intentional and unintentional) is one of the top 10 causes of injury-related death in the United States in all adult age groups. From the beginnings of written history, poisons and their effects have been well described. Paracelsus (1493-1541) correctly noted that “Alle Ding sind Gift, und nichts ohn Gift; allein die Dosis macht, daß ein Ding kein Gift ist,” which means, “Everything is a poison; there is nothing which is not. The dose differentiates a poison.” As life in the modern era has become more complex, so has the study of poisons and their treatments.

This chapter provides a general overview of Clinical Toxicology and the laboratory services necessary to support the care of poisoned patients. Because a comprehensive discussion of all aspects of toxicology is beyond the scope of this chapter, the clinical significance and toxicity of only a select number of common drugs, drugs of abuse, and other chemicals are discussed.

Basic Information

In practice, it is neither possible nor necessary to test for all of the hundreds or thousands of clinical toxins that may be encountered. In reality up to 24 drugs or agents account for 80% or more of cases of intoxication treated in most emergency departments.⁵⁸⁸ Moreover, some drugs are encountered very infrequently in some locations but with relatively high frequency in others. For example, phencyclidine (PCP) use is almost nonexistent in some areas but is responsible for a relatively high number of intoxications in a few large metropolitan cities. Thus the scope of clinical toxicology testing provided by the laboratory will depend on the pattern of local drug use and on the available resources of the institution and should be developed in consultation with the appropriate clinical staff.

The value of drug and/or substance testing (screening) is well established (1) in the workplace, (2) for some athletic competitions, (3) to monitor drug use during pregnancy, (4) to evaluate drug exposure and/or withdrawal in newborns, (5) to monitor patients in pain management and drug abuse treatment programs, and (6) to aid in the prompt diagnosis of toxicity for a select number of drugs or agents for which a specific antidote or treatment modality is required (Table 35-1). In many other instances of drug toxicity, the value of drug screening, especially on an emergency basis, is more controversial.^{94,411,552,695}

TABLE 35-1

Antidote or Specific Treatment for Intoxication

Toxin	Antidote/Treatment^{14,68,185,251,722}
Acetaminophen	N-Acetylcysteine
Aluminum, iron	Deferoxamine
Anticholinergic agents	Physostigmine
Arsenic	Dimercaprol; 2,3-dimercaptosuccinic acid; D-penicillamine
Barbiturates	Multiple-dose oral activated charcoal; alkaline diuresis (phenobarbital only)
Benzodiazepines	Flumazenil
Beta-blockers	Glucagon
Calcium channel blockers	Calcium; glucagon; high-dose insulin infusion
Carbamazepine	Multiple-dose oral activated charcoal; charcoal hemoperfusion
Carbon monoxide	Oxygen
Cyanide	Amyl nitrite, sodium nitrite, sodium thiosulfate; hydroxocobalamin
Digoxin	Anti-digoxin Fab fragments
Ethylene glycol, methanol	Fomepizole (4-methylpyrazol) or ethanol; hemodialysis
Isoniazid	Pyridoxine
Lead	Calcium disodium edetate; dimercaprol; 2,3-dimercaptosuccinic acid
Lithium	Hemodialysis
Mercury	Dimercaprol; 2,3-dimercaptosuccinic acid; D-penicillamine
Methanol	Fomepizole (4-methylpyrazol) or ethanol; hemodialysis; folate
Nitrites, nitrates	Methylene blue
Opioids	Naloxone
Organophosphate or carbamate	Atropine; pralidoxime (controversial for carbamates)
Salicylates	Bicarbonate; hemodialysis
Theophylline	Multiple-dose oral activated charcoal; hemodialysis
Tricyclic antidepressants	Bicarbonate; benzodiazepines

Approaches to drug testing vary from the provision of just a few specific tests (e.g., acetaminophen, salicylate, ethanol, digoxin, iron) to testing for additional targeted groups of drugs (e.g., stimulant panel and coma panel) or to a more comprehensive general drug screen that might include hundreds of drugs and/or substances. For all of these situations, it is imperative that the laboratory communicate with the physician concerning the scope (and limitation) of the service and the proper timing and selection of specimens; when possible, the laboratory should assist with interpretation of results. At a minimum, the laboratory request slip should clearly state the drugs that it has the capability of detecting. Otherwise, the report of a “negative” result for a drug screen could be misleading.

Clinical Considerations

To operate effectively, the laboratory should be closely associated with the healthcare team directly managing the patient. Through close and collaborative work, clinical information provided will help to guide appropriate ordering of tests and to ensure that interpretation of results is complete and accurate. For example, the team caring for the patient should provide the following information with the laboratory request:

1. The time and date of the suspected exposure along with the time and date of sample collection.

2. History from the patient or witnesses that might aid in identification of the toxin.

3. Assessment of the physical state of the patient at the time of presentation.

Such information is useful to guide test selection and interpretation of results.

Analytical Considerations

Because of the wide range of drugs of interest, no single analytical technique is adequate for broad-spectrum drug detection. Therefore several analytical approaches in combination are generally required. These may include simple, inexpensive, and rapid spot tests; immunoassays (see Chapter 16); and chromatographic and/or mass spectrometric techniques (see Chapters 13 and 14), including thin-layer chromatography (TLC), high-performance liquid chromatography (HPLC), gas chromatography (GC), gas chromatography-mass spectrometry (GC-MS or GC-MS/MS), and liquid chromatography-mass spectrometry (LC-MS or LC-MS/MS).⁷⁷² Currently, GC-MS is the most widely used definitive confirmatory procedure, although LC-MS/MS is becoming more popular. Confirmatory testing is mandatory for forensic drug testing (e.g., workplace drug testing).

Speed of analysis, or turnaround time (TOT), and availability are critical issues in clinical toxicology. A drug analysis that requires several hours to complete or that is not available at all hours of the day is of little value in a clinical emergency. Alternatively, a rapid test that provides false information could result in erroneous diagnostic and therapeutic decisions. For numerous agents, quantitative determinations guide management during a clinical emergency. These agents include (1) acetaminophen, (2) carbamazepine, (3) digoxin, (4) ethanol, (5) ethylene glycol, (6) iron, (7) isopropanol, (8) lithium, (9) methanol, (10) phenobarbital, (11) phenytoin, (12) salicylate, (13) valproic acid, and (14) theophylline, and in whole blood, (15) carboxyhemoglobin and (16) methemoglobin. Results for these determinations should be available within 1 hour of specimen receipt.⁸⁶⁰

Proper selection of analytical methods and interpretation of results require knowledge of the pharmacology and pharmacokinetics of the toxins of interest. For example, the potential hepatotoxicity of acetaminophen is related to the concentration of unmetabolized drug. Conversely, delta-9-tetrahydrocannabinol (THC)—a metabolite of marijuana—is measured in urine as an indication of marijuana use.

