Body Fluids

Introduction

The analysis of drugs of abuse in unconventional bio-fluids is a rapidly expanding area of toxicological science. Blood and urine have long been the dominant matrices for drugs of abuse detection. Incorporation of drugs into these biofluids is well understood and the analysis and interpretation of these findings has become routine for most commonly abused drugs. Recently, however, interest has shifted towards alternative specimens that may offer distinct advantages over conventional bio fluids. These benefits may include long-term, cumulative information on drug use, or the convenience of noninvasive sample collection.
Following absorption of a drug, distribution, metabolism and excretion pathways account for the sequential appearance of drug and metabolite in different tissues and fluids. Physicochemical characteristics of the drug and biofluid can be used to rationalize or predict the appearance of drug in a particular biofluid or compartment of the body. The pKa, lipid solubility, protein binding and biofluid composition determine the extent to which the drug is present. Transfer of drug from the circulating blood plasma (pH 7.4) to another biofluid involves transport across membranes, which are effective barriers against ionized, highly polar compounds. Following penetration of the membrane and transfer into the biofluid, the pH differential may result in ionization of the drug, restricting further mobility. Accumulation of the drug in this way is commonly referred to as ‘ion trapping’.
The presence of drug in a biofluid indicates exposure to the substance, perhaps unwittingly. Drugs of abuse are generally self-administered by oral, intrana-sal, intravenous or smoked routes. Passive smoke inhalation and ingestion of certain foodstuffs, such as poppy seeds or hemp oil, may result in detectable amounts of drug in certain biofluids. Unconventional drug exposure, such as that which occurs in utero or in nursing mothers, has also necessitated the use of alternative biological specimens. Our discussion of these unconventional samples is limited to amniotic fluid, breast milk, saliva, semen and sweat. The growing concern regarding the effects of drugs on health and human performance has highlighted alternative biofluid analysis in multiple forums: law enforcement, probation, parole, drug compliance and abstinence programs, employment, health and insurance, among others. Combined interest and growing expertise in alternative biofluid analysis has increased momentum in the field. The relative expectations, limitations and interpretation vary widely between these different applications. Recent advances in the analysis of alternative biofluids and matrices have accelerated drugs of abuse detection in some of these areas. Of critical importance is the choice of biofluid for analysis, as each may provide unique chemical and pharmacolo-gic information. There are a great many factors which may influence the choice of biofluid for drug analysis and some of these are listed in Table 1. The interpretive value, advantages and disadvantages of each biofluid are summarized in Table 2.

Biofluids

Amniotic fluid

Increased use of illegal drugs by expectant mothers has led to the need for prenatal toxicological testing. Exposure to drugs of abuse has demonstrated effects

Table 1 Factors influencing the choice of a biofluid

Sample collection
Invasiveness
Risk of infection, complication and hazards
Protection of privacy
Ease and speed of collection
Training of personnel (medical/nonmedical)
Likelihood of adulteration
Contamination
Volume of specimen
Analysis
Qualitative or quantitative Window of detection
Drug concentration/accumulation in biofluid
Parent drug or metabolite(s)
Stability of drug analytes
Biofluid storage requirements
Pretreatment of specimen
Limitations of the matrix
Likelihood of interferences
Inter-and intrasubject variability of the matrix
Use of existing analytical procedures
Speed of analysis
Personnel training requirements
Appropriate cut-off concentrations

Interpretation

Pharmacologic effects Behavioral effects Indicator of recent drug use (hours) Short-term drug exposure (days) Long-term drug exposure (weeks) Forensic defensibility on both the fetus and the neonate. Higher rates of fetal distress, demise, growth retardation and adverse neu-rodevelopment have been documented. Cocaine, heroin, amphetamines, and nicotine have been associated with impaired fetal growth and acute withdrawal syndromes. The greatest risk of neonatal abstinence syndrome occurs with narcotic drugs but has also been observed with cocaine and amphetamines. In a 1997 study conducted in Michigan, in which gestational drug exposure was measured in nearly 3000 newborns, as many as 44% of neonates tested positive for drugs. Of these, 30.5% tested positive for cocaine, 20.2% for opiates and 11.4% for cannabinoids.
Anatomy/physiology Amniotic fluid, which is produced by cells that line the innermost membrane ofthe amniotic sac (amnion), is the liquid that surrounds and protects the embryo during pregnancy. This fluid cushions the fetus against pressure from internal organs and from the movements of the mother. Production of fluid commences the first week after conception and increases steadily until the 10th week, after which the volume of fluid rapidly increases.

