Abstract
The process of toxicological analysis of postmortem specimens can reveal some special difficulty compared to the clinically derived specimens. Many drugs are not stable and the chemical changes that occur in the specimens, due to the hydrolysis processing, the time passed, the drug metabolism, and matrix effect, even when the postmortem interval is short, may affect the interpretation of the toxicological results. This interpretation may be critical, not only to the thorough investigation of different kind of forensic cases, but also to clinical or other cases as it provides very significant challenges to the scientists. This article reviews (a) particular toxicological issues associated with some toxic substances responsible for common lethal or nonlethal poisonings, such as opiates, cannabis, and cocaine and the vast number of factors that affect drug concentration; and (b) focuses on toxicological issues associated with the analytical findings of certain postmortem specimens. The toxic substances cited in the present paper are the most commonly found in forensic cases in Greece. The investigation of these drug-related deaths has revealed that heroin, alone or in combination with other psychoactive substances, such as cannabis and cocaine, is the main drug involved in these deaths.
The use of psychoactive drugs in Greece over the last 15 years has been significantly increased, with cannabis holding the first place (72.8%), followed by heroin by a much lower percentage (8.9%), whereas other drugs, such as amphetamines, cocaine, and other related substances hold a percentage of 7.4%, and pills such as benzodiazepines and other nonbarbiturate sedatives range around 5.9% (Greek Annual Report 2005). These national data do not coincide with the findings of March, Oviedo-Joekes, and Romero (2006) concerning the use of illicit drugs in 10 European cities, among which, Athens, the capital of Greece. This study involves 201 addicts from a drug therapy center, who used for their daily need heroin alone (92.04%) or in combination with other drugs. The above differences probably are due to the fact that these results concern a group of drug addicts from a therapy center and not general population. Nevertheless, the March, Oviedo-Joekes, and Romero (2006) observations coincide with the Greek national results concerning other rehabilitation centers all over Greece, as heroin addicts average around 85% (Greek Annual Report 2005).
Investigation of drug-related deaths by the Drug Prosecution Coordinating Agency of the Greek Police and the Greek Reitox Focal Point of the European Monitoring Center has revealed that heroin alone or in combination with other psychoactive substances, such as cannabis and cocaine, is the main drug involved in these deaths (Athanaselis et al. 1999, 2001; Greek Annual Report 2005; Greek Police Annual Report 2005). Concerning these deaths, one has to notice that although there has been a constant increase of these deaths during the years 1991–2001, they have been stabilized during the last 5 years. The deaths due to heroin represent the 98.6% of the total drug-related deaths, whereas deaths due to cocaine represent the 0.7%, and the same percentage (0.7%) holds for deaths due to other illicit drugs (Greek Annual Report 2005; Greek Police Annual Report 2005).
The Toxicology Laboratory of the Athens University, Medical School, is often asked to perform toxicological analyses and to interpret results having very little evidence available, i.e., no history, wrong kind of specimens, biological samples taken on nonproper time and from wrong site of the body, samples not kept under good laboratory practice (GLP) standards and often on very minute amounts. The toxicologist (as every toxicologist in the world) has to overcome all of the above related problems in addition to any other issues that may affect the interpretation of the toxicological results during the investigation of a specific case, using personal knowledge and experience in addition to all relative scientific facts referred to international literature.
In forensic medicine, a number of different factors such as the terminal events, the time interval of death and discovery of the corpse, storage, and the cooperation of various professions including the forensic toxicologist may be considered to have a potential influence on obtaining the prerequisites that have been recognized to play a substantial role in the quality, reliability, and interpretation of the results. Usually, a limited amount of specimen is available for analysis; therefore, for the process of the analysis and later for the interpretation of the results, it is imperative to know the relevant circumstances of the poisoning prior to beginning the analysis. The interpretation of the toxicological results is critical for the thorough investigation of the use and abuse of psychoactive substances. A positive analytical result for a sample taken could usually result in criminal proceedings and a punitive outcome for the defendant whose sample was analyzed. Moreover, the detection of drugs of abuse in postmortem cases can provide some special difficulty compared with clinically derived specimens. The extent of chemical changes due to the time passed and the drug metabolism may affect the interpretation of the results (Robertson and Drummer 1995; Moriya and Hashimoto 1996).
