SECTION II: LEGAL ASPECTS OF TOXICOLOGY

INTRODUCTION

The number and diversity of toxicologic emergencies faced by emergency department (ED) staff have increased steadily since the early 1970s and continue to rise today. This chapter discusses the medicolegal management of patients who are exposed to xenobiotics that alter their ability to think. It also addresses the legal and ethical dilemmas routinely encountered by emergency medicine practitioners.

Patients with toxicologic emergencies require immediate care, and yet are often unable to give consent because their impaired consciousness prevents them from making informed decisions. Treating patients who present with an acute organic impairment manifested by confusion, irrational thought, or even dangerous behavior is very challenging. Emergency practitioners must recognize the medicolegal problems created when the impaired patient refuses treatment and insists on leaving against medical advice. The issue is further complicated by the variations in relevant state laws. Emergency practitioners must become familiar with the legal requirements of informed consent and the essential management necessary to avoid liability for negligence and abandonment within the state in which they practice. Of particular concern are the risk management and liability issues that relate to impaired patients attempting to leave the ED before medical care is complete. The legal requirements of informed consent in emergency settings, the duty to treat, medical malpractice risk, battery, and negligence allegations, are examined here. Guidelines based on generally accepted common law principles are suggested for developing appropriate patient care plans and departmental policies. These issues and principles are best illustrated by case examples.

INFORMED CONSENT

Patient 1

An 18-year-old college student was brought by ambulance to the ED after a friend reported seeing her in the bathroom with slit wrists and an empty bottle of acetaminophen (APAP). In the ED, the patient was alert and oriented to time, place and person. Vital signs were as follows: blood pressure, 120/65 mm Hg; pulse, 95 beats/min; respiratory rate, 16 breaths/min; and temperature, 99.1°F (37.3°C). A rapid bedside glucose concentration was 120 mg/dL. The patient stated that she ingested the APAP approximately 5 hours earlier in an attempt to kill herself. The health care team wished to measure an APAP concentration to determine whether N-acetylcysteine should be administered. The patient refused venipuncture and stated that she would refuse any medications. The physicians informed the patient that she might suffer irreparable damage to her liver and possibly die if not treated immediately. This type of clinical dilemma is the ideal moment to discuss strategies with the local poison control center and medical toxicologist. The poison control center staff will have great experience in addressing complex unorthodox management strategies, suggesting how much time the physician has in the emergency department to achieve a solution without substantially compromising the patient’s care and how much of the normal strategy can be altered or eliminated and still render quality care.

Medically treating patients against their will poses a difficult problem. Forcible treatment violates the principle of autonomy and the patient’s right to privacy. However, harm to the patient may ensue if the appropriate evaluation and treatments are not performed. In the case above, the patient gives a history of ingesting a large amount of APAP, which risks causing hepatotoxicity and even death if not treated (Chap. 33). Her refusal to be evaluated raises the important question of whether a physician is ever justified in performing an assessment of someone who is alert and oriented, refuses treatment, yet is poisoned. Do most patients suffering from a mental illness such as depression require treatment against their will? As in this case, if the harm that the patient faces is not immediate, but certain in the near future, does the physician have the duty and authority to treat? General principles of patient autonomy, informed consent and battery will be elucidated here.

A patient’s right to choose a course of medical treatment was first recognized in the early twentieth century in a landmark case decided by the New York State (NYS) Court of Appeals in Schloendorff v. Society of New York Hospital.²⁶ Mary Schloendorff was admitted for abdominal pain and was found on physical examination to have a mass. Her physicians offered her a more thorough examination under anesthesia and the option for surgery. Although she consented only to the exploratory surgery, during the procedure the mass was removed. Postoperatively, Schloendorff developed an infection, gangrene, and several fingers had to be amputated. She sued the hospital. In its decision, the Court acknowledged Schloendorff’s right to self-determination, and the right to refuse treatment. It stated:

This decision formed the foundation for the “Doctrine of Informed Consent.” Informed consent serves to protect a patient’s autonomy, and creates the concept that each individual has the right to choose a personal course of medical treatment.⁴⁵ This includes the right to refuse care, and the right to terminate care already in process. In a nonemergent situation, it is the responsibility of the physician to obtain approval from a patient with decision making capacity or a legally recognized surrogate for those patients lacking such capacity before rendering treatment.

The core elements of the informed consent process consist of (1) an explanation of the treatment or procedure; (2) the disclosure of alternative choices to the proposed intervention; and (3) a discussion of the relevant risks, benefits, and uncertainties associated with each alternative; and (4) the likely outcomes with the various treatment options including the choice of no treatment. This discussion must take place whether the patient is either consenting to a procedure or refusing a recommended procedure. A patient does not automatically assume the risks of rejecting the recommendations of the physician if the patient is not fully apprised of the consequences of his or her decision.³⁵ Furthermore, it is the duty of the physician to assess how well the patient understands the above information. Before a physician accepts a patient’s consent or refusal of care, the patient must demonstrate to the physician an adequate understanding of the nature and consequences of the decision made.

Depending on the state, courts apply one of 2 standards to determine if the information communicated by the physician was sufficient to satisfy common law obligations. The reasonable person standard requires a physician to disclose information that a reasonable person in the same position as the patient would need in order to make an informed decision.¹¹ The alternative legal approach is the reasonable physician standard, which requires that a physician to reveal information that a reasonable physician in a similar circumstance would disclose to the patient.¹¹ States vary on which standard they apply, and it is incumbent on physicians to know the legal standard of the state in which they practice.

Courts recognize that the requirement for informed consent has exceptions for situations where physicians do not need to obtain permission before rendering treatment. In Schloendorff v. Society of New York Hospital, the Court recognized that consent is not required in emergency situations and is implied. Situations are generally considered emergent if care of the patient would be compromised with a delay in treatment. For example, in New York State, an emergency is defined as a situation that includes both the immediate endangerment of life or health and the need for the immediate alleviation of pain.³² Physicians need to determine the specific requirements of informed consent in their respective states.

Often well-intended efforts on the part of the physician to fully communicate treatment information to an impaired patient prove ineffectual and then present the practitioner with a medicolegal dilemma. Physicians are frequently unable to discuss in a meaningful way the implications of the proposed treatment with the patient. Nevertheless, there remains a duty to treat the patient who presents with a life-threatening condition or one with the potential for permanent disability. In these situations, consent is considered to be implied, and emergent treatment should be provided. The principle of implied consent is a general tenet of tort law.²² In our case, although the physicians have explained the risks, they must have a more extensive conversation to determine if the patient is of sound mind. Suicide ideation raises the concern that the patient is not exhibiting rational thought and a thorough investigation needs to take place to determine if the patient has decision-making capacity. Given the fact that timely evaluation and intervention are necessary to prevent hepatotoxicity in overdose, the physician needs to draw blood early even if the patient does not consent, as she is at risk of harm if there is a delay in her care. So in this case, it is essential to begin management by any number of approaches including administering activated charcoal, obtaining blood specimen to assess hepatotoxicity, coagulation profile, basic metabolic and toxicologic evaluation of acetaminophen concentrations and/or initiating N-acetylcysteine with or without agreement of the patient. These options are all possible appropriate recommendations for the clinician. Decisions of this complexity should be accomplished with the full support of the local poison control center, a medical toxicologist, and an excellent team in the emergency department.

RISK MANAGEMENT CONSIDERATION AND DOCUMENTATION

Patient 2

A 28-year-old woman was brought to the ED by police who believed she ingested a large quantity of drugs to avoid arrest minutes before their arrival. The triage nurse helped the patient onto a stretcher, brought her to the treatment area, and recorded normal vital signs. The patient became combative and uncooperative as the physician initiated the examination and he verbally ordered that the patient be restrained. The patient was given 40% oxygen via face mask, cardiac monitoring was initiated, and an intravenous line was inserted and 0.9% sodium chloride at a rate of 125 mL/h was administered. A fingerstick glucose was determined to be normal and 100 mg of thiamine were administered intravenously (IV). Three doses of IV diazepam (5 mg) in 5-minute intervals were administered until light sedation was achieved. The physician ordered a basic set of laboratory studies, a urine drug screen and ordered an abdominal radiograph. The laboratory values and the radiographic findings were normal and the urine drug screen was negative. The treatment plan was then to continue monitoring and to reassess the patient at a later time. This was an ideal time for the physician to contact the local poison control center or a medical toxicologist to better understand the management of a body stuffer (Special Considerations: SC5). There are critical decisions to be made immediately: protection of the patient’s airway, management of the ingested toxin, and a high level of supervision of the patient in an intensive care unit (ICU). This case demonstrates inadequate knowledge of the management of an unknown toxin. It demonstrates the risks of management without utilization of available hospital (ICU) and health system resources (Chap. 130).

One hour after arrival the patient’s vital signs were blood pressure, 120/70 mm Hg; pulse, 82 beats/min; and respiratory, rate 18 breaths/min. The patient was noted to be stable and was transferred to an observation unit. Oxygen and cardiac monitoring were discontinued. No further orders were written, and the patient remained restrained. Three hours later, the vital signs were blood pressure, 110/60 mm Hg; pulse, 92 beats/min; and respiratory rate, 18 breaths/min. A nursing note stated the patient was resting comfortably.

Forty-five minutes later, at midnight, the initial treating physician completed his shift and informed his replacement that the patient was stable and resting in the observation unit.

At 4:20 AM the patient was found unresponsive, hypotensive, with agonal respirations and with a weakly palpable pulse. The patient felt very hot to the touch and had a rectal temperature of 108°F (42.4°C). Resuscitative efforts were initiated but were unsuccessful. Thirty minutes later, the patient was pronounced dead.

Several important risk management questions frequently arise in emergency medicine malpractice litigation. To prove that a case constitutes medical malpractice, a plaintiff’s attorney must show by a preponderance of evidence that a deviation from accepted practice by the physician occurred. The attorney must further demonstrate that this deviation or negligent act or omission by the physician proximately caused the patient’s injury. Courts have held that where “there is substantial probability that the [defendant physician’s] negligent conduct caused the resulting injury, that sufficient evidence has been developed against [the] physician.”⁴²

When the attorney for the patient (plaintiff’s attorney) introduces evidence to advance the case, the ED record is likely to become the central focus in the medical malpractice trial. The problems associated with an improperly documented ED record are numerous, but they can be minimized if the practitioner is cognizant of risk management principles. Every entry in that record is scrutinized by both parties (plaintiff and defendants). The importance of completing the record with knowledge of risk management implications should be a priority for all health care professionals.

The physician is required to complete a medical record that amply supports the basis for the medical judgments exercised. When a physician opts to complete only a summary statement on the record without noting supporting clinical data or patient history, claims alleging a failure to diagnose will be extremely difficult, if not impossible, to defend. One of the basic tactics of the defense in a medical malpractice case is the argument that the physician’s judgment was appropriate given the clinical facts and the patient’s history available at that time. Therefore, physicians who do not record supporting clinical data and history undermine their “medical judgment” defense.

Notations in the medical record related to patient monitoring are considered inadequate when those notes do not provide insight into the patient’s clinical status. Thus, any note for a restrained patient in the ED for a lengthy period of evaluation, observation, or until an inpatient bed becomes available must include specific clinical data and observations overtime (including relevant laboratory results, radiographic findings, hemodynamic changes, and intravenous medications and solutions). Any deficiencies in this area would undoubtedly be noticed and highlighted at trial by a plaintiff’s expert, who usually is a board-certified physician in the same specialty.

Documentation supporting the restraint of an impaired patient against his or her will must include a description of the clinical situation to support such a forcible impediment to the patient’s right to liberty and freedom of movement. Such a description should specifically document any manifestation of agitation and/or behavior that might place the patient at risk. The record should incorporate a description of the specific acts of the patient of concern and, most importantly, should comment on the difficulties in providing care to the patient because of those acts. Documentation of patient conduct should be factual, accurate, detailed, and nonjudgmental.