Clinical Evaluation

Primary Survey

When a healthcare team initially evaluates a patient who presents with a potential toxicologically induced health problem, the final diagnosis is often determined by (1) reviewing the history, (2) performing a directed physical examination, (3) using ancillary tests (e.g., electrocardiogram, radiology), and (4) applying a rational approach to laboratory testing. Often no specific antidote or treatment is available for a poisoned patient and careful supportive care is the most important intervention.²³⁴

All patients who present with potential toxicity should be thoroughly assessed, and it is imperative that the clinician follow a standard “ABC” approach with attention to “airway, breathing, and circulation” respectively. The patient’s airway should be open and unblocked and adequate ventilation ensured. If the patient’s airway is not secure and endotracheal tube intubation is considered, the first diagnostic test that should be performed is a rapid bedside glucose concentration. Hypoglycemia can result in coma or new-onset seizures, thereby mimicking a toxic etiology. In addition, numerous toxins are clinically associated with hypoglycemia (e.g., sulfonylureas, iron, Mentha pulegium). Clinical effects induced by hypoglycemia can be rapidly reversed with intravenous glucose, thus preventing unnecessary and costly procedures and testing.

Too often healthcare providers are lulled into a false sense of security when a patient presents with altered mental status and oxygen saturations on pulse oximetry that are adequate on high-flow oxygen. If the patient has inadequate ventilation or a poor gag reflex, then the patient may be at risk for progressive CO₂ narcosis or aspiration, respectively, and yet may be maintaining adequate oxygen saturation on supplemental oxygen. It is imperative that the healthcare team avoid common complications of poisonings such as aspiration pneumonitis.138,359 An arterial blood gas (ABG) can rapidly aid the healthcare team in determining the need for intubation and mechanical ventilation. The ABG can also provide valuable information regarding the patient’s acid-base status and can help the clinician begin to generate a differential diagnosis. For example, in the scenario of a febrile toxic patient who presents with an altered mental status, a normal ABG (lack of acidosis) immediately rules-out uncoupling of oxidative phosphorylation as a cause of that patient’s fever. Finally, an ABG with co-oximetry can rapidly assist in determining other toxic etiologies, such as carbon monoxide poisoning (depending on the timing of the blood draw in relation to the exposure) and methemoglobinemia.

The initial treatment of hypotension in all toxic patients consists of the administration of intravenous fluids.³⁴⁰ The patient’s pulmonary status should be closely monitored to ensure that pulmonary edema does not develop as fluids are infused. Symptomatic toxic patients should be placed on continuous cardiac monitoring with pulse oximetry, and the healthcare team must perform frequent neurologic checks to ensure continued protection of the airway. Acutely poisoned patients should receive a large-bore peripheral intravenous line, and all symptomatic patients should have a second line placed in the peripheral or central venous system, depending on the severity of their clinical status. At this time in patient care, blood can be drawn and sent for appropriate laboratory diagnostic testing. Placement of a urinary catheter should be considered early in the care of hemodynamically unstable poisoned patients to monitor urinary output as an indicator of adequate perfusion. A rapid bedside urine dipstick (photometric chemical assay) provides helpful information rapidly as healthcare team members await further laboratory testing. For example, a urine specific gravity will give insight into the patient’s initial hydration status, and the appearance of tea-colored urine positive for blood may indicate the presence of myoglobinuria in a comatose patient with rhabdomyolysis.

Secondary Survey

Upon completion of the primary survey, the healthcare team should be assured that the patient’s airway is open and unblocked (also described as a patent airway), the ventilatory effort is adequate, and blood pressure is sustained in an appropriate range. At this point in management, the secondary survey can be performed. The secondary survey involves a thorough examination of the entire patient. For adequate access to a toxic patient, the patient must be completely undressed. Exposure of the patient ensures that a complete physical examination is performed.²³⁴ If the patient is not completely undressed, an important diagnostic clue may be missed. For example, skin lesions consistent with pressure necrosis on the back of a comatose patient may indicate the need to obtain a blood creatine phosphokinase activity or a urine myoglobin concentration. A comatose drug abuser may have attached transdermal drug patches (e.g., fentanyl, clonidine) in atypical locations (e.g., gluteal sulcus) that when found can rapidly lead to a diagnosis, avoiding the need for further laboratory testing. Besides completing a thorough physical review of all organ systems, the secondary survey involves reviewing items brought with the patient (e.g., medication bottles, drug paraphernalia). Searching carefully through the patient’s clothing may assist in providing clues that change the plan for specific laboratory tests or explain specific laboratory findings. For example, the discovery of a cough and cold product in the patient’s pocket that contains dextromethorphan could explain the clinical presentation of an agitated patient with hyperreflexia whose initial urine toxicology screen was positive for phencyclidine but later was found negative on confirmation.

Toxic Syndromes

Toxic syndromes (“toxidromes”) are clinical syndromes that are essential for the successful recognition of poisoning patterns. A toxidrome is the constellation of clinical signs and symptoms that suggests a specific class of poisoning. An important component of the secondary survey is to determine whether a specific toxic syndrome is present.²³⁴ The most commonly encountered toxidromes include (1) anticholinergic, (2) cholinergic, (3) opioid, (4) sedative-hypnotic, and (5) sympathomimetic (Table 35-2). Many toxidromes have several overlapping features. For example, anticholinergic findings are highly similar to sympathomimetic findings, with one exception being the effects on sweat glands: anticholinergic agents produce warm, flushed dry skin, but sympathomimetic agents produce diaphoresis. Toxidrome findings may also be affected by individual variability, comorbid conditions, and coingestants. For example, tachycardia associated with sympathomimetic or anticholinergic toxidromes may be absent in a patient who is concurrently taking beta-antagonist medications. Additionally, although toxidromes may be applied to classes of drugs, one or more toxidrome findings may be absent for some individual agents within these classes. For instance, meperidine is an opioid analgesic, but it does not induce miosis, which helps to define the “classic” opioid toxidrome. When accurately identified, the toxidrome may provide invaluable information for diagnosis and subsequent treatment, although the many limitations impeding acute toxidrome diagnosis must be carefully considered.