In postmortem cases the selected specimens are often rarely ideal. There are often of low quality or/and quantity, leading to a difficulty in the interpretation of the results. Experience and knowledge is necessary for the interpretation, because many factors play major role, such as the state and quality of the specimen, the stability of the drug and its diffusion away from or to other tissues. Postmortem toxicology is a complex area because changes in the composition of fluids such as blood depend on the site of collection from the body, and the time elapsed since death is among other factors that may also have an influence in the results obtained (Flanagan and Connally 2005).
Postmortem changes occur in all toxic substances during the decomposition and liquefaction of tissues occurring due to high temperature or other environmental conditions and may affect the interpretation of the results (Robertson and Drummer 1995; Moriya and Hashimoto 1996; Skopp et al. 2001; Skopp and Potsch 2004). Problems arise from interfering substances that are endogenously produced as well as from alterations of drug or poison due to the autolysis of putrefactive process during decomposition. In addition, degradation, as well as formation of drugs, has been observed during the postmortem interval (Stevens 1984; Fiedler and Graw 2003).
The autopsy and toxicological analysis should be started as soon after death as possible. The natural enzymatic and nonenzymatic processes of decomposition and microbial metabolism may destroy a poison that was present at death or produce substances or compounds with chemical and physical properties similar to those of commonly encountered poisons. Before the analysis begins, several factors must be considered, such as the amount of specimen available, the nature of the poison sought, and the biotransformation of the poison. The parent compound and any major physiologically active metabolites should be isolated and identified. In some instances, the metabolites provide the only evidence that a drug or poison has been administered (Poklis 2001).
The choice of specimen depends on the case being investigated; however, the most common specimens used for the analysis of toxic substances in postmortem cases are blood, liver, and urine, although vitreous humor is important in routine toxicological cases, particularly for the determination of ethanol and drugs of abuse (Drummer 2004). In cases of extreme putrefaction, muscular tissue, hair, and bone can be used successfully as alternative specimens for the toxicological analysis. In putrefied bodies quantitative results are of little value. In some cases analyses of drugs in fly larvae in decomposing cases provide an insight as to the presence of drugs in the corpse (Pounder et al. 1996; Drummer 2004).
Once the analysis of the specimens is complete, the toxicologist must interpret the findings in regard to the physiologic or behavioral effects of the toxicants on the deceased at the concentrations found. Specific questions may be answered, such as the route of administration, the dose administered, and whether the concentration of the drug of abuse present was sufficient to cause death or alter the deceased’s actions enough to cause death. No result can be interpreted without thorough examination of the relevant circumstances and factors that could impact on the drug concentration (Poklis 2001; Drummer 2004). Because the most frequently observed compounds in drug-related deaths, not only in Greece but also in other countries, are by far cannabis, opiates, and cocaine, this work focuses on the toxicological issues associated with the analysis of postmortem specimens of blood and urine.
CANNABIS
The most widespread abuse of cannabis is by smoking and it is the most illicit drug found in many jurisdictions, particularly those involved in fatal or nonfatal traffic accident; the use of any other psychoactive substance follows, with the exception of alcohol (Athanaselis et al. 2001; Mravcik, Zabransky, and Vorel 2005).
Although over 20 metabolites of Δ9-tetrahydrocannabinol (THC) have so far been identified, oxidation at the C-11 position and glucuronidation account for the major compounds appearing in urine (Wall and Perez-Reyes 1981; Widman, Halldin, and Agurell 1985). The major acidic metabolite is 11-nor-Δ9-tetrahydrocannabinol-9 carboxylic acid (9-carboxy-THC), which is converted to mono- and diglucuronide conjugates, the major forms of metabolites excreted in urine. Thus, the identification of 9-carboxy-THC in urine is considered the best indication of previous cannabis consumption, because the excretion of 9-carboxy-THC is prolonged due to the tendency of Δ9-THC to be absorbed and accumulate into fatty tissues. Thus, this major metabolite is often measured following hydrolysis of the glucuronide conjugates, in situations where past cannabis exposure can lead to sanctions particularly in cases such as prisons, pre-employment testing, and workplace (Smith-Kielland, Skuterud, and Morland 1999).