Table 2 Advantages and disadvantages of biofluids

Amniotic fluid, which may total about 1 liter at 9 months, contains cells and fat that may give the liquid a slightly cloudy appearance. Maternal drug use may expose the fetus to the drugs and their metabolites, the toxicity and effects of which may be detrimental to the normal development of the fetus. Amniotic fluid is constantly circulated, being swal- lowed by the fetus, processed, absorbed and excreted by the fetal kidneys as urine at rates as high as 50mlh_1. This circulation of fluid continuously exposes the fetus to compounds that may absorb in the gut or diffuse through fetal skin in the early stage of development. The encapsulation of the fetus in this fluid may prolong exposure to harmful drugs or metabolites. Reduced metabolic or enzyme activity in pregnant women or the developing fetus may decrease drug metabolism, and therefore, compound the risks. The pharmacokinetics of drug disposition in utero varies from drug to drug, and the acute and chronic effects that may result are the topic of continuing research. In addition, indirect effects of maternal drug use, such as those on uterine activity and maternal circulation, may be as damaging as the drug itself.
Sampling The collection of amniotic fluid (amniocentesis) usually takes place between the 16th and 20th weeks of pregnancy. The liquid is usually collected to test for fetal abnormalities. The presence of illicit drugs or their metabolites in amniotic fluid suggests that the fetus has been exposed to these substances via maternal blood circulation. A maternal serum sample taken at the same time as the test may provide complementary toxicological data and help assess the relative risk to the fetus. Of the alternative biofluids described here, amniocentesis is perhaps the most invasive specimen collection procedure. Prior to amniocentesis, an ultrasound scan is used to determine the position of the fetus. After receiving a local anesthetic, a needle is inserted through the abdomen into the womb where there is the least chance of touching the placenta or the fetus. Although complications are rare, miscarriage occurs in approximately 1% of women. Typically 5-30 ml of amniotic fluid is removed; it is slightly alkaline in nature. However, the pH of amniotic fluid decreases during pregnancy due to fetal urination, reaching near neutral pH at full term. Amniotic fluid, which is 99% water, contains dilute plasma components, cells and lipids.
Analysis Drugs present in amniotic fluid can be analyzed using well-established techniques that are routinely used for blood and urine. A variety of analytical methods that have been employed are listed in Table 3. Drugs of abuse have been detected in human amniotic fluid using screening tests, such as immunoassays, and confirmatory methods such as gas chromatography-mass spectrometry (GC-MS). Few interferences are encountered with amniotic fluid due to its high water content. Sample pretreatment may not be necessary before the immunoassay screening test. However, confirmation of presumptive positive results by GC-MSrequires isolation of the drug by liquid-liquid or solid-phase extraction.
Interesting/relevant findings A number of maternal, fetal and placental factors may affect fetal drug exposure. Of these, binding to serum proteins in the maternal and fetal circulation and fetal elimination are particularly important. Accumulation of unmetabo-lized drugs in amniotic fluid occurs as a result of fetal renal excretion, and perhaps via the fetal membranes. However, elimination of the drug by the fetal liver may be less important than in the adult. Polar conjugated drug metabolites may accumulate in the fetus or amniotic fluid owing to limited placental transfer.