The transition to electronic health records (EHRs) was introduced as a way to improve the health care process by enhancing communication, workflow, decision making, and increasing patient safety while simultaneously decreasing errors. Despite their many actual and potential benefits, EHRs have introduced new concerns not previously encountered in written record keeping. Some of the new errors encountered include poor data display, alert fatigue, wrong order/wrong patient error, and communication failure.¹⁰ An additional concern that could lead to errors and increase physician liability is the potential to overlook pertinent data given the large volume of information that is provided on every aspect of care for each patient regardless of when it took place. For example, every laboratory test that a patient has ever undergone in one hospital system will often be available to the physician as she or he opens the chart to document on a patient’s current visit, yet the physician will only want to focus on what is important for the patient’s current encounter. The copy and paste function that is often used by providers to save time risks leading to repeating information that is no longer correct.²⁸ Computerized provider order entry (CPOE) was introduced to improve all aspects of the medication process from prescribing to administration but ultimately introduced new problems. Common errors associated with CPOE include wrong medication selection, incorrect input of patient data, wrong dose or strength selection, and scheduling errors (Chap. 134).^25,43

To summarize, a well-documented ED record reflecting the practice of accepted risk management principles is the best course to follow for the physician and nursing team managing a difficult overdose situation and one in which legal concerns may appear to present problems in providing proper medical management.

FORCIBLE RESTRAINT OF THE IMPAIRED PATIENT

Patient 3

A 23-year-old woman was found unresponsive on the street and transported to the ED by ambulance. Friends on the scene reported that the patient regularly uses opioids recreationally. In the ED, she was unresponsive and apneic. Oxygen was immediately administered by bag–valve–mask and intravenous access was established. Naloxone was administered, and shortly thereafter the patient regained consciousness. After 20 minutes of care in the ED, the patient became fully alert and oriented, with no evidence of hypoxia or other clinical signs to suggest impaired judgment. The patient stated that she had taken a sustained-release oxycodone and demanded to be discharged immediately.

The ability of a hospital to retain and physically restrain a person who has an altered level of consciousness for evaluation and emergency intervention is generally well supported by state statutes and case law.¹¹ Reasonably clear guidelines for the management of such impaired patients have evolved from legal precedents governing appropriate medical assessment, from risk management considerations, and from the predictability of patient injury in the event of premature discharge.

A staff decision to allow a treated or partially treated patient with a drug overdose who subsequently becomes alert to return to the community must be based on an assessment of several factors. The initial concern is the capacity of the patient; is the patient able to fully comprehend the current medical situation and associated risks of not receiving treatment? To demonstrate capacity, patients must be able to (1) understand the facts involved in the decision; (2) appreciate the nature and the significance of the decision; (3) weigh the risks and benefits and potential alternatives; and (4) express their choice.¹⁶ The next consideration is that of medical stability. Has the initial process that caused the clinical scenario completed its course? The history of drug use in this patient raises the concern that the underlying toxic metabolic process is not yet resolved, and alteration in mental status, significant respiratory compromise, or other medical symptoms will recur when the naloxone is metabolized, again placing the patient at risk.

Common ED practice and sound legal principles suggest that both the hospital and its staff have a duty to prevent such a person from leaving if the duration of the effect of the involved xenobiotic is longer than that expected for the antidote. Because the duration of effect for naloxone is considerably shorter than that of sustained-release oxycodone, the physician can predict with reasonable certainty that coma or apnea will recur in the near future. The physician has the duty to inform the individual of the life-threatening nature of the condition, and then to retain, with restraints if necessary, the patient in the hospital until medically stable.

Legal liability in this situation is further reduced if the chart substantiates the medical judgment that was the basis for the decision to retain the patient and, if applicable, the use of restraints. Such documentation should specifically note the likely relapse of the patient into a symptomatic state and that this occurrence could place the patient in a life-threatening situation. When documented in a clear manner, legal challenges to the decision to restrain the patient have a limited chance of success. Sound risk management principles support treatment and detainment. Conversely, prematurely releasing a patient with a significant overdose exposes both the physician and the hospital to a claim of negligence on the grounds of failure to foresee a likely and harmful event. From a practical perspective, the legal risk and liability of the latter scenario is far greater than the former.

In the above case, the patient responded to the opioid antagonist naloxone. Given that there are many different opioids available with variation in half-lives and duration of action, the patient needs to be kept in an acute care setting until the naloxone is metabolized to ensure the patient has not been exposed to a long-acting opioid that will cause recurrent life-threatening toxicity. Once the antidote has worn off the patient must be reassessed to determine possible damage from the initial event (hypoxic brain injury, aspiration, end organ impairment) as well as decision-making capacity and possible underlying psychiatric illness, such as suicidal ideation and the need for a psychiatry consultation. If it is determined that the patient is safe for discharge from the hospital consideration should be made to ensure that available family and social support is arranged and harm reduction including overdose education, naloxone prescriptions for family and friends, and access to counseling and medication-assisted treatment modalities (methadone buprenorphine naltrexone programs) after discharge are performed (Special Considerations: SC6).¹²

BLOOD ALCOHOL CONCENTRATION AND EVIDENCE COLLECTION

Patient 4

A 41-year-old man who was a driver involved in a motor vehicle crash was taken to the ED by ambulance. Two motorists in another vehicle were killed. The patient had no physical complaints, but was transported to the hospital for medical evaluation. On arrival, he was alert and oriented, responded appropriately to all commands, and demonstrated normal motor function and had a normal gait. Police officers suspected that he was driving while intoxicated, but he refused a breath alcohol test at the scene.

The police officers informed the ED staff that he was arrested and might be charged with vehicular homicide. The officers then requested that the emergency physician draw a blood specimen to determine the blood alcohol concentration. The patient stated that he would not allow the ED staff to draw blood for a determination of an alcohol concentration.

In 2015, there were 10,265 people who died in alcohol-related motor vehicle crashes, which comprised 29% of all traffic-related deaths in the United States.⁹ In the United States, an alcohol-related motor vehicle crash kills someone every 51 minutes.⁹ In 2015, crashes resulting from alcohol-induced impairment were responsible for 16% of fatalities in children younger than 14 years who were killed in motor vehicle crashes.⁹ The estimated economic cost of alcohol-impaired motor vehicle crashes in 2015 was $44 billion.⁹ The judicial system has historically been one of the most effective tools to combat drunk driving and its effectiveness depends on the ability to identify and punish individuals who violate the law. See Special Considerations: SC11 for more in-depth discussion on “per se” limits. It is essential, however, that the collection of evidence does not violate the rights afforded by Constitutional law or specific state law. Does forced phlebotomy for patients suspected of driving while intoxicated violate these protected rights? This has long been debated in the courts, and legal authority specifically brought into question include the Fourth Amendment,³⁹ the right against unreasonable search and seizure; the Fifth Amendment,⁴⁰ the right against self-incrimination, and the Due Process clause of the Fourteenth Amendment.³⁸ Past court decisions applying Constitutional or statutory law on these issues help guide current clinical practices and also contribute to further evolution of legal standards.

Every state and the District of Columbia have driver “implied consent” laws. When a person obtains a driver’s license, he or she consents at the time of acquisition to a chemical alcohol test if suspected of driving while intoxicated. Under implied consent laws, when a person suspected of driving while intoxicated refuses to take an alcohol test, a penalty is imposed. Specific penalties for refusals vary from state to state. At a minimum, the refusal results in suspension or revocation of a driver’s license. Some states assign additional fines and penalties for this action. A few states allow the refusal itself to be submitted at trial in support of the prosecution, making it possible to be convicted of an intoxication charge without chemical evidence.⁵ Certain states (eg, Texas, Illinois) allow blood tests to be performed on patients as ordered by an officer of the law when there is probable cause that driving while intoxicated resulted in severe injury.^21,33 Other states, such as New York and Wyoming, allow forced blood samples with a warrant issued by a judge.^7,19 State laws in this area vary with respect to amount of force that may be used, requirements of warrants, who may perform the blood draw, and other blood draw restrictions and it is important that the ED staff be familiar with the specific requirements of law in the state in which they practice.²

How much force may be used to collect chemical evidence? Does a physician violate human dignity and privacy in obtaining evidence for the State? These issues were addressed by the United States Supreme Court. In Rochin v. California,²⁴ the Court overturned a conviction of drug possession based on violation of the Fourteenth Amendment. In this case, police were informed that Rochin was selling drugs. While entering the home of the defendant, the police witnessed the defendant swallow 2 pills that were lying on the night stand. When the officers failed to recover the pills on the scene, the officers took Rochin to the ED, where they directed the physician to administer an emetic through a nasogastric tube. The capsules were recovered from the vomit, and Mr. Rochin was convicted by the trial court for possessing morphine.²⁴ The Supreme Court reversed the lower court decision based on Fourteenth Amendment grounds stating that “nor shall any State deprive any person of life, liberty or property, without due process of law.”⁴⁵ The term due process is vague, and is defined on a case-by-case basis; essentially, however, it means that states must use fair legal procedures when depriving an individual of life, liberty, or property. In Rochin, the court concluded that forced emesis by a physician was believed to violate “due process,” stating:

The Supreme Court revisited the issues presented in Rochin 4 years later in Breithaupt v. Abram.⁶ Breithaupt was the driver of a truck who killed 3 occupants of another vehicle. In the ED, a police officer requested that a blood alcohol concentration be drawn. The blood, drawn while the patient was unconscious, was above the legal limit for alcohol and the patient was convicted of involuntary manslaughter. Making an analogy to the forced emesis scenario in Rochin, Breithaupt argued that the blood draw violated due process because he did not consent to its collection. Justice Clark disagreed, stating “the distinction rests on the fact that there is nothing ‘brutal’ or ‘offensive’ when done, as in this case, under the protective eye of a physician” and that the “blood test procedure has become routine in our everyday life.”^5,6 Phlebotomy while the patient is unconscious and unable to give consent was determined not to violate the due process clause of the Fourteenth Amendment.

The Supreme Court continued to expand the scope of permissible phlebotomy in Schmerber v. California.²⁷ Schmerber was involved in a motor vehicle crash in which the police officer suspected he was intoxicated. Unlike the Breithaupt case, Schmerber was conscious, and a physician drew a blood sample at the officer’s request despite the patient’s verbal refusal. The attorney for Schmerber asserted that forced phlebotomy violated several constitutional rights. Specifically, it violated the Fourth, Fifth, and Fourteenth Amendments of the US Constitution. He alleged that forced phlebotomy denied him due process of law, it violated his privilege against self-incrimination, and it violated his right against unreasonable search and seizure.^5,27

The Supreme Court rejected all of these arguments holding that the phlebotomy did not violate the due process clause because “the extraction was made by a physician in a simple, medically acceptable manner in a hospital environment. . We cannot see that it should make any difference whether one states unequivocally that he objects or resorts to physical violence in protest or is in such condition that he is unable to protest.”²⁷ Furthermore, the court found that the forced blood draw did not violate the Fifth Amendment’s Privilege Against Self-Incrimination because the Fifth Amendment only protects evidence of a “testimonial or communicative nature,” such as writings or speech. Finally the court found there was no violation of the Fourth and Fourteenth Amendments’ protections against unreasonable search and seizure as the delay necessary to obtain a warrant threatened the destruction of evidence.

Considering the “totality of the circumstances” in Schmerber, the blood alcohol test fell under the Court’s exigent circumstances exception to the general requirement of a warrant.

The percentage of alcohol in the blood begins to diminish shortly after drinking stops as the body eliminates it from the system. Particularly in a case such as this where time elapsed to bring the accused to a hospital and to investigate the scene of the crash, there was no time to seek out a magistrate and secure a warrant.

To ensure compliance with standards set forth in Schmerber, the states have tailored laws and regulations governing the seizure of blood for the purpose of blood alcohol testing. Laws generally require the procedure be (1) performed in a reasonable, medically approved manner; (2) incident to a lawful arrest; and (3) based on the belief that the arrestee is intoxicated. It should be remembered that the issues raised by any one case are complex and the application to situations in real time is difficult. Laws and regulations governing blood draws for alcohol testing vary from state to state and are the subject of frequent review and amendment. Medical staff should review with hospital counsel the local laws and regulations that pertain to these issues. The benefits of determining a patient’s blood alcohol concentration must be weighed against the risks of the procedure. For example, drawing blood in an agitated patient may place the staff at risk for a needle stick and the patient at risk of vascular injury, It is important to remember that the role of physicians are to be the patient’s advocate and they are not serving as representatives of the court. Physicians must order tests they deem to be of clinical benefit and that may assist the court as necessary in obtaining evidence. One way to assist is talking to the agitated patient; if the patient is agitated because of a defense emphasizing that no drinking of alcohol occurred then it would be a good idea to have blood drawn to support that, as refusal of a blood draw is frequently perceived as a presumption of guilt by the court.