TABLE 35-2

Symptoms of the Important Toxidromes

Hyperthermic Syndromes

Toxin-induced hyperthermic syndromes are potentially devastating and require rapid management. Even though the patient’s temperature is one of the vital signs, the temperature often is not obtained in clinical practice. Fever in a poisoned patient can be associated with several hyperthermic syndromes: sympathomimetic toxicity, uncoupling of oxidative phosphorylation, serotonin syndrome, neuroleptic malignant syndrome, malignant hyperthermia, anticholinergic poisoning, and withdrawal syndromes.⁶⁹² Sympathomimetics, such as amphetamines and cocaine, may produce hyperthermia as the result of excess serotonin and dopamine, leading to thermal deregulation.³⁷² Uncoupling of oxidative phosphorylation, as seen in severe salicylate poisoning, occurs when the process of oxidative phosphorylation is disrupted, leading to heat generation and a reduced ability to aerobically generate adenosine-5′-triphosphate (ATP).⁴⁰⁹ Serotonin syndrome occurs when a relative excess of serotonin is present at both peripheral and central serotonergic receptors.³⁶⁰ Patients may present with hyperthermia, alterations in mental status, and neuromuscular abnormalities (rigidity, hyperreflexia, clonus), although individual variability is noted in these findings. Serotonin syndrome is associated with drug interactions such as those associated with the combination of monoamine oxidase inhibitors and meperidine, but it may also occur with single-agent therapeutic dosing or overdosing of serotonergic agents. Neuroleptic malignant syndrome is a condition caused by relative deficiency of dopamine within the central nervous system.⁷¹⁷ It has been associated with dopamine receptor antagonists and the withdrawal of dopamine agonists such as levodopa/carbidopa products. Clinically, it may be difficult to distinguish from serotonin syndrome and other hyperthermic emergencies. Malignant hyperthermia occurs when genetically susceptible individuals are exposed to depolarizing neuromuscular blocking agents or volatile general anesthetics. Anticholinergic poisoning may result in hyperthermia through impairment of normal cooling mechanisms such as sweating. Withdrawal syndromes can produce excessive adrenergic responses (e.g., sedative-hypnotic withdrawal, ethanol withdrawal) and subsequent heat generation. It should be noted that opioid withdrawal is not associated with fever or altered mental status. Overall, differentiating between the toxic hyperthermic syndromes may be challenging, and additional causes of hyperthermia such as heat stroke and infection should be explored.

Screening Procedures for Detection of Drugs

Screening procedures are designed for the relatively rapid and generally qualitative detection of drugs or other toxic substances. In general, screening tests have adequate clinical sensitivity but may not be highly specific. Thus, a negative result yielded by a screening procedure may rule out with reasonable certainty the presence of clinically significant concentrations of a particular analyte. Because of possible interferences, a positive result should be considered “presumptive positive” and should be confirmed by an alternate procedure of greater specificity. Screening procedures may be designed to detect a particular drug or drug class. Tests for such purposes include simple visual color tests (spot tests) and immunoassays.

Spot Tests

Spot tests are rapid, easily performed, noninstrumental qualitative procedures that provide presumptive evidence for the presence of tested drugs. Any positive response must be followed by testing with a more specific method. They are potentially valuable to rule out the presence of drugs or to suggest (but not prove) the presence of a drug of a particular group. Spot tests are less frequently employed now because many have been largely replaced by rapid immunoassays that may be performed at the point of care or in the central laboratory.

Ferric Chloride and Trinder Tests

Two bedside tests have been proposed for rapid identification of the patient with salicylate toxicity. Both involve simply applying a few drops of a prepared reagent to a small sample of a patient’s urine and watching for a characteristic color change. The first such reagent is ferric chloride (FeCl₃). Applying a few drops of a 10% solution of ferric chloride to 1 mL of urine containing even very small amounts of salicylate will produce a characteristic purple color caused by the formation of an iron-salicylate complex.³³⁷ This color change also will occur if the urine in question contains acetoacetic acid and phenylpyruvic acid.²⁵⁰ The urine Trinder spot test uses a reagent composed of mercuric chloride, ferric nitrate, concentrated hydrogen chloride, and deionized water.⁴²¹ Applying 1 mL of this solution to 1 mL of urine with salicylates present will also lead to a purple color change.²⁵⁰

One study exploring the salicylate question ²⁵⁴ examined the use of the ferric chloride test in 187 patients presenting to the emergency department. These ferric chloride tests were subsequently followed with serum salicylate concentrations. The sensitivity of the ferric chloride was 93.2% for salicylate concentrations ≥3.0 mg/dL and 93.8% for concentrations ≥30.0 mg/dL. Specificity was 88.8% and 75.4%, respectively. Three false negatives were reported, one of which had a toxic salicylate concentration of 34 mg/dL. King and associates evaluated the clinical utility of the Trinder reagent.⁴²¹ Investigators enlisted 12 volunteers who ingested 975 mg of aspirin. The sensitivity of the test in this study was 100% with two false positives from controls.

Weiner and colleagues examined the application of these two bedside tests in patients presenting to the emergency department with suspected drug overdose with or without unexplained metabolic acidosis.⁸²⁷ Both reagents were added to urine samples from all 180 patients enrolled, and confirmatory serum salicylate concentrations were drawn. Different from the previous studies, however, any darkening of color was regarded as a positive test. Twenty of these patients (11%) had salicylate concentrations ≥5 mg/dL. Both tests were 100% sensitive for recognizing these patients. The specificities for both tests were relatively low (Trinder with 73% and ferric chloride with 71%).

Overall both of these tests can be used for rapid bedside testing. Each is relatively inexpensive and, when color change is interpreted as suggested by Weiner and coworkers, has a low limit of detection. Positive tests indicate the possible presence of salicylate and not toxicity. Both tests should be followed with determination of serum salicylate concentrations to confirm toxicity and quantitate the salicylate concentration.

Urine Fluorescence

Automotive antifreeze is a major source of ethylene glycol exposure. Sodium fluorescein is added to the antifreeze to aid in identifying cooling system leaks. Some have suggested that fluorescein excreted in the urine of a patient who has ingested antifreeze would fluoresce with the aid of a Wood’s lamp.⁸¹⁸ Unfortunately, it seems that little advantage is derived by checking for urinary fluorescence by Wood’s lamp in those suspected of ethylene glycol poisoning.^120,818

Anion Gap

Obtaining a basic metabolic panel in all poisoned patients is recommended and is an important initial screening test. When low serum bicarbonate is discovered on a metabolic panel, the clinician should determine whether an elevated anion gap exists. The formula most commonly used for the anion gap (AG) calculation is as follows ¹²⁴:

This equation allows one to determine whether serum electroneutrality is being maintained. The primary cation (sodium) and anions (chloride and bicarbonate) are represented in the equation.³⁶⁴ Other contributors to this equation are “unmeasured.”²⁶⁷ Other serum cations are not commonly included in this calculation because either their concentrations are relatively low (e.g., potassium) or assigning a number to represent their respective contribution is difficult (e.g., magnesium, calcium).²⁶⁷ Similarly, a multitude of other serum anions (e.g., sulfate, phosphate, organic anions) are also difficult to measure and quantify in an equation.^267,364 These “unmeasured” ions represent the anion gap calculated using the previous equation. The reference interval for this anion gap is accepted to be 8 to 16 mmol/L,²⁶⁷ but it has been suggested that because of changes in the technique used to measure chloride, the interval should be lowered to 6 to 14 mmol/L.³⁶⁴ Practically speaking, an increase in the anion gap beyond an accepted reference interval, accompanied by a metabolic acidosis, represents an increase in unmeasured endogenous (e.g., lactate) or exogenous (e.g., salicylates) anions.¹²⁴ A list of the more common causes of this phenomenon is organized in the classic MUDILES pneumonic (Box 35-1). Note: The “P” has been removed from the older acronym of MUDPILES, because paraldehyde is no longer available.