Table 3 Methods of analysis of drugs in body fluids

Purification
Liquid-liquid extraction Solid-phase extraction Supercritical fluid extraction
Drug detection by immunochemical techniques
Enzyme immunoassay Enzyme-linked immunosorbent assay Fluorescence polarization immunoassay Radioimmunoassay
Drug identification by chromatographic techniques
Capillary electrophoresis
Gas chromatography
Gas chromatography/mass spectrometry
Gas chromatography/mass spectrometry/mass spectrometry
High-performance liquid chromatography
Liquid chromatography/mass spectrometry
Thin-layer chromatography
Small lipid-soluble drugs can rapidly diffuse across the placental barrier, producing similar drug concentrations in amniotic fluid and fetal plasma. Larger, water-soluble compounds, which are transferred more slowly, are incorporated into the amniotic fluid from fetal urine. Basic drugs may accumulate in the amnion due to ion trapping, resulting in drug concentrations in excess of those found in fetal or maternal plasma.
Maternal cocaine abuse is known to decrease uterine blood flow, cause fetal hypertension, cardiovascular effects and a deficiency of oxygen in the arterial blood. Retarded growth, congenital abnormalities, withdrawal syndrome, and cerebral hemorrhage or infarction have also been observed. Prenatal drug tests have shown amniotic fluid to contain cocaine and its major metabolite, benzoylecgonine, at concentrations up to 24 and 836ngml_1, respectively, although concentrations as high as 3300 and 1600ngml_1 have been measured following maternal death. The fetus is exposed to cocaine by maternal circulation as well as placental and fetal metabolism. After crossing the placental barrier by simple diffusion, the drug distributes between fetal and maternal blood. The amniotic sac and its contents serve as a deep compartment, with restricted, slow equilibrium between adjacent compartments. As a result, amniotic fluid inside this protective sac may expose the fetus to potentially harmful drugs or metabolites. Ion trapping and under-developed renal function may result in an accumulation of the drug in fetal blood, thus compounding the risk.
Narcotic analgesics are reported to cross the placental barrier rapidly, but at physiological pH, when the drug is mostly charged, the concentration of the drug in the amniotic fluid is expected to be lower than that of the maternal plasma. Due to their high lipid solubility and lack of ionization, benzodiazepines readily cross the placenta. However, drug concentrations in the amniotic fluid remain low due to extensive protein binding in the maternal plasma and minimal renal excretion by the fetus. Drugs of abuse and their metabolites that have been detected in amniotic fluid and other matrices are summarized in Table 4.
Interpretation of toxicological findings is limited because drug dose, route, and time of administration are usually unknown. Long-term implications of prenatal drug exposure are limited and many consequences of fetal drug exposure are still unknown. Despite adequate understanding of the maternal consequences of drug abuse, fetal consequences for many drugs are poorly understood and this is a challenging area of maternal-fetal medicine.

Breast milk

The prevalence of drug abuse among pregnant women throughout urban America is reported to be between 0.4 and 27%, depending on geographical location. In the last two decades there has been nearly a threefold increase in the number of women who breast-feed their infants. These factors together significantly increase the likelihood of drug exposure that may result in acute toxicity or withdrawal syndrome. Although it is generally accepted that drug exposure of the infant from milk is less harmful than in utero exposure of the fetus, the overall effect on health and development is largely unknown. According to the American Academy of Pediatrics, amphetamine, cocaine, heroin, marijuana and phencyclidine (PCP) are considered unsafe for nursing mothers and their infants.
Anatomy/physiology The female breast consists of 15-20 lobes of milk-secreting glands embedded in the fatty tissue. During pregnancy, estrogen and progesterone, secreted in the ovary and placenta, cause the milk-producing glands to develop and become active. The ducts of these glands have their outlet in the nipple; by midpregnancy, the mammary glands are prepared for secretion. Colostrum, a creamy white to yellow premilk fluid, may be expressed from the nipples during the last trimester of pregnancy. This fluid, which is a rich source of protein, fat, carbohydrate and antibodies, is replaced with breast milk within 3 days of delivery of the fetus and placenta. Proteins, sugar and lipids in the milk provide initial nourishment for the newborn infant. The production of between 600 and 1000 ml of milk per day by the milk-secreting cells is stimulated by the pituitary hormone, prolactin.
Sampling/analysis Fluid is collected using a special device such as a breast-milk pump, after which established analytical techniques may be used to detect drugs of abuse. Breast milk, which contains protein (1%), lipid (4%), lactose (7%) and water (88%), varies in pH between 6.35 and 7.35. However, the high lipid content of milk may interfere or decrease the extraction efficiency or recovery of drug. Additional washing with nonpolar solvents such as hexane may be necessary to remove excess lipids before chromatographic analyses. The effect ofnatural emulsifying agents in breast milk, which have detergentlike activity, may interfere with antibody-antigen reactions which take place in immunoassay screening tests. The daily variation of breast milk composition, combined with drug dose and time of administration relative to the expression of milk, is likely to affect the amount of drug present and the effect on the infant. Metabolic function and overall health of the nursing mother also play an important role. The concentration of drug found in the breast milk is subject to both within- and between-subject variation, further confounding attempts to generalize infant risk assessment. The lipid content of the milk varies not only daily, but also during a single feed; the latter portion of expressed milk may contain a severalfold increase in fat, which in turn may increase or decrease the concentration of a particular drug.