In traffic cases, blood Δ9-THC, the main active constituent of cannabis, constitutes a better measure for recent use than urinary metabolites and constitutes the main analyte of interest, particularly in legal cases of traffic accidents and other offenders who act under the influence of Δ9-THC. The question that often arises in traffic accident cases is whether or not a driver was driving under the influence of cannabis (DUIC) or whether it is possible to determine the relationship between blood THC concentration and driver impairment. Various recent projects have been financed by the European Union, particularly because it has been determined that more than two thirds of drug users in Europe drive after having smoked cannabis (Raes and Verstraete 2006). Recent data suggest that drivers with a measurable Δ9-THC concentration (>1 ng/ml) have an elevated crash risk (Ramaekers et al. 2002; Mura et al. 2003; Drummer et al. 2004). Furthermore, this increased risk of responsibility in road crashes shows a significant dose-effect pattern as the Δ9-THC level rises up to 5 ng/ml (Laumon et al. 2005). Other studies have further concluded that serum Δ9-THC concentration between 2 and 5 ng/ml establish the lower and upper ranges of Δ9-THC limit for impairment (Ramaekers et al. 2006), whereas concentrations higher than 5 ng/ml are indicative of significant impairment. It is also interesting to state that if drivers are provided with enough factual information about the accident risk associated with DUIC and if the severity of penalties for DUIC is not increased, this wiliness for cannabis-intoxicated driving will reduce (Jones et al. 2006).
The issue of passive or inadvertent exposure to marijuana smoke is sometimes raised as an explanation of a positive urine assay result. Although this has been demonstrated to occur, achieving a sufficient dose of Δ9-THC by this route is difficult and unlikely in most instances, because it requires extreme situations of exposure, such as congested and nonventilated room space (Perez-Reyes et al. 1983; Morland et al. 1985). If screening assays with a cut-off 20 ng/ml are used, positive results can occur but very infrequently. With cut-off levels of 100 ng/ml, the possibility of a passive positive result is virtually eliminated, but they create many false negatives (Law et al. 1984; Moffat 1986; Cone et al. 1987; Liu 1992). Thus, most toxicological laboratories today use the cutt-off level of 50 ng/ml to overcome these analytical problems (Luzzi et al. 2004). That is why whenever minute amounts of Δ9-THC or 9-carboxy-THC are present in blood or urine samples, the discrimination between active and passive inhalation continues to cause some problems. The statement of a passive exposure by marijuana smoke has been scrutinized by reviewing the literature to date and it seems not only useful but also maybe urgent to enlarge the existing database, particularly in jurisdiction cases (Skopp et al. 2001).
One should always keep in mind that urine drug concentrations can be influenced by different factors and primarily by liquid intake, which may change the urine concentration of a drug 10-fold in a matter of hours. This means that caution must be exercised when interpreting a positive result occurring after a negative one in a daily sampling regime. This happens in the case of Δ9-THC and 9-carboxy-THC, free or glucuronidated, with a long half-life, where detectability is probable over several days but with the concentration varying significantly around the cut-off level. Positive samples following negative ones do not necessarily indicate additional use of marijuana (United Nations 1995). It has also been observed that Δ9-THC concentrations decrease in blood also with time, particularly when stored at −20°C (Drummer, Gerostamoulos, and Chu 2002) and moderate losses of 9-carboxy-THC have also been reported not only in urine stored frozen but also at room temperature for several days. Last but not least, both free and glucoronidated 9-carboxy-THC have shown variable stability with storage temperature and sample pH that may raise questions in advanced forensic interpretations (Skopp and Potsch 2004).
OPIATES
Opiate testing is one of the most frequently requested assays in the death investigation of suspected drug overdoses. Screening assays typically target morphine, whereas more specific assays are able to positively identify and quantify the particular opiate present in the specimen (Jenkins and Lavins 1998). A positive result in an initial immunoassay screen means that an opiate is present in the urine at a level above or equal to the cut-off level and should be confirmed by a method that is sensitive but more specific than the initial test. Retention of opiates in the body and the actual drug concentrations in urine depend on different factors, such as drug metabolism, the subject’s physical condition, fluid intake, and manner of ingestion. In general, by using the above approach, opiates may be detected in urine for up to 3 days (United Nations 1995; Baselt 2000).