Box 35-1 Common Causes of Increased Anion Gap (MUDILE)

Methanol

Uremia

Diabetic ketoacidosis

Iron, inhalants (i.e., carbon monoxide, cyanide, toluene), isoniazid, ibuprofen

Lactic acidosis

Ethylene glycol, ethanol ketoacidosis

Salicylates, starvation ketoacidosis, sympathomimetics

It is imperative that clinicians who admit poisoned patients initially presenting with an increased anion gap metabolic acidosis investigate the cause of that acidosis. Many symptomatic poisoned patients may have an initial mild metabolic acidosis upon presentation caused by processes resulting in elevated serum lactate. However, with adequate supportive care including hydration and oxygenation, the anion gap acidosis should improve. If, despite adequate supportive care, an anion gap metabolic acidosis worsens in a poisoned patient, the clinician should consider continued absorption of exogenous acids (e.g., salicylate), formation of acidic metabolites (e.g., ethylene glycol, methanol, toluene metabolites), and cellular ischemia with worsening lactic acidosis (e.g., cyanide) as potential causes.

Electrocardiogram

Interpretation of the electrocardiogram (ECG) in the poisoned patient can significantly facilitate appropriate laboratory testing, diagnosis, and management of the poisoned patient, because numerous drugs can cause ECG changes.⁸³² Despite the fact that drugs have widely varying indications for therapeutic use, many unrelated drugs share common cardiac electrocardiographic effects if taken in overdose. Potential toxins can be placed into broad classes on the basis of their cardiac effects. For example, agents that block cardiac potassium efflux channels and agents that block cardiac fast sodium channels can lead to characteristic changes in cardiac indices consisting of QRS prolongation and QT prolongation, respectively. The recognition of specific ECG changes associated with other clinical data (toxidromes) can be potentially lifesaving.³⁴¹

Radiographic Studies

Radiologic testing is sometimes used to diagnose complications associated with poisonings, such as aspiration pneumonitis and anoxic brain injury. The use of radiology has also been advocated to detect the presence of potentially radiopaque poisons. For example, O’Brien and associates studied the detectability of 459 different tablets and capsules using plain radiography.⁵⁹⁶ Investigators used a ferrous sulfate tablet as a control in grading the radiopacity of other tablets. Overall, of the wide variety of pills tested, only 6.3% were graded as having radiopacity the same as or greater than ferrous sulfate; 29.6% were regarded as having at least moderate opacity; and the largest remaining portion of pills (64%) was regarded as no more than minimally detectable. Based on this and other studies, the indiscriminate use of plain abdominal x-rays is not justified, and a negative film should not be relied upon to rule out potential toxic pill ingestion, especially if enough time is given to allow the pills to dissolve.

Osmol Gap

The main osmotically active constituents of serum are Na⁺, Cl⁻, , glucose, and urea. Several empirical formulas based on measurement of these substances have been used to estimate the serum osmolality.276,539,606,617,662 In practice, one has not shown itself to be superior to the others, yet each equation demonstrates significant differences in the osmol gap reference interval.⁴⁴³ Therefore, reference intervals must be validated on appropriate patient populations. Two commonly used formulas (in conventional and SI units) are presented here:

The difference between the actual osmolality (OSMm), measured by freezing-point depression, and the calculated osmolality (OSMc) is referred to as delta-osmolality, or the osmol gap (OSMg).

Elevated OSMg implies the presence of unmeasured osmotically active substances.⁴²⁷ Volatile alcohols (ethanol, methanol, isopropanol, acetone, and ethylene glycol) when present at significant concentrations, increase serum osmolality, thus resulting in an increased OSMg. The calculation of OSMg is commonly used as a screen.²⁷⁶ However, it is important to remember that volatile alcohols are not detected when osmolality is measured with a vapor pressure osmometer. Therefore, for the purpose of determining the OSMg, only osmolality measurements based on freezing-point depression are acceptable.

What constitutes a normal osmol gap is widely debated. Traditionally, a normal gap has been defined as 10 mOsm/kg or less. The original source of this value is an article by Smithline and Gardner,⁷⁴² which declared that this number was pure convention. Further clinical study has not shown this assumption to be correct. However, large variability is seen in the normal population.^456,558 Researchers have found the OSMg to vary from −9 to +5 mOsm/kg²⁸⁸; from −13.5 to +8.9⁵³⁶; and from −10 to +20 mOsm/kg,¹ depending on the population studied. An important point to consider is that the day-to-day coefficient of variance of sodium was 1%. This analytical variance alone may account for the variation found in patients’ osmol gaps.¹

One would expect that each 100 mg/dL (21.7 mmol/L) of ethanol (molecular weight = 46.068 g/mol) in serum results in an approximate increase of 21.7 mOsm/kg.⁶⁴⁰ However, this is not found to be the case. Applying a correction factor of 0.83 to the ethanol value will more closely approximate the contribution of ethanol to the OSMg.²⁷⁶ By considering this effect of ethanol on the serum osmolality, it is possible to determine what portion of an increased osmol gap is due to ethanol. The contribution of ethanol to the measured osmolality can be calculated (ethanol, mg/dL/4.6 × 0.83) and included in the preceding formula for delta osmolality calculation. However, it has been observed that ethanol and methanol do not follow a completely predictable relationship with OSMg. In severe ethanol and methanol intoxication, OSMg increases with increasing concentration, making it appear that something is present besides the alcohol.^{276,469,539,606,662}

A significant residual osmol gap (>10 mOsm/kg) after the correction for ethanol would suggest the possible presence of isopropanol, methanol, acetone, or ethylene glycol. This information, in conjunction with the presence or absence of metabolic acidosis or serum acetone, is helpful to the clinician when specific measurements of alcohols other than ethanol and of ethylene glycol are not available on an emergency basis (Table 35-3). It must be realized that ketones (diabetics) and substances administered to patients such as polyethylene glycol (burn cream),⁹⁸ mannitol (osmotic diuretic), and propylene glycol (solvent for diazepam and phenytoin) may increase serum osmolality.