Table 4 Partial listing of drugs and metabolites detected in biofluids

Interesting/relevant findings The transfer of drug into the milk depends on metabolism, protein binding and the circulation of blood in the mammary tissue. Passive diffusion transports the drug across the mammary epithelium into the milk. The mildly acidic pH of breast milk tends to trap weakly basic drugs, particularly those of low molecular weight, which can diffuse fairly rapidly through small pores in the semipermeable membrane. Although drugs that are extensively protein bound may not readily pass into the milk, emulsified fats serve to concentrate lipid-soluble drugs such as marijuana and PCP. For this reason, PCP has been detected in breast milk for as long as 41 days after cessation of maternal drug use. Not all lipophilic drugs produce deleterious effects in the nursing infant. Fentanyl, which was found to concentrate in lipid-rich colostrum at much higher concentration than maternal serum, has a low oral bioavailability, somewhat minimizing the risk to the child. However, inactive conjugated metabolites present in the mother’s milk may undergo reactivation by deconjugation in the gastrointestinal tract of the infant.
Opiate addiction and withdrawal symptoms have been reported in infants receiving milk from substance-abusing mothers. Low oral doses of morphine to nursing mothers may produce drug concentrations in milk as high as 100ngml_1. Methadone maintenance (50mgday_1) of a drug-dependent nursing mother produced breast-milk concentrations between 20 and 120ngml_1 in the first 24 h after the dose, substantially lower than maternal plasma concentration. Benzodiazepines, which are excreted in the milk, can accumulate in the infant owing to underdeveloped metabolic and excretory function. Water-soluble drugs appear least likely to partition into the milk and are less likely to accumulate in the infant. However, stimulants such as amphetamine have been detected in milk at concentrations 3-7 times higher than those found in maternal plasma owing to ion trapping of basic drugs. Cocaine is also believed to preferentially partition into milk, increasing the likelihood of infant toxicity.
Despite the fact that some lactating women have reported using marijuana, little is known of the effect of this drug on the infant. The concentration of the active ingredient, A9-tetrahydrocannabinol (THC), is reported to be as much as eightfold higher in breast milk than maternal plasma. Metabolites of THC and the parent drug were measured in breast milk at concentrations as high as 340ngml_1 in a chronic marijuana smoker.