The presence of morphine in urine does not indicate which opiate was consumed, because opium, heroin, codeine, or morphine itself share common metabolic pathways and may be sources of morphine and morphine-3- and -6-glucuronide in urine. Morphine is rapidly excreted in urine as glucuronides, with up to 85% of the dose recovered in urine within 24 h and only small amounts of morphine are excreted unchanged (2% to 10%) (Drummer 2004).
Heroin is converted within minutes to morphine through the intermediate 6-acetylmorphine (6-AM). In addition, other opiates such as ethylmorphine, pholcodine, and nicomorphine may also be sources of morphine (United Nations 1995). Hence, the presence of morphine does not indicate which opiate was consumed. Thus, it is necessary to differentiate heroin-related deaths from other opiate fatalities. The only definitive method of achieving this is to monitor heroin or 6-AM or both. Both heroin and 6-AM are only present in blood and tissues for a relatively short period (Wyman and Bultman 2004). Because of its short half-life and instability in aqueous media, heroin is not typically monitored. Therefore, the detection of its active metabolite, 6-AM, is of major importance in order for the pathologist to determine if recent heroin exposure caused or was a contributing factor in an individual’s death (Jenkins and Lavins 1998). Moreover, heroin is not only rapidly converted into its respective hydrolytic products during life, but it also undergoes rapid bioconversion in situ after death. Additionally, unless special precautions are undertaken, hydrolysis may even occur in the collection vessel (Drummer 2004). Recent studies have shown that 6-AM may persist in cerebrospinal fluid and vitreous humor and these specimens may be preferable for establishing heroin exposure (Wyman and Bultman 2004). Thus, in special forensic cases the analysis of vitreous humour and cerebrospinal fluid may be of high value (Skopp et al. 2001).
Small amounts of codeine are also present in the urine of heroin users because of the presence of acetylcodeine in the heroin as an adulterant. In the case of codeine, it is generally accepted that if the total codeine to total morphine ratio is less than 0.5 and the total morphine concentration in urine is greater than 200 ng/ml, codeine may be excluded as the source of the morphine present (El-Sohly and Jones 1989; Gerostamoulos, Staikos, and Drummer 2001).
Concerning blood samples, it has been accepted that finding a morphine/codeine ratio >1 furnishes evidence of heroin use. It has been proven that a high morphine/codeine ratio cannot be attributed to a person’s inherent ability to convert codeine to morphine (Hand, Moore, and Sear 1988; Yue et al. 1991; Ceder and Jones 2001). Because the distribution ratio of morphine between plasma and whole blood is close to unity, the ratio of morphine to codeine can also be applied to plasma or serum samples (Ceder and Jones 2001).
Morphine is relatively stable in specimens when stored frozen, but shows significant losses when stored at 4°C or higher for more than a few days, or in postmortem specimens (Carroll et al. 2000). Morphine and glucuronide concentrations from cases in the early stages of putrefaction or when prolonged storage has occurred may have substantially changed from the time of death (Skopp et al. 2001). 6-AM undergoes deacetylation to morphine at room temperature and according to pH of the specimen. However, 6-AM is stable in frozen urine (−20°C) for at least 12 months (Fuller and Anderson 1992; Skopp et al. 2001).
COCAINE
Cocaine acts as an inhibitor of reuptake in dopamine and norepinephrine nerve terminals in the central nervous system (CNS), as well as serotonin, and is considered a potent stimulant of nerve function. Cocaine is rapidly metabolized mainly to benzoylecgonine (BE) and ecgonine methyl ester (EME). Moreover, persons, who coconsume ethanol have the biologically active product of cocaethylene (CE) in their body fluids and relatively lower urinary concentrations of benzoylecgonine (Moriya, Hashimoto, and Ishizu 1995; Harris et al. 2003). This decrease in urinary BE, however, does not depend on a possible inhibition by CE but rather in the ethanol-mediated inhibition on cocaine metabolism (Parker et al. 1998). Furthermore, anhydroergonidine methyl ester (methylecgonidine), a pyrolysis product of cocaine, can only be found in body fluids of people who smoke cocaine (Erzouki et al. 1995).