TABLE 35-3

Laboratory Findings Characteristic of Ingestion of Alcohols

For the diagnosis of ethanol intoxication, OSMg poisoning has lost its usefulness because ethanol is measured quickly on most chemistry analyzers. However, because other toxic alcohols can be measured only by chromatographic techniques, it is still useful. Unfortunately, OSMg as a screening method is insensitive to low, yet clinically significant, concentrations of ethylene glycol (<50 mg/dL), and methanol (<30 mg/dL).⁶¹⁶

Immunoassay

Different types of immunoassays are useful in screening specimens for drugs (see Chapters 16 and 34). In some cases, these assays are relatively specific for a single drug (LSD), but in others, several drugs of a similar class are detected (e.g., opiates). The detection limit for various members of a class of drugs or the degree of cross-reactivity for similar drugs varies, and each manufacturer of immunoassay reagents should be consulted for specific information. These assays are easy to perform; many are available for use on automated instrumentation and may be able to provide “semiquantitative” results. Several portable, noninstrumental, immunoassay-based drug detection devices are available for use in point-of-care testing (POCT). For the vast majority of drugs of abuse, immunoassays are the methods of choice for initial screening. However, for a more comprehensive drug screening, chromatographic procedures complement immunoassays.

Planar Chromatography

Planar chromatography, commonly known as thin-layer chromatography (TLC), is a versatile procedure that requires no instrumentation and thus is operationally relatively simple and inexpensive (see Chapter 13). With this technique, a large number of drugs may be detected; it may be applied to the analysis of serum, gastric contents, or urine. Urine, however, is the specimen of choice because most drugs and drug metabolites are present in urine in relatively high concentrations. However, application of TLC to drug screening requires considerable experience and skill to recognize drug and metabolite patterns and various color hues for detection; it has largely been replaced by other chromatographic techniques. For a more comprehensive description, refer to previous editions of this textbook.⁷⁸

Gas Chromatography

Also known as gas liquid chromatography, GC is relatively rapid, is capable of resolving a broad spectrum of drugs, and is widely used for qualitative and quantitative drug analysis.³⁹² Capillary columns, because of their high efficiency, are analytical columns that are commonly used for drug detection by GC (see Chapter 13). In many instances, nonderivatized drugs have good GC properties when capillary columns are used; in some instances, derivatization to a less polar or more volatile compound is necessary. Common detectors for drug detection by GC are flame ionization and alkali flame ionization (nitrogen phosphorus) detectors and mass spectrometers, which provide the greatest accuracy of identification. Numerous methods for general drug screening by GC-MS spectrometry have been published, but one comprehensive method can be adapted to multiple body fluids and tissues.³⁹²

High-Performance Liquid Chromatography

The resolving power of HPLC (see Chapter 13) for separating widely divergent chemical constituents has been applied to the complex challenge of comprehensive drug screening in biological fluids. Advantages of HPLC over GC include the ability to analyze polar and thermally labile drugs without derivatization. The advent of diode array detectors that provide a spectral scan of compounds as they elute from the column greatly increased the discriminatory power of this technique.^{82,500,661,785} LC-MS or LC-MS/MS currently plays a limited role in comprehensive screening but is rapidly gaining in popularity.*

Point of Care (POC) Drug Testing

Numerous POC drug test devices for urine (and oral fluid) are designed for easy, rugged, and portable use by nontechnical personnel. Although these devices are relatively simple to use, proper training of nonlaboratory users is important for optimal performance (see Chapter 20).^279,445 These noninstrumental immunoassay test devices are designed for use at the site of collection; results are available within minutes and are variously configured to detect only one drug or many drugs simultaneously. The spectrum of drugs tested includes tricyclic antidepressants, barbiturates, benzodiazepines, methadone, MDMA (methylenedioxymethamphetamine), MDA (methylenedioxyamphetamine), MDEA (methylenedioxyethylamphetamine), oxycodone, and the traditional SAMHSA (Substance Abuse and Mental Health Services Administration) or NIDA (National Institute on Drug Abuse) 5 (amphetamine, cocaine, marijuana, opiates, and phencyclidine). As previously cited, such devices are also available for measurement of acetaminophen and salicylate in serum or whole blood. Evaluations for some of these test devices for urine and oral fluid^558,639 have been published. A comprehensive review of on-site drug testing is also available.⁶²⁹ The assay principles of these POC test devices include the following:

• Sequential competitive binding microparticle capture immunoassay

• Homogeneous microparticle capture immunochromatography

• Solid-phase competitive sequential enzyme immunoassay

• Latex-agglutination-inhibition immunoassay

A more detailed description of these methods can be found in Chapter 16 and in the package insert for each specific test kit.

Pharmacology and Analysis of Specific Drugs and Toxic Agents

The toxic, pharmacologic, biochemical, and analytical characteristics of several individual drugs and toxins are discussed in this section.

Agents That Cause Cellular Hypoxia

Carbon monoxide and methemoglobin-forming agents interfere with oxygen transport, resulting in cellular hypoxia. Cyanide interferes with oxygen use and therefore causes an apparent cellular hypoxia.

Carbon Monoxide

Carbon monoxide (CO) is a colorless, odorless, tasteless gas that is a product of incomplete combustion of carbonaceous material. Common exogenous sources of carbon monoxide include cigarette smoke, gasoline engines, and improperly ventilated home heating units. Small amounts of carbon monoxide are produced endogenously in the metabolic conversion of heme to biliverdin.⁴⁵⁸ This endogenous production of carbon monoxide is accelerated in hemolytic anemias.⁴⁹⁶

Toxic Effects

When inhaled, carbon monoxide combines tightly with the heme Fe²⁺ of hemoglobin to form carboxyhemoglobin. The binding affinity of hemoglobin for carbon monoxide is about 250 times greater than that for oxygen. Therefore high concentrations of carboxyhemoglobin limit the oxygen content of blood. Moreover, the binding of carbon monoxide to a hemoglobin subunit increases the oxygen affinity for the remaining subunits in the hemoglobin tetramer. Thus at a given tissue PO₂ value, less oxygen dissociates from hemoglobin when carbon monoxide is also bound, shifting the hemoglobin-oxygen dissociation curve to the left. Consequently, carbon monoxide not only decreases the oxygen content of blood, it also decreases oxygen availability to tissue, thereby producing a greater degree of tissue hypoxia than would result from an equivalent reduction in oxyhemoglobin due to hypoxia alone.368,778 Carbon monoxide may also bind to other heme proteins, such as myoglobin and mitochondrial cytochrome oxidase a₃; this may limit oxygen use when tissue PO₂ is very low.^368,778