Saliva

Saliva, which can be collected noninvasively, conveniently and without invasion of privacy, is perhaps one of the most appropriate biofluids for on-site drug testing. The active constituent of marijuana, which is the most widely abused drug in America, was detected in the saliva of 9% of motorists who displayed erratic driving behavior. Law enforcement agencies are particularly interested in saliva testing as a complementary tool for identifying impaired drivers with roadside drug testing devices. This approach, which is already in widespread use for employment, health and insurance drug testing, is rapidly gaining in popularity.
Anatomy/physiology Saliva is the clear viscous fluid that is secreted by salivary glands and mucous membranes that line the mouth. It serves to lubricate the oral cavity, assists in the swallowing of food and facilitates our sense of taste. This biofluid, which is comprised of 90% water, contains only about 0.3% protein. The remainder is made up of electrolytes, mucin, urea, lipids, digestive enzymes, white blood cells and debris from the lining of the mouth. Three pairs of glands secrete saliva via ducts in the mouth. Each gland consists of thousands of saliva-secreting sacs and a network of canals that transport the fluid into the main ducts in the mouth. The parotid glands, which are the largest of the salivary glands, lie inside the cheek, just below and in front of the ear. These glands secrete serous fluid that is derived from blood plasma. Serous fluid and mucin are secreted by the sublingual glands, located on the floor of the mouth beneath the tongue, and by the submandibular glands, which are just below the jaw on the front of the neck. Mixed saliva is comprised mostly of submandibular secretions (71%). Parotid and sublingual glands are responsible for 25% and 4% of the remaining volume. This biofluid, which may be secreted at rates of 1.51 day-1, has an average pH of about 6.8.
Sampling Mixed saliva is collected noninvasively by expectoration, aspiration by vacuum, or by saturation of a cotton swab. Secretions from a specific gland may be collected using a special device or by cannulation, but this is uncommon. Salivation for the purpose of specimen collection may be increased by chewing or sucking an inert substance, such as PTFE tape, parafilm (wax) or a rubber band. It is necessary to ensure that no adsorption takes place between the drug and the chewed substance. Acidic candy or citric acid has also been used to stimulate glandular secretions. Care must be taken that residual food or drink in the mouth does not interfere with the analysis. It is possible that interfering substances could be placed in the mouth to adulterate the saliva, but there have been no known reports as yet.
Analysis Saliva contains few interferences and is amenable to most types of toxicological analyses. It is possible that endogenous enzymes in saliva could interfere with colorimetric or fluorescence immuno as-says that detect drug indirectly from the activity of a labeled enzyme. If saliva is collected by expectoration or free flow, sample pretreatment may not be necessary before immunochemical testing. Dilution of the sample with buffer may be sufficient, although centrifugation may be used to remove any solid debris. Liquid-liquid or solid-phase extraction is necessary before chromatographic analysis.
Interesting/relevant findings The detection times for most drugs in saliva are fairly short relative to urine and do not provide a history or long-term profile of drug use. However, the parent drug may be detected for several hours following drug use at concentrations proportional to those measured in plasma. Saliva, which is an ultrafiltrate of interstitial fluid, contains the unbound fraction of drug, which is pharmacologically active. As a result, saliva tests may allow pharmacologic interpretation of results, based on the concentration of drug that was circulating at the time of the test.
Salivary ducts in the mouth are separated from the systemic circulation by a layer of epithelial cells. Transfer of drug from the plasma into saliva requires transport across the lipid membrane of these cells. Passive diffusion is perhaps the most important route of passage for most drugs, although ultrafiltration and active secretion from the blood may also occur. Transfer across cell membranes is restricted by size, ionization and macromolecular binding. Therefore the pH of the saliva, pKa of the drug and plasma protein binding of the drug, strictly control the passage of drug into this biofluid. At fixed salivary pH (6.8), theoretical estimates of the partition of a drug between the saliva and plasma (S:P ratios) can be predicted from the Henderson-Hasselbach equation. Increasing the saliva flow rate can increase pH from 5.5 to 7.9, greatly affecting the disposition of drugs that are ionized at normal plasma pH. Stimulation of saliva flow by chewing waxed film or sour candy may decrease the concentration of weakly basic drugs, such as cocaine (pKa 8.7), by severalfold. This weakens the reliability of saliva testing for pharmacologic interpretation and accounts for discrepancies in S:P ratios between authors. Although salivary pH is the principal determining factor for ionizable drugs, compounds, which are uncharged in plasma, are unaffected. In contrast to cocaine, the concentration of zwitterionic metabolite, benzoylecgonine (pKa 2.3, 11.2), is unchanged by salivary pH.
Passive diffusion through lipid membranes requires the drug to be in a lipid-soluble form. Glucuronides or polar metabolites may be too hydrophilic to cross the membranes. Saliva has a minimal protein binding capacity compared with plasma. In order to be retained in saliva, the drug must be water-soluble, a property largely augmented by ionization of the drug. This transformation prevents back-diffusion of the drug into the plasma. Drugs of abuse that are strongly basic in nature may be preferentially distributed into saliva, as is the case with amphetamine. Salivary concentrations may exceed those measured in plasma by as much as two- to threefold. These types of drugs may be detectable for slightly longer periods of time, up to 2 days for amphetamine and methamphetamine.
The range of saliva pH is typically much narrower (6.5-7.2) than that of urine (4.5-8.0), suggesting that this alternative biofluid may be of diagnostic value for drugs whose excretion is heavily dependent on urinary pH, such as the amphetamines or PCP.
Numerous other drugs have been measured in saliva, including a number of opiates and synthetic opioids. Methadone, which is used to maintain opioid-dependent individuals, has been measured in saliva, where its concentration correlates well with that in plasma. Phenobarbital, amobarbital and other barbiturates have been detected in saliva, typically with S:P ratios of about 0.3. Other sedative drugs, methaqualone and meprobamate, have S:P ratios of 0.1 and 1, respectively. Owing to their widespread abuse, a number of benzodiazepines, including diaze-pam, have been investigated. S:P ratios were typically low (0.04 or less), which is very much lower than predicted. This is likely to be the result of the extensive plasma protein binding that takes place with these compounds.
Cocaine concentrations in saliva have been shown to correlate well with plasma concentrations and behavioral effects. Metabolites of cocaine are present in very low concentration in saliva. Inconsistencies in S:P ratios have been attributed to contamination of saliva following intranasal and smoked routes of administration. Peak cocaine concentrations in saliva increased from 428-1927ng ml-1 after a single oral dose to between 15 000 and >500 000ngml-1 after smoking. Contamination of the oral cavity following these types of exposure precludes pharmacologic or behavioral interpretation of results. However, these results may indicate recent exposure to the drug, which could be of equal forensic importance.
Elevated THC concentration (50-1000ngml-1) after smoking marijuana is also the result of oral contamination. Shortly after exposure, the concentration rapidly declines in a dose-dependent manner similar to that of plasma. However, some cannabi-noids can be detected in saliva for longer periods than plasma, suggesting that some drugs are actually sequestered in the buccal cavity during smoking.