It should also be noted that the US Supreme Court in Missouri v. McNeely limited the holding of Schmerber for warrantless blood alcohol testing.¹⁸ In McNeely, the state court in Missouri argued that the metabolism of alcohol over time, by itself, was justification to obtain warrantless blood alcohol testing in the hospital. The US Supreme Court rejected this argument. While the dissipation of alcohol from the bloodstream over time is one of the factors to consider when evaluating the “totality of the circumstances,” it does not meet the exigency exception to the warrant requirement by itself. The majority opinion held that “when officers in drunk driving investigations can reasonably obtain a warrant before having a blood sample drawn without significantly undermining the efficacy of the search, the Fourth Amendment mandates that they do so.” This does not suggest that a physician should insist on a warrant when an officer requests a blood alcohol test as this is a protection afforded retrospectively by the courts. Familiarity with local regulations is necessary to properly manage requests by law enforcement for blood alcohol testing.

When patients agree to have their blood drawn, it is important that a chain of custody process be carefully followed. The chain of custody process in a legal context refers to the chronological documentation showing the collection, transfer, analysis, and storage of a blood sample from the time it is collected until it is presented at trial. It is used to prevent contamination of evidence. If the proper chain of custody process is not maintained, the evidence collected in general will not be admissible at trial. If the evidence is admissible, and if there are deficits in the chain of custody, the defense attorney will try to cast doubt on the reliability of the information presented.

CONFIDENTIALITY

Patient 5

A 32-year-old woman was fired from her job as a high school mathematics teacher. The school board called for her termination after learning that she had a history of alcohol dependence and had a previous hospitalization for detoxification at the local community hospital. A parent on the school board was also employed as a nurse at the hospital where the teacher had received therapy. The school board member (and nurse) had inadvertently accessed the medical record of the teacher while caring for a patient with the same last name.

The Health Insurance Portability and Accountability Act (HIPAA) was enacted by Congress in 1996 and signed by President Clinton. Initially, the purpose of HIPAA was to increase the portability of health insurance and allow employees to maintain insurance when they changed jobs. The Act called for the establishment of several provisions, among them an electronic database designed to facilitate the exchange of information between health care professionals, insurance companies, and those involved in financial and administrative transactions.¹⁷ However, the idea of developing an electronic database brought to light already growing concerns regarding the maintenance of patient privacy. For the first time, medical records would be accessible to an unlimited number of people working in health care, everyone from bill processors to pharmacists to clinicians. Could the right to privacy be jeopardized by a system designed to increase efficiency?

Prior to HIPAA, individual hospitals or physician offices designed their own methods for maintaining confidential patient information. Records were maintained on computers in some circumstances and on paper in others. Accessibility to that information was largely regulated by state laws and some federal regulations. Ethical codes of conduct also provided guidance. During the 1990s, however, the weaknesses of the existing systems gained attention as multiple high-profile breaches of confidentiality became public. For example, the medical records of a congresswoman were released to the media during her campaign, making her history of depression and a past suicide attempt public knowledge.³⁰ There were also several cases in which the medical records of hospital employees were read by staff members not involved in the employees’ care, health insurance companies released health care information to employers without permission, and physicians released information to pharmaceutical companies that subsequently solicited the patient.⁴⁶ These breaches of ethics were each a testament to the fact that the right to privacy needed more stringent regulation. If such breaches occurred in the previous systems for recording personal health information, then it was clearly apparent that further opportunities for transgressions would be created with increased access through an electronic database.

The Privacy Rule of HIPAA has become the most well-publicized aspect of this Act among health care personnel. The Privacy Rule governs the use and disclosure of protected health information in the hands of health care professionals, health plans, and health care clearinghouses.¹⁷ The terminology used in the Privacy Rule is extensively defined. The following is a brief summary of the terms used.

Protected Health Information

Protected health information includes any individually identifiable information concerning the past, present, or future health of an individual including medical information pertaining to assessment and treatment of an individual and payment and billing information. All forms of information, written, oral, or electronic, are protected by this rule. Deidentified data is not considered protected health information.

Covered Entities

The term “covered entities” includes any person or business that provides health-related services or products. All those providing health-related services fall within this provision, including, for example, clinicians, pharmacists, medical equipment providers, and other health care professionals. Companies that provide disability insurance, car insurance, or casualty insurance are not included in the rule.^17,37

Health Care Clearinghouses

Health care clearinghouses are entities that compile health care information, such as billing companies or data processing centers.

Institutions are required to provide all individuals with written notice of the institutional privacy policy when each individual first seeks medical care. Patients must be informed of how the institution may use and disclose information. The notice must also describe the patients’ rights, including the right to access their medical information and their right to file a complaint if they believe their rights were violated. The notice must be written in plain language, and the patient must provide written acknowledgment of receipt of the information in the notice.

The Privacy Rule of HIPAA was not intended to impede health care. There are several exceptions to the nondisclosure rule. A covered entity is permitted to use and disclose protected health information for the purposes of ongoing evaluation and treatment. For example, physicians have the freedom to consult with each other, both within and outside their own institution in order to provide clinical care. Additionally, there are several specific exceptions to the Privacy Rule delineated within the legislation—situations in which protected health information will be and often must be disclosed, even without an individual’s permission. For example, activities related to public health, such as reporting communicable diseases; information necessary to report actual or suspected abuse, neglect, or domestic violence; or information pertaining to cadaver organ or tissue donation are all specifically exempt from the Privacy Rule.¹³

Examples of the application of the HIPAA rule occur in everyday emergency medicine. For example, only persons directly involved in a patient’s care are allowed to access and view the chart. Occasionally well-recognized politicians, actors and actresses, or even work colleagues seek medical treatment, and previously hospital workers were able to read their medical records. Today, these patients would be protected from the curiosity of medical personnel not directly caring for them. Another example is communication with family, friends, and even law enforcement. Whereas in the past clinical teams periodically discussed questions about a person’s health care with anyone who asked, today clinicians may only speak with those individuals to whom the patient specifically has given permission.

The HIPAA Privacy Rule specifically addresses consultations with poison control centers. It states:

This is important to remember, and individuals seeking assistance from poison control centers to assess risk after exposures to xenobiotics or who need assistance in managing complex cases should never be concerned that they are violating the HIPAA rule by seeking help. The same is true for managing medication safety. The FDA and pharmaceutical industry rely on postmarketing surveillance to obtain further safety information regarding the toxicity of a medication after it has been granted approval for use. The FDA’s program MedWatch is a postmarket surveillance tool that relies on reports by health care professionals and/or patients regarding the occurrence of harmful effects associated with the use of a medication or medical device. For a more in-depth discussion of the FDA and MedWatch, see Chaps. 134 and 135. Reports made to MedWatch are in compliance with the HIPAA Privacy Rule, as the rule recognizes the need for public health authorities to have access to protected health information to carry out their mission of ensuring safety and health to the public at large.³⁶

Violations of the Privacy Rule are subject to penalties, the severity of which is dependent on the type of infraction. Simple noncompliance will result in financial penalties. Moreover, significant or intentional disclosures of information carries steeper fines in addition to criminal charges and potential imprisonment.⁴⁶ In 2009, as part of the economic stimulus plan, the Health Information Technology for Economic and Clinical Health Act (HITECH) under the American Recovery and Reinvestment Act (ARRA) was passed.²³ This Act was primarily created to encourage the use of electronic health records and supporting technology. However, it also widened the scope of privacy and security protections defined by HIPAA. Several notable changes brought forth by this act include an increase in civil penalties for willful neglect, extending up to $250,000 for a single violation, to as much as $1.5 million for repeated or uncorrected violations. Furthermore, HIPAA provisions now apply to business associates. Business associates are defined as persons who, on behalf of a covered entity, perform or assist in performing a function or activity that involves the use or disclosure of individually identifiable health information.

OTHER LEGAL CONSIDERATIONS FOR POISON CONTROL CENTERS AND CERTIFIED SPECIALISTS IN POISON INFORMATION

Patient 6

The poison control center received a call from a concerned mother that her daughter might have ingested one of her grandmother’s diabetes medications. The mother stated that the child was acting normally all day at the grandmothers’ house, but when the family returned home, the child had become drowsy. When contacted, the grandmother had confirmed that one pill was missing from her purse although she did not know the name of her medication. The poison control center advised that the parents give the child juice and closely observe her over the next 6 hours. Approximately 2 hours later, while sleeping, the child had a seizure. The child continued to seize in the hospital, where medical evaluation revealed hypoglycemia. The patient subsequently suffered permanent neurologic damage. The medication was later identified as glyburide. A negligence action was brought against the poison control center, alleging inappropriate advice and failure to recommend transport to a hospital.

As a general rule, any physician who decides to treat a patient enters into a physician–patient relationship and this creates well-established legal duties. Courts have ruled that the physician–patient encounter need not be a face-to-face interaction to have legal consequences. For example, the absence of physical contact between a physician and patient, as in the practice of radiology and pathology, does not preclude a patient from asserting that a duty of care exists.⁶ More particularly and quite relevant to the practice of a poison control center, a New York State court ruled that an initial telephone call from a patient to a physician can be sufficient basis to hold that physician responsible for inappropriate advice or a significant error in judgment.²⁰ Given these legal precedents, it is clear that contact with a certified specialist of poison information provides sufficient grounds for a subsequent legal action if inappropriate advice was given and that advice was a proximate cause of patient injury.

Standards of Care Applicable in Poison Control Centers

Standards of care applicable to medical toxicologists who supervise or direct a patient’s care in a poison control center are examined under the same legal framework as other medical specialists. The basic medical malpractice concepts are universal, with some variations from state to state. Generally, for patients to prevail in medical malpractice cases, they must demonstrate “by a preponderance of the evidence: (1) that the medical toxicologist’s supervision of the care fell below the ordinary standard of care expected of a physician in his [or her] medical specialty, and (2) the existence of a causal relationship between the alleged negligent treatment and the injury sustained.”⁴⁴ While the standard of care is recognized not to ensure perfection, what constitutes the “ordinary standard of care” has much room for interpretation.

In the 19th century, courts determined what the “ordinary standard of care” was by introducing testimony from physicians practicing in the community where the event occurred. This was known as the strict locality rule.⁴¹ The strict locality rule was intended to prevent the inequities of comparing rural physicians working with limited resources and under exigent conditions from physicians working in large urban hospital settings.¹⁵ In many jurisdictions, however, the strict locality rule was rejected because (1) it was difficult to find an expert witness in a small community to testify against another physician in the community, and (2) the strict locality rule permitted some small medical communities to set unacceptably low standards of care.⁴¹ In response, some states adopted the modified locality rule, which compares the physician in question with physicians practicing “in similar localities.”⁴¹ Over time, the basis for the modified locality rule was also called into question. Advances in transportation, communication, and education continued to minimize the disparity between rural and urban medical practice. Most states have abandoned the locality rule altogether, permitting the introduction of evidence of nationwide medical practices. As described by one court:

Although there is movement toward reviewing nationwide medical practices, depending on the jurisdiction where the event occurs, any one of the above rules may apply. In New York, courts allow evidence from the specific locality where the event occurred, from statewide practice, or from nationwide practice.¹⁴

A discussion of standard of care for certified specialists in poison information should also mention several operational aspects of poison control centers. Certified specialists in poison information are required to have rapid and accurate access to a standard information resource system that contains both basic information and recommendations to deal with most toxic exposures. If a patient were to bring a negligence action against the poison control center, the plaintiff might attempt to demonstrate deviations from the standard recommendations in these resources.

It would be inaccurate to suggest, however, that the care required by a certified specialist in poison information should be measured only by how closely the advice given compares with these standard resources. Frequently, a certified specialist in poison information encounters situations that cannot be managed in accordance with an information system alone, and may seek consultation from a clinical pharmacist or a medical toxicologist working with the poison control center. For example, in this chapter, patient 2 and patient 6 are examples of complicated cases that would involve the expertise of a medical toxicologist. In the above case, the report of only one pill missing and possibly ingested can create a feeling of false reassurance. Life-threatening hypoglycemia can occur after the ingestion of a single pill in children. Additionally, the onset of hypoglycemia may be delayed up to 21 hours after ingestion (Chap. 47).²⁹ Knowing this, the parents should be advised to bring the patient to a hospital where fingerstick glucoses can be monitored every hour, and dextrose can be administered if necessary. If the certified specialist in poison information were to enlist the help of a medical toxicologist then any subsequent legal proceeding would also carefully review the accuracy of the content of the information provided to the consultant.