The toxic effects of carbon monoxide are a result of hypoxia. Organs with high oxygen demand, such as heart and brain, are most sensitive to hypoxia and thus account for the major clinical sequelae of carbon monoxide poisoning. It must be emphasized that the carboxyhemoglobin concentration, although helpful in diagnosis, does not always correlate with the clinical findings or prognosis.569,702 Factors other than carboxyhemoglobin concentration that contribute to toxicity include length of exposure, metabolic activity, and underlying disease, especially cardiac or cerebrovascular disease. Moreover, low carboxyhemoglobin concentrations relative to the severity of poisoning may be observed if the patient was removed from the carbon monoxide–contaminated environment several hours before blood sampling.³¹⁸

An insidious effect of carbon monoxide poisoning is the delayed development of neuropsychiatric sequelae, which may include personality changes, motor disturbances, and memory impairment. These manifestations do not correlate with the length of exposure or with the maximum blood carboxyhemoglobin concentration.⁴⁵

Treatment for carbon monoxide poisoning involves removal of the individual from the contaminated area and administration of oxygen. The half-life (t_1/2) of carboxyhemoglobin in the body is variable, and attempts to determine the exact elimination t_1/2 for CO based on the inhaled oxygen concentration have not been validated. Hyperbaric oxygen therapy for CO is highly debated, and current position papers have found no evidence to support its use.¹⁰¹

Analytical Methods

Carbon monoxide may be released from hemoglobin and then measured by GC, or it may be determined indirectly as carboxyhemoglobin by spectrophotometry. Gas chromatographic methods are accurate and precise even for very low concentrations of carbon monoxide. Spectrophotometric methods are rapid, convenient, accurate, and precise, except at very low concentrations of carboxyhemoglobin (<2 to 3%).

Gas chromatographic methods measure the carbon monoxide content of blood. When blood is treated with potassium ferricyanide, carboxyhemoglobin is converted to methemoglobin, and carbon monoxide is released into the gas phase. Measurement of the released carbon monoxide may be performed by GC using a molecular sieve column and a thermal conductivity detector.²¹³ A lower detection limit is achieved by incorporating a reducing catalyst (e.g., nickel) between the GC column and the detector to convert carbon monoxide to methane. The methane may then be detected with a flame ionization detector.³⁰⁷ A very low detection limit may be achieved with the use of a heated mercuric oxide reaction chamber between the GC column and an ultraviolet light detector. As carbon monoxide elutes from the column, it reacts with mercuric oxide to form mercury gas, which has a high molar absorptivity at 254 nm.⁸⁰⁹ In practice, the carbon monoxide binding capacity is also determined after an aliquot of the blood specimen is treated with carbon monoxide to saturate the hemoglobin. The results are then expressed as percent of carboxyhemoglobin:

GC methods are accurate and precise and are considered to be reference procedures. Normal values for carboxyhemoglobin in rural nonsmokers are about 0.5%; for urban nonsmokers, 1 to 2%; and for smokers, 5 to 6%.⁴⁹ Values may be increased by about 3% in cases of hemolytic anemia.⁴⁹⁶

Spectrophotometric methods rely on the characteristic spectral absorption properties of carboxyhemoglobin.197,879 Among several such methods, the most popular are based on automated, multiwavelength measurements of several hemoglobin species. These methods are rapid and convenient for the determination of carboxyhemoglobin and other hemoglobin species. Spectrophotometric methods generally compare favorably with gas chromatographic procedures at carboxyhemoglobin concentrations greater than 2 to 3%, but their precision is poor below these concentrations.⁸⁰⁹ Therefore, they are sufficiently accurate and precise for measurement of carbon monoxide after exogenous exposure but are too insensitive to detect the increased endogenous production of carbon monoxide that occurs in hemolytic anemia.

Fetal hemoglobin has slightly different spectral properties than adult hemoglobin. Consequently, falsely high carboxyhemoglobin values of 4 to 7% may occur when blood from neonates is measured by some spectrophotometric methods utilizing fewer wavelengths.⁸¹⁰ Moreover, erroneous results may occur with lipemic specimens, with bilirubin, and in the presence of methylene blue (see section on “Methemoglobin-Forming Agents”).

Cyanide

Cyanide is a chemical group that consists of one atom of carbon bound to one atom of nitrogen by three molecular bonds (C≡N). Inorganic cyanides (also known as cyanide salts) contain cyanide in the anion form (CN⁻) and are used in numerous industries, such as metallurgy, photographic developing, plastic manufacturing, fumigation, and mining. Organic compounds that have a cyano group bonded to an alkyl residue are called nitriles. For example, methyl cyanide is also known as acetonitrile (CH₃CN). Hydrogen cyanide (HCN) is a colorless gas at standard temperature and pressure with a reported bitter odor. Cyanogen gas, a dimer of cyanide, reacts with water and breaks down into the cyanide anion. Many plants, such as Manihot spp. (cassava), Linum spp., Lotus spp., Prunus spp., Sorghum spp., and Phaseolus spp., contain cyanogenic glycosides. Iatrogenic cyanide poisoning may occur during use of nitroprusside as a vasodilator given to reduce blood pressure and afterload. Each nitroprusside molecule contains five cyanide molecules, which are slowly released in vivo. If endogenous sulfate stores are depleted, as in the malnourished or postoperative patient, cyanide may accumulate even with therapeutic nitroprusside infusion rates (2 to 10 mcg/kg/min).

Toxic Effects

Hydrocyanic acid binds to hemoglobin. The hydrocyanic acid bound in the erythrocyte is in equilibrium with free hydrocyanic acid in the serum at a ratio of 10 : 1. Cyanide in serum readily crosses all biological membranes and avidly binds to heme iron (Fe³⁺) in the cytochrome a-a₃ complex within mitochondria.715,803 When bound to cytochrome a-a₃, cyanide is a competitive inhibitor that causes decoupling of oxidative phosphorylation. Patients exposed to toxic concentrations of cyanide exhibit rapid onset of symptoms typical of cellular hypoxia—flushing, headache, tachypnea, dizziness, and respiratory depression—which progress rapidly to coma, seizures, complete heart block, and death if the dose is sufficiently large.