Semen

Analysis of semen for drugs of abuse has not been widely adopted. Privacy issues surrounding collection of this biofluid restrict its widespread use. Despite limited reports of drug detection in semen, it is known that certain drugs, such as cocaine, can influence the characteristics of the spermatozoa, causing decreased motility and anatomical abnormalities. It has also been shown that cocaine binds to sperm with high affinity, suggesting a unique mode of transport to the ovum. Although the concentration of drug in semen is not sufficient to elicit a response in a sexual partner, it has been suggested that insemination of drug-laden sperm into the egg could result in abnormal development.
Anatomy/physiology Semen is the viscous fluid released from the male upon ejaculation. The fluid contains spermatozoa as well as auxiliary sex gland secretions. Two seminal vesicles, the prostate gland and the bulbourethral glands contribute 95% of the gelatinous secretion. Seminiferous tubules of the testes contribute less than 5% of the seminal fluid volume, which is typically between 3 and 5 ml. Paired seminal vesicles, which are thin-walled, pear-shaped structures, secrete a thick, slightly alkaline fluid that mixes with the sperm as they pass into the ejaculatory ducts and urethra. These secretions, which are expelled when the seminal vesicles contract during orgasm, constitute about 60% of the seminal fluid volume. This secretion is rich in fructose, a sugar that stimulates the sperm to become mobile. The prostate gland, which sits just below the bladder, secretes a thin, milk-colored fluid, which accounts for about 30% of the total volume. This fluid helps activate sperm and maintain their motility. Bulbourethal glands located below the prostate produce mucuslike secretions that lubricate the terminal portion of the urethra, contributing about 5% of the total volume.
Sampling/analysis Following ejaculation, seminal fluid can be analyzed for drugs of abuse using techniques that are widely used for serum and other biofluids. Diluted seminal fluid may be used in immunoassay screening tests, and chromatographic analyses may be performed after extraction of the drug using conventional techniques. Semen pH typically ranges between 7.3 and 7.8, depending on the differential contribution of fluids. The overall alkalinity of the fluid helps protect spermatozoa from the acidic environment of the female reproductive tract.
Interesting/relevant findings Lipid solubility and pKa of the drug play an important role in the transport of drugs of abuse into seminal fluid. Ion trapping may be responsible for the transport of certain drugs from the seminal plasma to the genitourinary tract. Drug ionization depends on the pH difference between plasma (pH 7.4) and prostatic fluid (pH 6.6); prostatic fluid can trap basic drugs in the prostate. In contrast, the vesicular fluid, which is more alkaline in nature, is likely to contain much lower concentrations of these drugs. Cocaine was detected in semen at concentrations typically 60-80% of those measured in plasma, independent of the route of administration. Following a 25 mg dose of cocaine, parent drug and benzoylecgonine concentrations in semen were 45 and 81 ugg-1, respectively, at 1h. One study involving opioid-maintained individuals indicated that methadone was present in semen at concentrations in excess of those found in blood. Amphetamine concentrations in semen were reported to correlate well with whole blood concentration following drug exposure. However, the disposition of drugs of abuse in this biofluid is widely variable due to the heterogeneous nature of the seminal fluid.