State Statutes Limiting Liability of Poison Control Center Consultants

A comprehensive review of state-enacted legislation providing additional liability protection to poison control center consultants is outlined elsewhere.⁸ These state statutes provide either immunity from malpractice claims or indemnity for a successful claim against a poison control center consultant. The statutes are state specific and poison control center consultants should review them with local counsel, particularly if the statute has not been challenged in the jurisdiction in question.

Several states have enacted laws granting immunity to poison control center consultants. Immunity is an affirmative defense to medical malpractice liability. In other words, even if a lawsuit were to satisfy all of the elements for a successful malpractice claim, the lawsuit would not lead to compensation for injuries sustained because of this state-conferred protection. Examples of states that have enacted statutes providing immunity include Arkansas, California, Florida, Louisiana, Ohio, Tennessee, and Washington.¹ These statutes are not absolute protections from liability. State statutes granting immunity usually stipulate that the poison control center consultant act in good faith while performing the duties required. In some instances, the poison control center consultant must have been providing information in accordance with protocols established by the poison control center. These laws do not protect against gross negligence, nor do they provide protection against intentional misconduct by the poison control center consultant.

Other states have enacted laws that require the state itself to indemnify poison control center consultants in the case of a successful malpractice claim. In other words, a patient may bring a medical malpractice claim against a poison control center consultant, but if the claim is successful, the state itself compensates the patient rather than the poison control center consultant. States that have enacted indemnity statutes include Illinois and Texas.⁴⁸ Statutes providing state indemnification for medical malpractice claims are also limited in that they do not provide protection in the case of gross negligence, nor do they provide indemnity in the case of intentional misconduct.

In addition to state statutes that specifically limit liability for poison control center consultants, it has been argued that some states may hypothetically provide protections under common law or general “public immunity” statutes.⁴⁷ Neither poison control centers nor poison control center consultants are specifically delineated under these state laws, but may still be afforded immunity protection. States with general common law or “public immunity” statutes include North Carolina, Missouri, Illinois, Connecticut, and Georgia. Finally, many states have sovereign immunity statutes that afford liability protection to any entity created, funded, and operated by the state. Poison control centers and staff may be considered such an entity and, therefore, granted such protection.

Practices of Regional Poison Control Centers That Can Reduce Potential Liabilities

There are some inherent risks creating potential liability for a poison control center. To minimize these risks and the subsequent risk of civil action against a poison control center, quality assurance and risk management programs should be a regular function. Daily audits or monitoring of the advice given by certified specialists in poison information should be conducted. Such processes enhance care and ensure patient safety for the individual and establish a higher general standard of performance.

The medical toxicologists and clinical pharmacists responsible for supervising the certified specialists in poison information must be able to adequately assess the competence and capabilities of the staff, make recommendations, take corrective actions, and provide suggestions for improvement to involved members. This process is facilitated by such actions as audiotaping calls made to the poison control center, and the subsequent advice given, and reviewing written records maintained by the information specialist on each particular case. Documentation is extremely important because in the event of a lawsuit, the most likely area of dispute will be what was actually said to the patient.

REPORTING IMPAIRED PHYSICIANS

It is well recognized that physicians suffer from substance abuse and misuse disorders just like the general population. Though there are few formalized studies on physician impairment, there are certain factors unique to the medical community, such as knowledge of the medications and ability to prescribe that leads to easier access.³¹ Impairment in the medical profession is a serious issue as physicians make daily decisions that affect the health of their patients, and their conduct has the potential to affect the quality and safety of the health care delivered. In 1973, the American Medical Association (AMA) formerly recognized physician impairment in its landmark paper “The Sick Physician: Impairment by Psychiatric Disorders, Including Alcoholism and Drug Dependence.”³⁴ This paper declared that physicians had an ethical obligation to recognize a colleague’s impairment due to substance abuse.³ Subsequently, local, state, and national programs designed to help physicians struggling with impairment were formed. Today there are physician health programs (PHPs) in all 50 states, and they are administered by the state medical boards or medical societies. Physicians may self-refer to these programs or be referred by a colleague or an administrator. These programs seek to provide detection, assessment, and referral to treatment facilities. Physicians entering contracts with PHP can maintain confidentiality and may keep their license after completing treatment and avoids formal disciplinary action.³¹ Identifying an impaired colleague may be challenging as physicians can be good at hiding such disorders until it becomes severe. Signs to look for include deteriorating social and interpersonal behaviors, inaccessibility to patients and staff, frequent conflicts with staff members, unexplained absenteeism, and frequent moves to different locations for jobs.³¹ Colleagues may be hesitant to report such behaviors because of fear of damaging a career or the fear of confrontation. However, in addition to an ethical obligation as determined by the AMA, many states have laws in place requiring the reporting of impaired colleagues.

SUMMARY

• The risk management and legal issues of an ED with patients exposed to xenobiotics, often with impaired judgment, are complex. The substantial practical challenges in initiating emergent treatment to an individual unable to provide proper informed consent vary depending on individual state laws.

• Medical care is further complicated by the difficulties inherent in documenting these interactions with our patients.

• Both forcible restraint (often desired by the physician) and evidence collection (requested by law enforcement) have significant potential legal ramifications for the care provider.

• Maintaining patient confidentiality outside the ED, despite circumstances often visibly volatile within the department, often seem futile, yet legal obligations to protect patient privacy are stringent.

• A hospital must ensure a close working relationship among the legal department, risk management, and medical personnel. Only in this manner can everyone learn, cooperate, and meet each other’s needs in the ever-evolving and challenging clinical scenarios confronted.

Acknowledgments

Walter LeStrange, RN, MPH, MS, Kevin Porter, Esq, and Danius A. Drukteinis contributed to this chapter in previous editions.

REFERENCES

INTRODUCTION

One of the potential areas of collaboration for toxicologists is the medicolegal setting. This includes working with forensic toxicologists, medical examiners, law enforcement, lawyers, and regulators. In this arena, interpretive toxicology, whether based on analytical findings or theoretical concepts, is routinely used or required to explain issues in criminal and civil proceedings and as the basis for policy making. Often, the concepts associated with forensic interpretive toxicology are presumed to be associated with deceased individuals. In practice, the preponderance of cases involving toxicologic interpretation involves living people and includes specialty testing such as human performance toxicology. Regardless, interpreting the role, or potential role, of xenobiotics in an adverse outcome is typically not straightforward. For example, the use of generally applied pharmacokinetic equations is usually inappropriate, especially in the postmortem setting, yet it is in widespread practice by individuals unfamiliar with the nuances of case-related issues.²⁷ Rarely is any given case accurately interpreted solely based on toxicologic assessment. As such, forensic interpretive toxicology generally involves an integrative approach that draws on an understanding of case history in addition to analytical, toxicologic, pathophysiologic, and specimen-related issues. Individuals called on to interpret the role of xenobiotics in any given case must understand the factors that affect such interpretations and associated limitations. Furthermore, although difficult in many situations, they should attempt to have their opinions firmly supported by science. Other chapters in this text (Chaps. 7, 41 and 140, Special Considerations: SC11 and SC12) cover much of the basic science associated with interpretive issues. This section is meant to focus on specific issues as they relate to medicolegal (forensic) interpretive toxicology.

CASE HISTORY

It is reported that case history plays a role in interpreting toxicologic findings in about 70% of cases.²⁶ A number of specific aspects of a case history will lead a seasoned toxicologist in a direction that narrows the scientific investigative focus and avoids a shotgun approach to defining the role of toxicology in an adverse outcome. Table 139–1 lists some of the more important aspects of the case history that should be evaluated from the outset to begin the interpretive process. Detailed information about the individual should optimally include anthropomorphic information, vocation, hobbies, medical history, timing of events to the extent possible, and other contextual information.¹⁹ As an example of the importance of this information, consider a case in which an individual suspected of drunk driving worked at a company that manufactured chemicals. One of the chemicals known to be handled by the individual was a substance that co-eluted with ethyl alcohol in the gas chromatographic assay used to quantitate the blood alcohol concentration. Without his employment history, the potential for an incorrect interpretation of the data might have occurred.

Unfortunately, toxicology is often not initially considered with respect to field investigation of a case. Therefore, important information is often lost, especially in the case of purposeful poisonings. The longer the time between suspicion of a toxicologic aspect to a case and proper investigation, the greater the chance of lost evidence to achieve an understanding of the situation. The toxicologist can play a key role in mitigating this occurrence. As a significant point of contact with the poisoned individual, the toxicologist has the opportunity to not only ask the individual questions if he or she is conscious, but to also ask police, paramedics, and others questions that could assist in not only treatment of the patient, but also address serious questions that often later arise.

Unlike the confidence in the medical diagnosis of poisoning supported by clinical findings or analytical data, it should be recognized that there is rarely certainty in the use of toxicologic analysis for medicolegal purposes. However, a toxicologist’s conclusions and opinions in this setting can only be attained by integrating pieces of information into a coherent analysis of the events.

PREANALYTICAL ISSUES

Ample evidence exists that the major source of error in analytical toxicology is based not on laboratory error, but on preanalytical issues.^20,41 Preanalytical issues are defined as those events that affect an analytical result prior to the specimen(s) arriving at the laboratory. In accordance with ISO (International Organization for Standardization) requirements, mathematical uncertainty should be able to be determined by the analytical testing laboratory.²⁸ It is an expectation in certain areas of toxicologic testing that such errors be reported for acceptance into legal proceedings.³⁴ Although laboratory error can usually be relatively easily quantified, the error associated with events surrounding samples before arrival at the laboratory is not so easily measured. The myriad of potential preanalytical factors is summarized in Table 139–2.

Clinical Specimens

With respect to living individuals, where and how specimens are collected from patients, in addition to specimen storage, can adversely affect not only the analytical test but its interpretation as well.⁵⁴ For example, although perhaps not protocol, occasionally samples for analysis are collected proximate to the site where medications are administered, such as close to an indwelling antecubital fossa catheter. Contamination of such specimens cannot be excluded and will lead to misinterpretation of findings. Although the treatment of patients is the primary concern during medical emergencies, consideration of potential medicolegal issues should be part of the treating professional’s assessment, especially in cases involving children, victims of drug-facilitated sexual assault, motor vehicle operation, workplace incidents, suspected poisoning, and other ill-defined situations. Adequate analysis requires that specimen collection be performed appropriately and consideration given to the types of preservatives used in the collection device. Not all collection devices fit the need for all xenobiotics. For example, although fluoride-containing tubes work well to prevent microorganism-induced ethanol formation, this same preservative will lead to degradation of some organic phosphorus compounds and make the sample totally useless in cases of suspected fluoride poisoning.^1,49 In general, it is good practice to collect both preserved and unpreserved blood samples. Serum samples are generally a good choice from living individuals, but this is analyte specific and certainly not preferred for xenobiotics sequestered in red blood cells. Perhaps the most important precaution in this respect is the collection of specimens for legal blood alcohol determination. Not only are tubes containing fluoride (1%–5% w/v) and oxalate (anticoagulant) necessary, but skin-cleansing swabs containing ethanol should be avoided.² Specimen storage conditions can significantly adversely affect analyte stability and specimen integrity. Except for specimens such as hair and nail clippings, specimens for toxicologic analysis should not be maintained at room temperature. Minimally, refrigeration is required, but if it is anticipated that a delay between sample collection and testing will occur, then freezing the samples at −20° or −60°C (the choice of temperature being analyte specific) should be considered, making sure to take precautions to prevent tube cracking.¹¹ It is recommended that if a specific xenobiotic or class of xenobiotics is suspected, that laboratory personnel should be contacted before specimen collection for guidance on proper collection, preservation, and storage of specimens. Hospitals should have in place a mechanism that ensures retention of specimens beyond the routine time frame for suspect cases. Typically, hospitals maintain specimens from a few days to a few weeks after collection to up to one year or longer in most forensic cases.^46,61 In this regard, all available specimens should be maintained. This is especially important in cases in which death is preceded by a relatively long clinical course. Although it cannot be expected that a hospital maintain specimens under a legal chain of custody, maintaining tubes with original identifier labels often suffices for this purpose.²¹