Hydroxycobalamin or the cyanide antidote kit should be administered as soon as cyanide poisoning is suspected. Hydroxocobalamin, a vitamin B₁₂ precursor, is a metalloprotein with a central cobalt atom that complexes cyanide, forming cyanocobalamin (vitamin B₁₂). Cyanocobalamin is eliminated in the urine or releases the cyanide moiety at a rate sufficient to allow detoxification by rhodanese. The cyanide antidote kit contains amyl nitrite, sodium nitrite, and sodium thiosulfate. Thiosulfate donates the sulfur atoms necessary for rhodanese-mediated cyanide biotransformation to thiocyanate. The mechanism of nitrite is less clear. The traditional rationale relies on the ability of nitrite to generate methemoglobin. Because cyanide has a higher affinity for methemoglobin than for cytochrome a₃, cytochrome oxidase function is restored.

Analytical Methods

Following microdiffusion, whole blood CN⁻ is measured by photometric analysis ¹⁷⁹ or by headspace gas chromatography.¹¹¹

With the spectrophotometric method, a sealed, two-well microdiffusion cell is used to separate hydrocyanic acid from blood by mixing a sample of whole blood with strong acid in a sealed chamber and allowing the hydrocyanic acid gas generated to be absorbed into a strong base located in another part of the sealed chamber. One well of the cell contains the blood specimen and strong acid (unmixed until the cell is sealed), and the other well contains a strong base to absorb the hydrocyanic acid gas. After the hydrocyanic acid is collected in the aqueous base medium, pyridine, barbituric acid, and chloramine-T are added to generate a red complex, with the intensity of the color proportional to the concentration of CN⁻. A good quality spectrophotometer is required to measure the absorbance. Quick and easy methods for plasma thiocyanate analysis have been described.⁸⁰²

Methemoglobin-Forming Agents

The heme iron in hemoglobin is normally present in the ferrous state (Fe²⁺). When oxidized to the ferric state (Fe³⁺), methemoglobin is formed, and this form of hemoglobin cannot bind oxygen (Figure 35-1). The principal physiologic system that maintains hemoglobin iron in the reduced state is nicotinamide adenine dinucleotide (NADH)-methemoglobin reductase. The NADH for this enzyme is supplied by normal glycolysis (Embden-Meyerhof pathway). A minor pathway for methemoglobin reduction involves nicotinamide adenine dinucleotide phosphate (NADPH)-methemoglobin reductase, and the NADP for this enzyme reaction is derived from the hexose-monophosphate shunt. Congenital methemoglobinemia may result from a deficiency of NADH-methemoglobin reductase or, more rarely, from hemoglobin variants (hemoglobin M) in which heme iron is both more susceptible to oxidation and more resistant to reduction by the methemoglobin reductase system.

Figure 35-1 Enzymatic pathways for methemoglobin reduction.

Toxic Effects

An acquired (toxic) methemoglobinemia may be caused by various drugs and chemicals (Table 35-4). The normal percentage of methemoglobin is <1.5% of total hemoglobin. In otherwise healthy individuals, methemoglobin percentages up to 20% may cause slate-gray cutaneous discoloration, cyanosis, and chocolate-brown blood. Percentages between 20% and 50% may cause dyspnea, exercise intolerance, fatigue, weakness, and syncope. More severe symptoms of dysrhythmias, seizures, metabolic acidosis, and coma are associated with methemoglobin percentages of 50 to 70%, and >70% may be lethal.^659,858 All of these symptoms are a consequence of hypoxia associated with the diminished O₂ content of the blood, and with a decreased O₂ dissociation from hemoglobin species in which some, but not all, subunits contain heme iron in the ferric state (i.e., shift of dissociation curve to the left). The PO₂ is normal in these patients, and therefore so is the calculated hemoglobin oxygen saturation. Thus, a normal PO₂ in a cyanotic patient is a significant indication for the possible presence of methemoglobinemia. Direct measurement of methemoglobin is important in these cases and may be performed by the manual spectrophotometric method of Evelyn and Malloy²³⁵ or by automated multiwavelength measurements with a co-oximeter (see section on “Carbon Monoxide”).

TABLE 35-4

Examples of Acquired Causes of Methemoglobinemia

Drugs	Chemical Agents
Amyl nitrite	Aniline
Benzocaine	Amyl nitrite
Chloroquine	Butyl nitrite
Dapsone	Chlorobenzene
Nitroglycerin	Naphthalene
Phenacetin	Nitrates
Phenazopyridine	Nitrites
Primaquine	Nitrophenol
Sulfonamides	Nitrous oxide

Specific therapy for toxic methemoglobinemia involves the administration of methylene blue, which acts as an electron transfer agent in the NADPH-methemoglobin reductase reaction, thereby increasing the activity of this system several-fold.649,658 Methylene blue and sulfhemoglobin cause spectral interference in the measurement of methemoglobin with some co-oximeters^412,875 but not with the Evelyn-Malloy method.⁴¹²

Analytical Methods

Methemoglobin is measured in blood manually,235,412 or by automated multiwavelength measurements with a co-oximeter.³²⁵ Methemoglobin interferes with the noninvasive pulse oximetry method, measuring the absorbance of light at 660 nm (oxyhemoglobin) and 940 nm (deoxyhemoglobin). Because methemoglobin is not stable at room temperature, specimens should be kept on ice or refrigerated but not frozen.⁴¹² The stability of methemoglobin at 4 °C has not been well studied. Some sources indicate significant decreases in methemoglobin concentration after 4 to 8 hours,⁴⁷⁰ whereas others report little or no change after 24 hours.⁴¹² Freezing results in an increase in methemoglobin concentration.⁴¹²

Alcohols of Toxicologic Interest

Several alcohols are toxic and medically important.^197A They include ethanol, methanol, isopropanol, acetone, and ethylene glycol.

Ethanol

Ethanol is the most widely used and often abused chemical substance. Consequently, measurement of ethanol is one of the more frequently performed tests in the toxicology laboratory. Although less frequently encountered, it is important to include methanol, isopropanol, acetone (a metabolite of isopropanol), and ethylene glycol in a test battery for alcohols for proper evaluation of the acutely intoxicated patient.

The principal pharmacologic action of ethanol is central nervous system (CNS) depression. CNS effects vary depending on the blood ethanol concentration (Table 35-5) but are also heavily influenced by an individual’s tolerance. Symptoms vary from euphoria and decreased inhibitions, to increased disorientation and incoordination, and then to coma and death. A blood alcohol concentration of 80 mg/dL (0.08%) has been established as the per se limit for operation of a motor vehicle in most countries.