Sweat

A study performed by the Michigan State Department of Corrections evaluated sweat and urine as possible drug-detection matrices. Participants in the study were prisoners, subject to residential or electronic monitoring, who might have had access to street drugs. Sweat patches, which were applied and worn for 7-14 days, were more effective indicators of drug use than repeated urinalysis every 3 days. Sweat analysis is becoming increasingly popular in drug compliance programs, such as rehabilitation, probation or parole, because it is a noninvasive, convenient means of specimen collection and requires less frequent testing.
Anatomy/physiology About 3 million tiny structures deep within the skin are responsible for the elimination of between 300 and 700 ml of fluid in a 24 h period. These glands secrete sweat, which is then transported through narrow passageways to the surface of the skin, whereupon evaporation has a cooling effect that helps maintain body temperature. The glands themselves are controlled by the autonomic nervous system, which is responsible for increased rates of secretion during times of anxiety or fear. Elimination of this body fluid, which occurs during normal breathing, is known as insensible sweat. Sensible sweat, which refers to perspiration that is actively excreted during stress, exercise or extreme temperature, may be eliminated at rates of 2-4 lh-1. About half of the total volume of sweat is eliminated from the trunk of the body. The remaining fluid is lost from the legs or upper extremities and head in approximately equal amounts. The fluid consists of water (99%), sodium chloride, phosphate, protein, urea, ammonia and other waste products. The average pH of sweat is about 5.8, but increased flow rates increase the pH to between 6.1-6.7 in a manner analogous to saliva.
Sweat is produced by two types of gland, eccrine and apocrine. The former are coiled, tube-like structures that are placed throughout the skin. These glands, which are concentrated in the palms of the hands and soles of the feet, open up directly to the surface through tiny pores in the skin. Apocrine glands, which become active only after puberty, are large, deep glands in the axillae, pubic and mammary regions. These produce cellular material as well as sweat, which results in the secretion of a more viscous fluid with a characteristic odor. Apocrine glands often open up into hair follicles before reaching the surface of the skin. Contamination of the exterior surface of the hair has been demonstrated as a result of this biofluid exchange.
Parts of the body’s surface are also bathed in sebum, an oily secretion of the sebaceous glands. Composed of keratin, fat and cellular debris, sebum mixes with sweat to form a moist, acidic film on the surface of the skin, which protects it from drying. This waxy lubricant, which consists mostly of fatty acids, maintains a pH of about 5 and is mildly harmful to bacteria and fungus, providing additional protection against infection.
Sampling Sweat is usually collected using an adhesive absorbent patch, which is placed on the surface of clean skin or by wiping the skin with a swab or gauze. Careful preparation of the skin is necessary before placement of a sweat patch to minimize external drug contamination or bacterial degradation of the drug once it has been retained. Occlusive sweat collection devices typically consist of an absorbent pad with an adhesive polyurethane exterior, similar to a waterproof bandage. These allow sweat to diffuse into the patch but prevent water or compounds from the environment from penetrating the device. In some instances, discomfort and inconvenience led to high rates of noncompliance among sweat patch users, resulting in tampering, loss of the patch or refusal to wear. Due to the relatively small volume (microliters) of insensible sweat secreted from a small absorbent area (typically 3 x 5 cm), patches are typically worn for several days on the outer portion of the upper arm or back. Sweating may be induced by occlusive wrapping or by diaphoretic treatment, but these efforts, which are more invasive, are not routinely used.
Occlusive bandages have been replaced with more advanced technology which improves user comfort. The PharmCheck® sweat patch is a specimen collection device for nonvolatile and liquid components of sweat, including drugs of abuse. Nonvolatile components in the environment cannot penetrate the pad externally, and a semipermeable membrane that covers the absorbent pad allows oxygen, water and carbon dioxide to diffuse through. This helps maintain healthy skin, improves user comfort and increases the likelihood of compliance. Salts, solids and drugs that pass through the skin are trapped in the absorbent pad, where they are temporarily stored in situ until the patch is removed.
Most skin wipes and patches contain a mixture of sweat and sebum, both of which are secreted from the surface of the skin. Most reports of drugs of abuse in sweat actually refer to the collection of the mixed matrix. As yet, there have been relatively few reports of drugs of abuse in sebum alone. This biofluid is typically collected from the forehead, which is rich in sebaceous glands, as are the face and scalp. Unlike sweat, which is predominantly water, fat-soluble drugs can be sequestered in sebum due to the high lipid content of this fluid.