Postmortem Specimens

In postmortem situations, the potential for sample contamination is substantial, especially in the face of significant trauma. For example, the spread of gastric or intestinal contents containing xenobiotics of interest can contaminate sources of blood collection and visceral organs.¹⁷ Additionally, other postmortem phenomena can mitigate the potential value of laboratory-based testing (Table 139–3) for interpretation purposes (see the Postmortem Redistribution section later). Specimen collection should be done by an experienced pathologist who understands the intricacies of site-specific sample collection. Postmortem blood can be obtained from 2 broadly categorized sources, central (eg, heart) and peripheral sites (eg, femoral and subclavian veins). Not all peripheral sites have the same interpretive value.⁵⁶ “Blind stick” samples should not be considered appropriate surrogates for proper dissection and visualization of the specimen source.⁴² For example, heart blood should be collected from the right atrium after opening of the pericardial sack, removal of the pericardium, and drying of the heart.^13,65 Although there is no ideal postmortem specimen, femoral vein blood tends to better reflect circulating concentrations of xenobiotics closer to the time of death, but this is not always the case. Even femoral blood is subject to postmortem redistribution (PMR) in a manner similar to other blood sources. Whenever possible, both cardiac and femoral blood should be collected. Because of postmortem changes, heart blood tends to contain higher concentrations of xenobiotics than peripheral blood sites, thus making it a valuable screening specimen but limiting its actual interpretive value. To properly collect femoral vein blood, a dissection and visualization of the femoral vein followed by ligation or clamping of the vessel above the point of collection is recommended to avoid contamination by iliac blood, the latter seemingly representing a cross between peripheral and central blood.¹³ Visceral organ specimens are of value in the interpretation of the meaning of blood concentrations in certain situations.^5,43 Although a plethora of data exist regarding toxicologic findings in blood, there tends to be less data regarding tissues. Nevertheless, certain organs are of substantial value when considering specific xenobiotics (Table 139–4). Specimen collection tubes and storage conditions, although similar to those described with clinical specimens, tend to be more important with postmortem specimens because of continuous postmortem changes in blood pH, autolysis, and cell lysis, which can affect analyte stability, recovery, and interpretation.⁶¹

Hair collection must be considered in practically all cases of toxicologic concern, especially in those involving children because this specimen can circumvent arguments of incidental exposure and other issues.⁴⁷ This specimen, after being collected, requires no special storage conditions and often reflects exposure history for the length of hair growth (∼1 cm/mo). Controversy regarding external contamination of hair leading to positive findings, although of some concern, does not impugn the value of this specimen.^27,36

Lastly, iatrogenic toxicologic injury or death, whether unintentional or purposeful, are often not considered or poorly investigated. The involvement of hospital-based risk management often occurs late and valuable evidence that can either incriminate or exculpate individuals gets destroyed before collection and preservation. Protocols in suspicious cases should exist for the area where the individual was located and considered to be a “crime scene.” That is, potential useful evidence, including devices, medications, and other xenobiotics should be gathered and stored.⁵⁸

SPECIAL LABORATORY CONSIDERATIONS

Testing for xenobiotics of interest is varied and inconsistent among laboratories. Hospital-based laboratories tend to be limited in scope compared with toxicology reference laboratories. Even in this latter group, wide variances exist with respect to capability and scope. No interpretation of toxicologic findings should take place without a complete understanding of the testing performed and its limitations in addition to a review of the analytical data by a toxicologist. With the development of robust, widely available analytical tools, inadequate testing with respect to medicolegal interpretive matters should no longer occur.

Ideally, all toxicologic analyses are performed under chain of custody, and all findings are confirmed by a separate and distinct analytical test, with at least one, such as mass spectrometry, that provides specific molecular identification.⁴⁸ This process reduces uncertainty in the utility of a finding for interpretive purposes. For example, the use of gas chromatography to identify ethylene glycol in a hospital-based laboratory led to the wrongful conviction and jailing of a mother for poisoning her infant son. The gas chromatographic peak was actually propionic acid. Remarkably, the reporting of ethylene glycol was duplicated by another laboratory. Post-conviction analysis determined the misidentification of the peak. In this case, the propionic acid resulted from the inborn error of metabolism disease, methylmalonic acidemia. 2,3-Butanediol also co-elutes with ethylene glycol in some gas chromatographic systems.^17,63

It is uncommon for hospital-based drug testing to be analytically confirmed. However, a positive screen using an immunoassay is rarely sufficient for medicolegal purposes given the general cross-reactivity of these tests. Most hospital laboratory reports include a disclaimer to this effect. Also, most of this testing is performed using urine, so such screening tests are neither quantitative nor interpretable with respect to effects on a given individual.

Today, advanced toxicology laboratories use a combination of tools to help identify xenobiotics of toxicological concern. Many of these tests are performed rapidly and, in some sense, inexpensively compared with the offered breadth of testing. Instruments such as liquid chromatography time of flight mass spectrometry (LC-TOF) have revolutionized broad-based toxicology testing. This technique allows for the rapid screening of a patient sample in minutes covering hundreds of chemically dissimilar compounds with identification using exact mass.²⁵ Liquid chromatography/tandem mass spectrometry (LC-MS/MS), although somewhat less amenable to broad-based screening, has a similar utility with an easier approach, in general, to quantitative analysis.⁴⁴ Both devices will be found in some larger hospital-based laboratories and should help the toxicologist significantly in evaluating potentially poisoned patients.²⁴ Other instruments, such as inductively coupled plasma mass spectrometry and inductively coupled plasma optical emission spectroscopy used for the analysis of the majority of metals, are typically found in specialty laboratories and most likely will not be cost effective in the hospital laboratory (Chap. 7). In all cases, however, the toxicologist should understand the limitations of these devices that can lead to false-positive and false-negative findings.

CORRELATING FINDINGS OR THEORY TO CASE-RELATED ISSUES

Ultimately, the toxicologist’s role in forensic interpretive toxicology is to integrate analytical findings and theory to develop an explanation for an adverse outcome. Although explanations abound, both in concept and practice, plausible explanations are often few. It is rare that findings are simply taken at face value. For example, it is common for identical findings in 2 different cases to have 2 completely distinct interpretations, which should not be surprising or disconcerting. For example, the finding of a high concentration of morphine in the blood of a hospice patient would be expected and often not related to the cause of death, whereas the same concentration in a naïve heroin user would represent a cause of death. Toxicology has been described as both science and art, with the interpretive aspect based in science, knowledge, and experience.²² Many of the following factors need to be considered in rendering rational opinions.

Problems With Analytical Interpretations

Commonly, individuals render toxicologic opinions without careful review of the analytical data that help form the basis of the opinions. Acceptance of reported results at face value is a fundamental flaw in medicolegal interpretations. It is helpful to review all available information that led to a reported result before interpretation. Table 139–5 lists some of the important facets of analytical data review that lead to incorrectly reported results. Although there is no expectation that the toxicologist be an expert in analytical chemistry, familiarity with the key elements that lead to false-positive, false-negative, or poorly quantified findings should become part of the review process. Errors ranging from the wrong patient number on the analytical data to poor control and calibration of an analytical run to poorly integrated chromatographic peaks are all examples of errors that invalidate any given finding.

Postmortem Redistribution

There are many factors related to postmortem changes that complicate the interpretation of well-performed toxicologic test results (Chap. 140).^18,19,40,64 However, the importance of PMR, a long-established principle in forensic toxicology, cannot be ignored or underestimated. Although recognized in the early 1980s, Prouty and Anderson authored the first codified report of this phenomenon in 1990.⁵⁷ Succinctly, PMR is the term given to changes in site-specific xenobiotic concentrations after death.⁵³ These xenobiotic movements postmortem generally result in elevated concentrations in common collection sites such as heart blood. The elevation for some xenobiotics are dramatic, giving the appearance of overdose when no such conclusion is supported by other evidence, including analysis of sites of blood collection where PMR may occur to a lesser degree, such as femoral blood. Although instructed to do otherwise, many pathologists still collect heart blood as a sole source of postmortem blood. In attempting to assist in such situations, investigators often publish ratios of heart blood concentrations to femoral blood concentrations over a series of cases from a given laboratory. The common observation is a widespread range of values often encompassing 3- to 5-fold or greater differences in magnitude.¹⁰ The use of such ratios to convert a heart blood concentration to a femoral blood concentration, and by further analogy to a premortem circulating blood concentration, should not be performed. The greatest use of such ratios is to give an idea of the likelihood of PMR for any given xenobiotic based on the mean and range of ratios.⁴⁵ Because it cannot be predicted if PMR occurred in any given case and if so to what degree and over what time period, such calculations have no basis for antemortem blood concentration determination. Another confounder to the use of ratios is the possibility that site-specific differences are not due to PMR but merely a reflection of incomplete distribution before death.¹⁹ It is unclear how many of the reported ratios of heart blood to femoral blood actually reflect this process as opposed to true PMR. The proposed mechanisms and influencing factors resulting in PMR are varied and are listed in Table 139–6. It is most likely a combination of factors that lead to PMR in any given case. Interestingly, as noted earlier, PMR is also reported to take place in femoral blood; thus, it should not be taken at face value that any postmortem blood concentration accurately reflects the circulating concentration at and around the time of death.

At one time it was believed that PMR was associated with basic xenobiotics with a large volume of distribution. Seemingly, however, most xenobiotics, including acidic and neutral xenobiotics, undergo PMR. Furthermore, PMR occurs for xenobiotics with wide-ranging volumes of distribution. Still other xenobiotics that would be predicted to undergo PMR have experimental evidence to the contrary.⁵³ Thus, a priori predictions of PMR for any given xenobiotic without experimental evidence should explicitly state the caveats associated with such a conclusion. It must be stated, however, that PMR does not preclude the interpretation of findings for cause of death determination because the findings represent only one piece of data in a potential myriad of other relevant information.

Another similarly related phenomenon is postmortem diffusion of xenobiotics. In this situation, diffusion of xenobiotics from an area of high concentration to that of a lower concentration occurs. This is most noted when the gastric contents contain a substantial amount of a xenobiotic that migrates across the wall of the stomach into neighboring tissues and blood sources, such as abdominal aorta and iliac vessels. Additionally, perimortem aspiration of gastric contents can lead to significant esophageal, tracheal, and lung concentrations of a xenobiotic, further facilitating diffusion processes.^12,55 One last concern is the administration of drugs via such mechanisms as a central line, whereby it can be reasonably expected that drugs will continue to accumulate in and around the heart postmortem.

Alternative Matrices

Although most discussions regarding interpretive toxicology tend to focus on blood concentrations of xenobiotics, clearly in both the living and the dead, alternative matrices are extremely valuable. It should be recognized that there is no perfect specimen to assess exposure or provide analytical findings in support of impairment. In the living, alternative specimens including hair, oral fluid, and breath are useful. Hair gives a longer window of detection than most other easily obtained specimens.⁶⁵ Analytical precautions, such as rinsing procedures, are used to minimize some concerns with hair testing.⁵⁹ The utility of hair should not be underestimated, and despite some of its limitations, it can represent the best suited specimen in some cases, such as drug-facilitated sexual assault, especially when more than 24 to 48 hours passed before specimen collection from the time of event, thus often limiting the value of urine or blood testing.³⁸ In such cases, waiting 1 to 3 months before collecting hair will capture the potential exposure period in growing hair. It is especially useful in children to demonstrate acute versus chronic exposure to xenobiotics.⁴⁷ Collection of hair should be from the posterior vertex of the head, which can be easily approximated by drawing an imaginary line over the head connecting the tops of both ears and a line from the bridge of the nose to the nape of the neck.^13,59 The area where the 2 lines meet in the back of the head provides hair that grows most consistently on the head.³⁷ About a pencil thickness worth of hair should be clipped as closely to scalp as possible.¹⁴ The root ends should be identified either by tying with string or wrapping in foil.¹³ Virtually any other sources of hair are suitable for testing, such as pubic and axillary hair, although practical reasons and shorter growth rates tend to make such samples less suitable for testing.