TABLE 35-5

Stages of Acute Alcoholic Influence/Intoxication

Because of many factors, not all individuals experience the same degree of CNS dysfunction at similar blood alcohol concentrations. Moreover, the CNS actions of ethanol are more pronounced when the blood ethanol concentration is increasing (absorptive phase) than when it is declining (elimination phase), in part because of the phenomenon of acute tolerance.²⁹¹ In addition, heavy alcohol use leads to a more chronic form of tolerance. When consumed with other CNS depressant drugs, ethanol exerts a potentiation or synergistic depressant effect. This can occur at relatively low alcohol concentrations, and numerous deaths have resulted from combined ethanol and drug ingestion.²⁷⁴

The pharmacologic mechanisms for the CNS depressant actions of ethanol are complex and incompletely understood, but probably involve both enhancement of major inhibitory neurons and impairment of excitatory neurons. The principal CNS inhibitory neuronal system is mediated by the neurotransmitter γ-aminobutyric acid (GABA). When GABA binds to its postsynaptic receptor subtype GABA_A, this oligomeric ion-gated complex “opens” to allow inward flux of Cl, leading to membrane hyperpolarization and subsequent decreased electrical response. This GABA-mediated inhibitory response is enhanced by ethanol and sedative, hypnotic, and anesthetic agents, including barbiturates, benzodiazepines, and volatile anesthetics.²⁴⁸ Neuronal nicotinic acetylcholine receptors also may be prominent molecular targets of alcohol.⁵⁷⁶ Both enhancement and inhibition of nicotinic acetylcholine receptor function have been reported depending on receptor subunit concentration and the concentrations of ethanol tested. Ethanol also inhibits the function of the N-methyl-D-aspartate (NMDA)- and kainate-receptor subtypes; AMPA receptors are largely resistant to alcohol.¹¹⁹

The aforementioned chronic tolerance to ethanol is considered to be mediated by ethanol-induced increased responsiveness and upregulation in the synthesis of NMDA receptors, attained by concomitant downregulation and desensitization through phorphorylation of GABA_A and glutamate receptors.248,452,819 Largely because of these adaptive changes, abrupt withdrawal from chronic, heavy ethanol use leads to a physical abstinence syndrome that has prominent features of CNS excitation. Included among these withdrawal symptoms are anxiety, irritability, insomnia, muscle tremor and cramps, seizures, hallucinations, and increased temperature, blood pressure, and heart rate.

Ethanol is metabolized principally by liver alcohol dehydrogenase to acetaldehyde, which is subsequently oxidized to acetic acid by aldehyde dehydrogenase (Figure 35-2). The rate of elimination of ethanol from blood approximates a zero-order process. This rate varies among individuals, averaging about 15 mg/dL/h for males and 18 mg/dL/h for females.^210,211 At both low (<20 mg/dL)⁸¹³ and high (>300 mg/dL) ethanol concentrations, elimination becomes more nearly first-order; it is accelerated at high concentrations.⁸¹ The elimination rate is also influenced by drinking practices (e.g., alcoholics have increased elimination rates caused by enzyme induction).⁸⁴⁷

Figure 35-2 Metabolism of ethanol.

Ethanol is a teratogen, and alcohol consumption during pregnancy can result in the birth of a baby with fetal alcohol spectrum disorder (FASD). FASD is an umbrella term that describes the variety of effects that can occur in an individual whose mother drank alcohol during pregnancy (http://www.nofas.org/accessed June 13, 2011). These effects may include physical, mental, behavioral, and/or learning disabilities with possible lifelong implications and are 100% preventable when a woman completely abstains from alcohol during her pregnancy.

Methanol

Methanol is used as a solvent in several commercial products, as a constituent of antifreeze and window cleaning fluids, and as a component of canned fuel. It may be consumed intentionally by alcoholics as an ethanol substitute or accidentally when present as a contaminant in illegal whiskey. Accidental ingestions have occurred in children.

The CNS effects of methanol are substantially less severe than those of ethanol. Methanol is oxidized by liver alcohol dehydrogenase (at about one tenth the rate of ethanol) to formaldehyde. Formaldehyde in turn is rapidly oxidized by aldehyde dehydrogenase to formic acid, which may cause serious acidosis and optic neuropathy, resulting in blindness or death.513,534 Serum formate concentrations correlate better with the degree of acidosis and the severity of CNS and ocular toxicity than do serum methanol concentrations.⁷¹⁸ Therefore, some investigators recommend the measurement of serum formate to assess the severity of toxicity and to guide appropriate therapy in cases of methanol ingestion. The mainstay of therapy for methanol toxicity includes the administration of ethanol or fomepizole as a competitive alcohol dehydrogenase inhibitor, either folate or folinic acid, and dialysis.

Isopropanol and Acetone

Isopropanol is readily available to the general population as a 70% aqueous solution for use as rubbing alcohol. It has about twice the CNS depressant action as ethanol, but it is not as toxic as methanol.⁵⁸ Isopropanol has a short t_1/2 of 2.5 to 3.0 hours,⁵⁸ as it is rapidly metabolized by alcohol dehydrogenase to acetone, which is eliminated much more slowly (t_1/2, 3 to 6 hours).⁴⁷ Therefore concentrations of acetone in serum often exceed those of isopropanol during the elimination phase following isopropanol ingestion. Acetone has CNS depressant activity similar to that of ethanol, and because of its longer t_1/2, it may prolong the apparent CNS effects of isopropanol.^47,58 Supportive care is the mainstay of treatment, with rare reports of dialysis in severe intoxication.

Ethylene Glycol

Ethylene glycol, present in antifreeze products, may be ingested accidentally or for the purpose of inebriation or suicide. Because it tastes sweet, some animals are attracted to it. Veterinarians are often familiar with ethylene glycol toxicity because of cases involving dogs or cats that drank radiator fluid.

Ethylene glycol itself is relatively nontoxic, and its initial CNS effects resemble those of ethanol.³⁸⁶ However, metabolism of ethylene glycol by alcohol dehydrogenase (ADH) results in the formation of numerous acid metabolites, including lactate, oxalic acid and glycolic acid.^53,386 These acid metabolites are responsible for much of the toxicity of ethylene glycol.^334,371 Serum concentrations associated with death from ethylene glycol ingestion have been observed to vary from 0.06 to 4.3 g/L,^53,651 highlighting the lack of correlation between ethylene glycol concentration and severity of toxicity. It is thus impossible to define a serum ethylene glycol concentration associated with a high probability of death. The serum concentration of glycolic acid correlates more closely with clinical symptoms and mortality than does the concentration of ethylene glycol.^334,651 Because of the rapid elimination of ethylene glycol (t_1/2, 2 to 5 hours),⁵³ its serum concentration may be low or undetectable at a time when glycolic acid remains elevated.^257,334,651 Thus the determination of ethylene glycol and glycolic acid provides useful clinical and confirmatory analytical information in cases of ethylene glycol ingestion. The mainstay of therapy for ethylene glycol toxicity includes administration of ethanol or fomepizole as a competitive alcohol dehydrogenase inhibitor and dialysis.