Analysis Surface secretions of sweat and sebum must be extracted before an initial drug screen. Drugs, which are present in the absorbent collection material, are generally extracted with alcohol, buffer or a combination of both. Recovery of commonly abused drugs during this process is usually high (80% or more) and limits of detection by GC-MSare generally less than 1 ng per patch. The increased work necessary to analyze sweat patches is counteracted by the relative ease of sample collection and the ability to determine cumulative drug exposure over an extended period.
Interesting/relevant findings Quantitative analysis of drugs of abuse in sweat is rarely attempted because the volume of biofluid collected is generally uncertain. If an occlusive collection device is used, the volume of sweat can be estimated from the increased weight of the patch. Newer nonocclusive devices, which can be worn for longer periods, give no indication of specimen volume, but it has been suggested that the sample volume could be estimated indirectly from the concentration of sodium or lactate in the patch, both of which are excreted in sweat at relatively constant rates.
One of the principal advantages of sweat analysis is that it can provide information on long-term continuous exposure to drugs of abuse. Over a period of days, sweat saturates the absorbent pad, which retains the drugs. As most drugs are eliminated in the urine within 2-3 days of a single exposure, effective drug monitoring necessitates urinalysis every few days, which is inconvenient and labor intensive. Sweat patches, on the other hand, can be worn for as long as 2 weeks, are less likely to be adulterated and require drug testing to be performed only once, when the patch is removed.
Incorporation of drugs into sweat by transdermal migration or passive diffusion is favored by small, lipid-soluble compounds. Drugs that are highly bound to plasma proteins, such as benzodiazepines, are present in sweat at low concentrations. Basic drugs with high partition coefficients and pKa values close to sweat appear to be maximally excreted. The patch itself operates as an ion trap towards weakly basic drugs that ionize as a result of the pH differential between plasma (pH 7.4) and sweat (pH 5-6). The predominant analyte in sweat appears to be the parent drug, which is generally less polar than the metabolites commonly found in urine. However, relative amounts of parent drug and metabolite vary between drugs and subjects, and this may be attributed to differences in skin enzymes or excretory pathways. Unlike many other biofluids, heroin, which is metabolized very rapidly by the body, is excreted in the sweat. Both heroin and its metabolite, 6-acetyl-morphine, have been detected, both of which are conclusive markers of illicit opioid use.
Using a sweat patch, one-time exposure to cocaine can be detected for up to 7 days. Cocaine concentrations greater than 100 ng ml-1 were detected in sweat 72 h after a 2 mg kg -1 intranasal dose. Appearance of the drug in sweat occurs within 1-2 h of exposure and peaks within 24 h. Ion trapping may increase the detection window of certain basic drugs in sweat compared with in urine. In one study, methampheta-mine was detected in sweat for up to 140 h after a 10 mg dose, compared with 96 h in urine. Individual skin wipes obtained from methamphetamine users indicated nanogram quantities of both methamphet-amine (20-164 ng) and amphetamine (3-13 ng) on the surface of the skin.
Adulteration of patches is rare. Adulterants, which are commonly used for urinalysis, are difficult to apply without causing noticeable disturbance to the patch. Introduction beneath the adhesive layer using a hypodermic needle is possible, at the risk of causing substantial irritation to the skin as a result of the reactive or caustic nature of many adulterants. Once patches are sealed, the skin is free from further environmental drug contamination. However, studies have shown that external deposition of drug on the surface of the skin may result in detectable amounts of drug for several hours after normal hygiene practices have been performed. Cleaning the skin with alcohol before placement of a sweat patch may not sufficiently remove residual drug in these instances. In situ storage of excreted drugs in the sweat patch before removal makes drug stability in this biofluid an important consideration. Presence of drug metabolites in sweat may not reliably indicate in vivo drug exposure. Sweat contains esterases and other enzymes that may degrade or transform a drug on the surface of the skin. Hydrolysis of drugs such as cocaine and heroin to benzoylecgonine and 6-acetylmorphine, respectively, may take place inside the sweat patch.
Variation between subjects and environmental effects suggest that a correlation between sweat and blood concentration is unlikely, except perhaps under conditions of controlled sweating. Pharmacologic interpretation of drugs in sweat is not possible. The presence and quantity of drug present in a sweat sample can only be used to indicate exposure to that substance and is not indicative of impairment. However, this biofluid is unique, in as much as it offers convenient and noninvasive drug monitoring in drug compliance programs, which are becoming increasingly popular and necessary owing to increasing trends in drug use.