Oral fluid (OF) is gaining ground as a viable specimen type to assess exposure and impairment because obtaining it is noninvasive, and it can be collected at the site of an event, thus eliminating time delays.⁵ Such delays will result in clearance of a xenobiotic from other specimen types before collection. Oral fluid is composed of secretions from the parotid, submaxillary, sublingual, and other smaller glands. Numerous specimen collection devices exist today that circumvent the need to expectorate, which is not a desirable means of specimen collection.⁵¹ For the most part, non–protein-bound parent compounds appear in OF, although there are exceptions (eg, benzoylecgonine).¹⁵ Factors that affect how much of a xenobiotic gets into OF include pH of the blood and OF, the pKa of the xenobiotic, and the degree of protein binding. In general, basic xenobiotics get into OF more easily than acidic xenobiotics.⁶⁸ Oral fluid has been used for therapeutic drug–monitoring purposes (eg, theophylline and digoxin).⁸ Although some correlations between blood concentrations and a number of xenobiotics have been made, it must be remembered that OF is susceptible to contamination from residual drug in the oral cavity, smoked xenobiotics, and passive exposure.^5,15 Despite its promise, there is variability in OF concentrations of xenobiotics not only within the same individual but also among individuals based on influencing factors that facilitate or inhibit secretion of a xenobiotic into OF.⁵ Finally, based on the relatively small specimen volumes after OF collection, specialized testing, including tandem mass spectrometry, is often needed to detect the concentrations of xenobiotics present.

Breath testing for alcohol is a mature subject matter, yet when it comes to impairment, there is a current trend for states to move back to blood alcohol determinations. Legal arguments are the force behind some of this movement.⁵² The United States Supreme Court, however, rendered a decision in 2016 whereby blood collection related to driving under the influence can only be obtained after issuance of a search warrant, declaring blood collection to be invasive compared to breath testing. It is unclear how this latter decision will ultimately affect the decision as to what specimen is collected for analysis in driving under the influence cases.⁴ In living individuals, particularly in the hospital setting, serum alcohol is typically measured. As such measurements for alcohol are often used in the medicolegal setting, it is imperative that toxicologists understand the differences between blood and serum alcohol concentrations and that such differences are related to attorneys (Special Considerations: SC11 for greater detail).

Breath testing for other xenobiotics is still in its fledgling state. Despite this, the future holds promise for the detection of other xenobiotics in breath with the possibility of roadside breath drug-testing devices.³ The major advantage of this specimen is not only its lack of invasive nature, but rapid screening capability.

Alternative matrices from decedents are varied and include virtually all fluids and tissues (Table 139–4). Practically speaking, however, other than providing qualitative information, for most matrices, there is little data to support utility of a measured concentration to either allow for independent interpretation or correlation to blood concentrations. Although numerous studies exist that suggest such correlations, rarely is there consistency between studies. Even so, situational use of alternative matrices is sometimes warranted (eg, the use of fat or brain to detect volatiles). A specific exception is vitreous humor (vitreous), a specimen that is often extremely valuable postmortem. Vitreous electrolyte measurements provide data that cannot be gleaned from virtually any other specimen type (Table 139–7).⁹ For most xenobiotics, vitreous concentrations lag behind blood concentrations by about 1 or 2 hours. Thereafter, some correlations have been made between vitreous concentrations and those in blood, with the greatest example being ethanol, where there is an approximately 1:1 correlation when equilibrium is reached. Vitreous represents a relatively pristine specimen during decomposition, embalming, exsanguinations, severe peritoneal trauma, and other less ideal situations.²³

Toxicokinetics, Toxicodynamics, and Drug Interactions

The basics of toxicokinetics (TK) and toxicodynamics (TD) are covered in Chap. 9. The large and unpredictable degree of interindividual variability places practical limits on the use of TK and TD in postmortem interpretation. These same variations affect the pharmacologic response in living patients. Multiple elements related to TK and TD affect the ability to render interpretive opinions of xenobiotics related to an adverse outcome (Table 139–8) and the use of pharmacokinetic and toxicokinetic equations warrants special attention.

Toxicologic emergencies and overdoses alter normal physiological functions. Liver and kidney dysfunction significantly alter normal TK parameters.^62,66 Coupled with pharmacogenetics, drug–drug or drug–xenobiotic interactions, pathophysiology, and other significant factors, the use of routine TK equations usually leads to grossly under- or overestimates of a specific measured value. Nevertheless, in a living patient, TK equations are useful to estimate certain parameters (eg, dose and, in part, form the basis of such useful tools as nomograms). After death, because of PMR, postmortem diffusion and other issues, the use of TK formulas to predict premortem concentrations or dosing is not generally an acceptable practice. For example, the use of the following formulas will lead to either an absurd dose calculation or an inappropriate estimate of predicted concentration compared with that reported for compounds undergoing significant PMR:^16,29

Nevertheless, such calculations, although ill-advised, are routinely performed and can provide grossly misleading information. “Back extrapolating” from a postmortem concentration to some concentration earlier in time using half-life is fraught with the same perils in that the initial assumption of an unchanged postmortem blood concentration should not be made.¹⁸ Additionally, it is never known how much of any measured blood concentration resulted from single or multiple dosing exposure to a xenobiotic. Despite these and other issues, the toxicologist will be pressured into providing dosing information or predicting a blood concentration during life. The safest avenue in this respect is not to accede to such pressures because the scientific underpinnings are unreliable.

It should be stressed that metabolic drug interaction considerations alone, together, or in conjunction with additional information informs the toxicologist of many varied aspects of exposure and potential effect and outcomes. Often, the metabolic fate of a substance is under genetic control, but it must be understood that genotype does not often relate to phenotype, thus complicating interpretation of toxicologic findings or adversely affecting theoretical interpretive toxicology. Drug interactions occur through varied means, including metabolic inhibition or induction, receptor competition, allosteric modulation, etc. (Chap. 11). All such interactions affect the efficacy and toxicity of a given compound. In some cases, these effects reveal themselves through such factors as ratios of metabolite to parent compound, metabolite to metabolite ratios, etc. One such example is the determination of heroin use from postmortem blood and blood from impaired drivers. In most cases, a ratio of free morphine to free codeine greater than unity is strong evidence that the decedent used heroin. The same appears to hold true in the blood of living people. These latter findings will not hold true when exposure to both heroin and codeine together occurred or when the individual is exposed to both morphine and codeine simultaneously.^7,31

Assessing Impairment

There exist 2 broad areas of concern with respect to impairment, that from ethanol and that from other xenobiotics. Impairment caused by ethanol has been extensively studied, and general conclusions are usually reached with some confidence. The same cannot be stated for impairment caused by other xenobiotics. Ultimately, conclusions of xenobiotic impairment are based on a number of factors that include physical examination by a trained clinician, dosing information, length of time taking a xenobiotic, co-administered xenobiotics, pathophysiology, observed behaviors, measured concentrations in blood or serum or plasma, and any other useful information.³⁵ The use of the term “consistent with” is common in assessing impairment in any given individual and might very well be the best opinion that can be offered. Mitigating factors, including tolerance, single versus multiple dosing, reason for the presence of the substance, timing and route of administration, and pathophysiology, all make firm conclusions difficult, especially when blood findings are the first indication of potential impairment (ie, no visible signs of impairment).^30,60 Great care must be exercised in evaluating or predicting impairment from xenobiotics given the implications of such conclusions. There is currently no scientific support for back extrapolation of xenobiotic concentrations based on blood findings. It cannot be stressed enough that urine findings should never be used to assess impairment because the predictive value is limited.⁶⁷ Similar to blood ethanol concentration, many states and countries have developed per se xenobiotics laws that presume guilt by the presence of certain drugs or metabolites over some prescribed reportable concentration, even in urine, thus mitigating the need for interpretation in most cases. For example, the presence of Δ⁹-carboxy-tetrahydrocannabinol, an inactive metabolite of marijuana, is used to convict drivers of impaired driving; for clarity, this is a legal issue, not a scientific, issue.^33,39 Tables and texts that list xenobiotics and determined concentrations in blood and other specimens should only be used as a guide for interpretive purposes because every human toxicologic experience is unique.

REASONABLE MEDICAL CERTAINTY

It is common for experts in the medicolegal setting to be asked to present opinions and conclusions with a reasonable degree of medical or scientific certainty. Although on the surface such a phrase appears to connote some level of confidence in what one opines or concludes, it has been declared a nebulous phrase without true meaning and at its worst, misleading for a trier of fact.⁵⁰ The derivation of such a phrase is most likely a legal construct intended to persuade a trier of fact that the particular opinions of the retained expert are somehow weighted to such a degree that no other opinion or conclusion would make sense. However, that 2 toxicologists can give 2 distinctly differing opinions in the same case, both to a reasonable degree of medical certainty, demonstrates the lack of utility of the phrase. It is not common parlance for physicians or scientists to discuss matters with a reasonable degree of medical or scientific certainty and it is unclear why the term should carry into the medicolegal setting. Undoubtedly, there will be a slow retraction from the use of the phrase, but toxicologists can push the measure forward through discussions with attorneys prior to expert report generation or testimony by declaring that use of the phrase cannot be supported in a medical or scientific context.

SUMMARY

• Toxicologists have the opportunity based on both contact with patients and as consultants to be involved with medicolegal cases.

• To properly assess such cases, a holistic approach should be taken that involves the integration of detailed historical information, preanalytical factors and understanding of the analytical work performed (including its limitations), specimen-specific utility, and special aspects that can weaken or strengthen the ability to form medically and scientifically sound conclusions.

• It should be understood that the burden on the toxicologist is not generally “absolute certainty” because this is something that is rarely achievable. However, the toxicologist must develop opinions and conclusions based upon sound medical and scientific principles and, if asked, be able to identify factors that weaken the stated opinions and conclusions.

• The purpose of the proffered opinions and conclusions offered should not be lost, that is, to educate someone or a group of individuals who have to make difficult decisions, some of which infringe on civil liberties and freedoms. In that respect, the burden assumed by toxicologists should be recognized as significant and warrant the same attention to detail with medicolegal opinions as treatment and diagnosis of a poisoned patient.

REFERENCES

INTRODUCTION

Myriad medical texts describe the medical consequences of acute and chronic ethanol use. The purpose of this discussion is to describe the history and use of assessments of ethanol-induced impairment, primarily in a legal context. Some of the principles described, however, are extrapolated from the legal to the medical setting. Throughout this chapter, the terms ethanol and alcohol are used interchangeably in order to retain the forensic, legislative, and medical contexts in which the terms are used. Similarly, words such as drunk and punishable are also occasionally used in this chapter in order to preserve the specific language used in the early studies and legal writings.

HISTORY

Numerous well-designed laboratory studies assessing alcohol-induced impairment are published. However, application of study results to actual cases of possible impaired driving or over service of alcohol in a drinking establishment may be inaccurate without an understanding of the origin of laws governing these activities. Although the intoxicating effects of ethanol have been known for centuries, the advent of mechanized transportation spawned increased public and legal scrutiny. Concern was expressed over the potential adverse safety implications of driving while intoxicated, not just for the impaired driver but also for other persons (passengers, other drivers, pedestrians) and property. Although railroads had regulations against the operation of equipment while intoxicated dating back to at least the 1850s, the first arrest for drunk driving in an automobile was that of a London taxicab driver in 1897.²² The realization that alcohol intoxicated motor vehicle operators were a public health issue worthy of legal scrutiny soon followed, and legislation was enacted to combat “drunken driving.” In 1910, New York was the first US state to enact such a law.²²

In early laws, “drunken driving” was poorly defined and relied heavily on observable signs of gross intoxication. Attempts at legislative clarification of language resulted in terms that were nearly as nebulous, such as “alcohol-impaired” and “under the influence.” Prior to the landmark work by Widmark⁶⁷ in Sweden, and by Heise²⁷ in the United States, evaluations of driver impairment were predominantly based on the “expert” testimony of an evaluating physician and the arresting police officer, as well as behaviors reported by witnesses. These evaluations were observational, rather than scientifically based, and such nonsystematic and nonobjective testimony was often dubious and frequently plagued by exaggeration of behaviors (eg, staggering gait, incoherent speech).

The development of analytical technology capable of measuring the concentration of alcohol in blood gave rise to the idea that diagnosis of intoxication might be assisted by an objective chemical test result. The theory behind this idea being simply, the higher the blood alcohol concentration (BAC), the more drinks the individual must have consumed and, therefore, the greater the degree of impairment. It was, therefore, reasoned that there is a legally punishable limit of alcohol in blood or other biological matrix. Although the assignment of a specific clinical effect to a given BAC is not rigorously applicable to evaluation of an individual case, the general trends when examined across a population formed the foundation of many modern driving while intoxicated (DWI) laws.