After a single dose of cocaine, the unchanged drug can be detected for up to 24 h and the metabolites BE and EME for up to 48 h (Ambre 1985; Hawks and Chiang 1986). After chronic use, the detection time can be longer, up to 5 days or more (Cone and Weddington 1989; Parker et al. 1998; Preston et al. 2002). Furthermore, the terminal elimination half-life of cocaine ranges from about 40 min to 4 h, depending on dose (Cone 1995). The presence of BE and EME and the absence of cocaine in biological fluids indicate the past use of cocaine. BE and EME are the major metabolites detected in urine, whereas cocaine accounts only for 1% to 9%. Nevertheless, in most if not all cocaine-related death, blood concentrations of the parent drug cocaine is found in addition to one or both of its two major metabolites, BE and EME (Lora-Tamayo, Tena, Rodriguez 1994).
Minimal differences are found in the relative amounts of metabolites excreted following administration of cocaine intranasally, intravenously, or by smoking. In addition, it is not possible to draw conclusions from urinary concentrations of cocaine and its metabolites concerning the route of exposure, the amount of drug administered, the time since the last dose, or the level of impairment (United Nations 1995).
Plasma concentrations of cocaine and BE following therapeutic administration of cocaine are normally less than 0.5 and 0.1 μg/ml, respectively. In overdose cases, levels in autopsy blood are in the range of 1 to 20 and 1 to 10 μg/ml, respectively (Baselt 2000; United Nations 1995). Nevertheless, as with other illicit drugs, there is no defined “safe” or “therapeutic” blood concentration (Drummer 2004).
Cocaine and its metabolites show poor stability with respect to hydrolysis. Cocaine is not only rapidly converted into its respective hydrolytic products during life, but it also undergoes rapid bioconversion in situ after death. This is the reason why concentrations of cocaine and its metabolites are unlikely to yield useful interpretive information (Karch, Stephens, and Ho 1998). Moreover, unless special precautions are undertaken, hydrolysis may even occur in the collection vessel (Drummer 2004). For this reason, blood and plasma samples should be collected in tubes containing sodium fluoride and the pH adjusted to pH 5 with acetic acid. They can then be kept in a refrigerator at 4°C or frozen if possible for some months, although degradation occurs even at −20°C (Isenschmid, Levine, and Caplan 1989; United Nations 1995). In addition, due to the above hydrolysis process, the analysis of brain tissue may yield more useful interpretive information, because cocaine in brain seems more stable to hydrolysis than in other tissues (Moriya and Hashimoto 1996). This by no means can lead the analyst to underestimate the analysis of blood and urine.
CONCLUDING COMMENTS
The interpretation of the analytical results of the biological specimens in poisonings, lethal or not, must always take place with extreme caution and always taking into account the circumstances of poisoning or death, the autopsy findings, and the recent history of the poisoned patient or the descendent. The analysis of the proper postmortem specimens can provide special challenges for forensic toxicologists. The choice of specimens depends on the case being investigated. A number of errors can be made when one attempts to estimate antemortem drug concentrations and the ingested dose from postmortem measurements. The chosen site of the body and the technique applied for postmortem blood sampling can greatly influence the concentration of drug measured. Thus, applying the proper attention to the preanalytical period of a forensic case may provide not only the proper specimens, but it may also play a very substantial role in the quality and reliability of analytical results.
The most common specimens used for the analysis of drugs of abuse in postmortem cases are blood, liver, and urine. However, the collection of several alternative specimens to guard against the possibility of poor specimen collection is warranted. Therefore, it is of critical importance that anyone who attempts to interpret the results of a specific toxicological analysis must have adequate knowledge of the pharmacokinetics and the toxicokinetics as well as of the relative rate and pattern of biotransformation of the drugs and poisons involved. Furthermore, increasing our knowledge on the degradation mechanisms may enable us not only to target the right substance, which may be a major breakdown product in the investigation of highly labile compounds, but also facilitate us to assess in the best possible way sample quality and interpretation of analytical results.