Despite centuries’ worth of knowledge of the effects of alcohol and decades of efforts to articulate drunk driving standards, a single universal definition of “alcohol intoxicated” does not exist. A typical medical definition focuses on altered mental status, ataxia, or the ability to care for oneself, whereas the focus of a legal definition is on the ability to safely operate a motor vehicle or successfully perform field sobriety tests, or whether the patron of a bar or restaurant should be served additional alcohol. A lay public definition may include parts of both.

Although alcohol intoxication is an important factor in a nearly infinite number of settings, 2 of the most common legal circumstances involving toxicology consultation for alcohol-related issues are (a) alcohol-impaired operation of a motor vehicle and (b) so-called “dram-shop” cases when an already intoxicated individual is served additional alcohol. Discussion of alcohol use in the workplace and alcohol-abstinence monitoring are beyond the scope of this chapter.

PER SE LIMITS

With the advent of analytical measurements capable of quantitatively determining BAC and the increasing public awareness of the dangers associated with drinking and driving, legislation based on BAC began to appear. Although the initial laws did not define the specific BAC that established illegality in the operation of a motor vehicle, they did allow for the use of BAC results as supportive evidence of intoxication. The first states to incorporate BAC into drunk-driving statutes were Indiana and Maine, which did so in 1939.²⁸ The approach taken by the Indiana legislature resulted in a 3-tiered statute, which stated that a BAC of less than 50 mg/dL was considered presumptive evidence of no intoxication, a BAC between 50 and 100 mg/dL was considered supportive evidence of intoxicated driving, and a BAC of greater than 150 mg/dL was considered prima facie (ie, obvious and evident without proof) evidence of guilt. From a legal perspective, prima facie evidence shifts the burden of proof from the accuser having to substantiate the charge to the defense to rebut the allegation. It is this legal perspective from which the so-called per se standards are derived—that is, an individual whose BAC exceeds a predetermined concentration is deemed guilty of driving while impaired by alcohol, even without any other evidence of intoxication or impairment.

Early drunk-driving standards established in the Uniform Vehicle Code made reference to a BAC at which there was “no doubt of obvious intoxication”; this BAC was defined as 150 mg/dL. However, by 1960, as data regarding alcohol-related motor vehicle crashes became available, most (but not all) states adopted a more rigorous per se BAC driving standard of 100 mg/dL.²⁸ This reduction in the per se legal BAC for an alcohol-impaired driving charge was supported by powerful medical and political groups, including the American Medical Association (AMA). Interestingly, the AMA recommendation also noted that some persons were “under the influence” or impaired in their ability to safely operate a motor vehicle at BACs of 50 to 100 mg/dL. Despite newer drinking and driving laws defined on the basis of an objective chemical test, many state laws still contained older vague language such as “intoxicated,” “visibly intoxicated,” and “obviously intoxicated,” which legislators were then forced to define more clearly and objectively.

Different states use different acronyms to apply to alcohol-impaired driving, including DUI (driving under the influence), DUIL (driving under the influence of liquor), DWI (driving while intoxicated), OUI (operating under the influence), OWI (operating while intoxicated), and OMVI (operating a motor vehicle while intoxicated). Regardless of the acronym used, the definitions are effectively the same. Depending on the state, the same terms may also be used in drug-impaired driving, whereas in other states, a separate charge, such as DUID (driving under the influence of drugs) may be applied to nonalcohol drug-related driving impairment.

In 2000, the National Highway Traffic Safety Administration (NHTSA) reported that of 41,471 motor vehicle fatalities, 38.4% were alcohol-related; this corresponded to an average of one alcohol-related traffic fatality every 31 minutes.⁶⁹ The 2014 NHTSA data suggest a slight decrease in alcohol-related fatalities, with 31% of fatal motor vehicle crashes involving a driver with a BAC of 80 mg/dL or greater, corresponding to one alcohol-impaired-driving fatality occurring every 53 minutes.⁴³ However, it must be noted that NHTSA considers any fatal crash in which a driver has a BAC of 80 mg/dL or greater to be an alcohol-impaired driving crash; and, fatalities in these crashes are considered to be due to alcohol-impaired driving.⁴³ Critics of the NHTSA data claim this method for counting likely overestimates the number of actual fatalities involved in alcohol-related crashes as the mere fact that a driver had consumed alcohol and had a BAC of greater than or equal to 80 mg/dL does not prove crash causality. Nonetheless, there is little argument that alcohol use increases crash risk. Epidemiologic assessments demonstrate that the probability of causing an alcohol-related motor vehicle crash increases slightly at a breath alcohol concentration (BrAC) of 50 mg/dL. The risk of causing a crash is increased by roughly 4-fold at a BrAC of 80 mg/dL, 7-fold at a BrAC of 100 mg/dL, and 25-fold at a BrAC of 150 mg/dL.^5,28 In 1994, a successful campaign, vigorously supported by the AMA and Mothers Against Drunk Driving (MADD), encouraged all states in the US to lower the BAC used to define per se intoxicated driving to 80 mg/dL.²² Although all states technically have the right to set drunk-driving statutes at their discretion, political advocates for more strict alcohol-impaired driving statutes were successful in convincing the US government to require a minimum legal drinking age of 21 years and a mandatory per se BAC statute of 80 mg/dL in order to receive federal highway funding.²² Consequently, as of July 2004, all US states, the District of Columbia, and Puerto Rico conform with federal recommendations defining a BAC of 80 mg/dL as a violation of motor vehicle code. Lower per se BACs are applied to interstate commercial drivers (40 mg/dL) and pilots of aircraft (40 mg/dL, with no alcohol consumed within 8 hours prior to acting as pilot in command), as well as minors (10–20 mg/dL, depending on the state, and 20–50 mg/dL in some European countries). Additionally, some US states, such as Colorado, have lower per se statutes for the slightly lesser charge of driving while ability impaired (DWAI), which may be invoked when a driver’s BAC is between 50 and 80 mg/dL. Some US states also have legal standards defining an “aggravated DUI.” Definitions of “aggravated DUI” are variable by state, but some of the actions that may lead to this charge include, committing a DUI while under court order to use an approved ignition interlock device, committing a DUI while the offender’s driver’s license or driving privileges are suspended, canceled, or revoked because of a prior DUI violation, or committing multiple DUIs in a specified time period (eg, 10 years). States such as Montana have an additional standard that charges an individual with aggravated DUI if the driver’s BAC is found to be 160 mg/dL, or more.

Laws addressing per se driving under the influence (DUI) limits are based on ethanol concentrations measured in whole blood. BrAC limits (which are arithmetically linked to BAC) are also frequently specified in per se statutes. As such, the use of appropriately conducted breath alcohol measurements eliminates the need to collect a blood specimen from a suspect. Per se limits, as described above, apply to the ability to safely accomplish specifically defined tasks such as driving an automobile or flying an aircraft. It is, however, imperative to understand that per se BACs do not define drunkenness or alcohol intoxication in nondriving situations, and their use in circumstances not involving motor vehicle operation is generally inappropriate.

The legal significance of a per se limit is that there is no requirement for behavioral evidence of intoxication, as long as the measured BAC exceeds that established by the legislature. Since their enactment, numerous legal arguments have challenged the constitutionality of per se drunk-driving laws. Issues including a lack of due process through application of the “void for vagueness” doctrine, reliability of breath testing results, and the distinction between blood alcohol and breath alcohol concentrations have all formed bases for legal challenges to per se laws.²² However, despite such challenges, the courts consistently uphold these laws.

ANALYTICAL CONSIDERATIONS

Accurate measurement of ethanol concentration in various biological matrices can be done using a number of analytical techniques. Historically, Widmark employed wet chemical oxidation, using potassium dichromate and excess sulfuric acid, followed by iodometric titration of the remaining amount of oxidizing agent.⁶⁷ Although this method is effective, it is laborious and lacks specificity if other volatiles (eg, methanol, acetone, ether) are present, as these substances are also oxidized, causing a falsely elevated ethanol result. Such wet chemical methods are now obsolete and no longer used in forensic and clinical laboratories.

Enzymatic ethanol assays based on alcohol dehydrogenase (ADH) are commonly used, especially in high-throughput laboratories. The oxidation conditions of enzymatic methods are milder than wet chemical methods. Additionally, acetone, which caused the most troublesome interference with wet chemical methods, is not oxidized by ADH. Interferences by other aliphatic low-molecular-weight alcohols such as methanol and isopropanol occur with ADH isolated from humans and other animals; this interference is reduced by using yeast-derived ADH, which shows greater selectivity for ethanol.^31,64 Specific enzymatic methods used for ethanol analysis in body fluids include enzyme-multiplied immunoassay technique (EMIT), fluorescence polarization immunoassay (FPIA), and radiative energy attenuation (REA), which is related to FPIA (Chap. 7). Comparative study of ethanol determination by REA and gas chromatography shows excellent agreement in precision and accuracy between the 2 techniques.¹² Both high serum lactate and elevated lactate dehydrogenase concentrations interfere with ADH-based methods of ethanol analysis, including providing false-positive results in serum specimens from alcohol-free patients.^3,44

The greater selectivity for ethanol of gas chromatographic (GC) methods make this technique the mainstay for quantitative analysis in most forensic laboratories. Typically, the GC method involves head space sampling (HS-GC), which capitalizes on the volatility of ethanol, eliminates some potential interferences, and shortens run time. Samples to be run by HS-GC are first diluted (typically 1:5 or 1:10) with an aqueous solution of internal standard. This mixture is then sealed in a crimp-top vial with a rubber septum and gently heated to 122°F to 140°F (50°C–60°C) for 30 to 60 minutes in order to achieve equilibrium of volatiles between the gas and liquid phases. The vapor is then sampled with a syringe by puncturing the rubber septum and the withdrawn vapor injected into the instrument. Heating the sample for too long at a high temperature can result in oxidation from oxyhemoglobin.⁵⁶ This problem is solved by the addition of sodium azide or sodium dithionite to block the oxidation or simply reducing the equilibrium temperature to 104°F to 122°F (40°C–50°C).⁵⁶ Once chromatographic separation is achieved, various detection methods, including flame ionization (FID), electron capture (EC), and electrochemical sensing are used (Chap. 7). Dual detection (eg, FID and EC) can also be useful in circumstances where a larger number of volatiles may need to be screened.⁵⁷

As ethanol distributes based on total body water, the water content of the matrix will affect the amount of ethanol present in a given volume or mass of biological sample. A common circumstance in which this is observed is in the difference between a whole blood ethanol determination and a plasma or serum ethanol determination. The water content of serum and plasma is 10% to 12% higher than whole blood, meaning that serum or plasma ethanol concentrations will be correspondingly higher than whole blood concentrations. Clinical laboratories most often use serum and plasma for analysis, whereas per se DUI statutes are written in terms of whole blood. Therefore, if a comparison is to be made between a serum or plasma analytical result and a legal standard, the result must be converted to an approximated whole blood ethanol concentration. Experimentally determined serum-to-whole blood ethanol ratios range from 1.12 to 1.17, whereas plasma-to-whole blood ethanol ratios range from 1.1 to 1.35.⁴⁵ Typically, a ratio of 1.16 is used for this conversion, and no significant difference appears to exist between the serum-to-whole blood and plasma-to-whole blood ratios.⁶⁹ Ratios between whole blood and other biological matrices such as urine, saliva, cerebrospinal fluid, vitreous humor, brain, liver, kidney, and bone marrow are published, and a table listing these ratios and the sources from which they are derived is available.¹¹

Breath testing is commonly performed to assess BAC because of its relative simplicity and less invasive collection compared to urine or blood. The analytical basis for sampling exhaled breath is that, at equilibrium, alcohol in expired air is present at a predictable ratio with blood. The gas exchange process in the lungs is complex, with significant theoretical variability.⁴⁰ In the United States, a blood-to-breath alcohol ratio of 2100:1 is commonly used in the calibration of breath alcohol testing devices, although some experimental evidence suggests that the ratio is actually closer to 2300:1.¹⁷ As such, a systematic underestimation of BAC is expected when breath alcohol results are converted to whole blood alcohol results. Several studies document this underestimation with data from suspected impaired drivers and evidential breath alcohol testing instruments.^20,25 In 1983, Britain adopted a legal limit in breath of 35 mcg/100 mL, which corresponds to a blood alcohol of 80 mg/dL, using a 2300:1 blood-to-breath ratio.²⁸ Legal statutes that include in their offense definition breath alcohol results expressed in units of grams/210 L of exhaled air eliminate the need to convert breath alcohol to blood alcohol, largely mitigating arguments based on the breath-to-blood ratio.

BREATH-TESTING DEVICES

Three general analytical detection principles have been extensively validated and are currently used in breath alcohol–testing instruments. These are infrared spectroscopy, electrochemical oxidation/fuel cell, and chemical oxidation/photometry.²⁶ Additionally, combination of multiple technologies is sometimes used, as in infrared/fuel cell dual detectors. Breath-testing instruments are generally divided into 4 broad categories: passive alcohol sensors (PASs), screening devices (preliminary breath testers, {PBTs}), breath alcohol ignition interlock devices (BAIIDs), and evidential breath testers (EBTs).²⁶

A PAS device may be concealed in a device such as a modified police flashlight and is used to detect alcohol on or in the immediate vicinity of a subject through passive means (ie, with no requirement for subject cooperation). Breath alcohol ignition interlock devices are used to prevent drinking and driving by requiring the driver to blow into a sensor in order to start the ignition of the vehicle. The devices are typically installed on a court order designed to modify drinking and driving behavior in habitual DUI offenders, with the cost of installation typically borne by the offender. A number of mechanisms are employed to prevent substitution of breath samples, including preset patterns of exhalation or humming while blowing into the sampling tube. Additionally, random rolling retests are required, failure of which result in the vehicle’s lights flashing or horn blowing. Neither PAS nor BAIID results are useful for quantitative measurement of breath or blood alcohol for prosecution of a per se DUI case.

The PBT and EBT devices are the most common breath alcohol–testing devices encountered in cases of DUI arrest. In this setting, PBT results are used in conjunction with observation and field sobriety tests to establish probable cause for DUI arrest. Although a measured BrAC is often provided as a digital readout on a PBT, in most jurisdictions these results are not admissible as evidence for proceedings other than probable cause hearings. Use of PBT devices is also common in nonlegal settings such as hospital emergency departments, alcohol detoxification units, homeless shelters, and workplaces. As with DUI prosecution, these results are generally used for screening for alcohol presence and not for establishing impairment.

In contrast to PBT results, measurements from an EBT device are admissible in court and administrative proceedings and can be used as the basis for establishment of per se DUI cases without the necessity of blood collection and analysis. Although mobile EBT devices are available, most often the EBT is maintained in a fixed location such as a police station. For the results to be considered valid and admissible, the operator of the EBT must be trained and certified, and the operation must be performed using an accepted testing protocol.

Required procedures for the use of EBT devices exist for the subject as well as the instrument. The subject must have a period of alcohol deprivation of at least 15 to 20 minutes during which trained personnel observe him or her to ensure not only that no additional ethanol is consumed but also that no regurgitation, emesis, or eructation occurs, which could result in residual ethanol in the mouth prior to breath alcohol analysis. With respect to the instrument, both blank and control analyses must be performed prior to analysis of the subject sample. A blank analysis, typically done with room air, purges the instrument of contamination from previous samples and demonstrates a lack of environmental contamination. A control analysis is performed on a gaseous ethanol sample of known concentration, usually an ethanol gas canister or wet bath simulator, and demonstrates proper instrument calibration and maintenance. Certification of instrument maintenance of calibration must be documented, demonstrating compliance with applicable rules, regulations, laws, and standards for routine maintenance, troubleshooting, and corrective actions. These documents are then retained for a relevant time period after inspections are complete. Additional documentation for individual cases should include written verification that all steps in the accepted protocol were followed. This documentation also includes automated printouts from the analyzer used, and well as copies of manual checklists.²⁶

Beyond the required procedures, additional recommendations include the use of grams/210 L as the reporting units, rather than reporting breath alcohol concentration as a converted blood alcohol concentration of %weight/volume, grams/100 mL, or grams/deciliter. Serial collection and analysis of at least 2 separate sequential breath samples taken 2 to 10 minutes apart should be done in order to demonstrate the absence of residual mouth alcohol, instrument artifacts, frequency interference, and spurious results. Agreement of the serial results must be within prescribed limits (usually, 0.02 g/210 L) to be considered acceptable.²⁶

STANDARDIZED FIELD SOBRIETY TESTS

A driver whose BAC exceeds the established legal standard is considered per se intoxicated and often convicted without behavioral evidence of intoxication. However, some objective findings of impairment assists in establishing probable cause to demand chemical testing, initiate a DUI arrest, or prosecute a charge of impairment, without invoking per se limits. In these settings, results of a specific group of behavioral tests are of value in discriminating and prosecuting or refuting an impaired driving charge, although submitting to these behavioral tests is typically voluntary. Additionally, an objective measurement of alcohol effect is helpful in assessing alcohol-related impairment in nondriving situations, where it is inappropriate to directly apply per se standards.

Under a contract with the NHTSA, a group of 15 candidate sobriety tests were evaluated in a laboratory study.⁹ From this original group, the investigators developed a series of 3 specific tests that have subsequently been standardized as a test battery, known as standardized “field sobriety tests” (FSTs), for assessing driver impairment in the United States. The 3 tests comprising the standardized FSTs are the one-leg stand (OLS), walk-and-turn (WAT), and horizontal gaze nystagmus (HGN). Descriptions of these tests are provided in Table SC11–1.

A 1981 study of 297 drinking volunteers with BACs ranging from 0 to 180 mg/dL who were evaluated by trained police officers showed adequate interrater reliability (correlations, 0.6–0.8) and test-retest correlations (0.40–0.75).⁵⁹ Using all 3 field sobriety tests, the officers were able to distinguish whether BAC was above or below 100 mg/dL in 81% of participants. Test results were observed to generally correlate with BAC. However, the specificity and sensitivity of each individual test was not evaluated. Field sobriety tests are typically associated with physical, rather than mental, ability other than understanding the instructions for each test. Some aspects of variability in cognitive performance correlate moderately and positively with facets of FSTs, especially those tests evaluating reaction time.¹⁶

In the few studies evaluating the performance of individual FSTs, general correlation between test performance and BAC is observed. Although the correlation magnitudes for the WAT and OLS tests are low, the HGN test showed excellent sensitivity and specificity across a range of BACs.⁴⁶ In the control arm (ie, no alcohol ingested and no detectable ethanol in the blood), approximately 50% of the participants were judged to be impaired using the WAT test, whereas in the OLS test, 30% of the participants were rated as impaired. When the BAC was greater than 150 mg/dL, performance on the WAT and OLS tests improved to 78% and 88%, respectively. The rate of false positives (ie, those with a zero BAC who failed the HGN test) was only 3%. As seen with the other FSTs, the sensitivity of the HGN test increased with increasing BAC, being 81% for those with a BAC between 100 and 149 mg/dL and 100% for those with a BAC greater than 150 mg/dL. The presence of horizontal gaze nystagmus, even in alcohol-tolerant drinkers, gives this test an advantage over the WAT and OLS tests in detecting the presence of alcohol.¹⁰ It is noteworthy that balance and coordination can be affected by factors unrelated to alcohol consumption, such as physical disability, age, and nervousness about potential arrest; however, chronic drinkers with substantial alcohol tolerance are reportedly able to complete balance tests with few errors, even with moderate to high BACs.⁹

The 3 tests utilized for field sobriety assessment have some advantages, in that they are standardized and are easily understood by test takers. In addition, personnel administering the tests can be trained in a matter of hours.³⁹ However, because the test results are only approximately correlated with BAC, their use is limited by the threshold BAC they are able to detect. Predictably, various scientific and legal challenges to the use of the FSTs have also been made. For example, the tests were designed and validated at a time when the per se DUI limit in most of the United States was 100 mg/dL; however, that threshold is now 80 mg/dL in every state (and 20–50 mg/dL in many European countries).³⁹ Because of the variability of psychomotor effects of alcohol as a result of individual tolerance, FST assessments are generally not allowed as a means of estimating BAC in court proceedings. The effects of fear, fatigue, rehearsal of test performance, the arresting officer’s knowledge of estimated BAC prior to administering the standardized FSTs, and various medical conditions have not been fully addressed. It has further been argued that the studies most strongly supporting use of standardized FSTs are those conducted by NHTSA-affiliated investigators and reported in non–peer-reviewed government publications. So contentious has the debate been that some have charged that “the United States Department of Transportation indulged in deliberate fraud in order to mislead the law enforcement and legal communities into believing the test (HGN) was scientifically meritorious and overvaluing its worth in the context of criminal evidence.” A comprehensive review detailing this, and other, legal and scientific issues surrounding standardized FSTs was published in 2008.⁵⁰ The author concluded that SFTs are not useful in predicting BAC, and that further research is needed to determine whether improvements can be made to the current battery of tests included in FST assessments. Despite arguments of FSTs having limitations, these assessments are still commonly used in arrest and prosecution of alcohol-impaired driving cases.

ESTIMATING THE AMOUNT OF ALCOHOL INGESTED

The kinetic profile of ethanol in the blood has been extensively studied. In general, it is known that when blood ethanol concentration is greater than 20 mg/dL, it follows a zero-order elimination profile and then converts to first-order elimination when the concentration drops below this threshold.³⁰ In the zero-order portion of the curve, typical elimination rates range from 10 to 20 mg/dL/h in social drinkers. In one study of adult patients in an urban hospital emergency department, nonchronic users of alcohol (n = 9) demonstrated a mean elimination rate of 18.7 mg/dL/h (range, 16.1–21.4), whereas chronic users (n = 15; defined as more than 2 drinks per day, history of delirium tremens, or prior detoxification) had a significantly greater mean elimination rate of 20.3 mg/dL/h (range, 16.1–24.6).⁶ Other studies report elimination rates of 30 to 50 mg/dL/h in chronic heavy drinkers due, at least in part, to metabolic enzyme induction.^6,7 Pharmacokinetic calculations in healthy persons often employ a range of elimination rates of 10 to 20 mg/dL/h in order to best bracket individual differences.⁷

The estimation of the number of drinks ingested in order to achieve a given BAC has been extensively studied and is predicated on knowing several case-specific pieces of information. According to the US Department of Agriculture, a standard drink consists of 12 ounces of beer (5% v/v), 5 ounces of wine (12% v/v), or 1.5 ounces of 80-proof spirits (40% v/v); each of these drinks contains approximately 14 g of ethanol.^7,63 The time period over which the ingestion occurred, the time of the last drink, and time of blood alcohol specimen collection must be known. It is helpful to know whether the subject was fasting or had a full stomach at the time of drinking because the presence of food in the stomach slows ethanol absorption.^34,65 Finally, the gender, age, height, and weight of the individual should be known and considered.⁷

Using this information, an estimation of the number of drinks is possible. The original work in which the prediction of ethanol concentrations from known doses was performed by Widmark in 1932 using 20 healthy volunteer moderate drinkers (10 of each gender).⁶⁷ This work was translated into English from the original German by Baselt in 1981.⁶⁸ The basis of Widmark’s work was that the BAC was directly proportional to the administered amount of alcohol (“dose”); the body weight of the subject; and a unitless scaling factor that Widmark referred to as “ρ” (rho), which corrected the subject’s body weight for water content. Of note, in present-day pharmacokinetics, Widmark’s ρ effectively represents the volume of distribution of ethanol. Because of the differences in body water content between sexes, ρ was determined to be approximately 0.6 in women and 0.7 in men. The original Widmark equation is therefore written as

where A represents the total amount of ethanol equilibrated in all fluids and tissues at the time of sampling (ie, total dose ingested), C is blood ethanol concentration in grams/kilogram, ρ is the previously described Widmark ρ factor, and p is subject body weight in kilograms. Although the calculation has served well as an estimating tool for more than 80 years, it does have some limitations, principally in the ρ factor. As a result, refinement of Widmark’s original ρ was undertaken by others.^7,66 One method for the estimation of total body water (TBW), which takes into account not only gender but also individual age, weight, and height (abbreviated ΣV_d) for men aged 17 to 86 years yields the following equation:

For women aged 17 to 84 years, the TBW equation is:

In all persons, the height is measured in inches and the weight is measured in pounds. Note that the equation for men includes a mathematical term derived from age, and the equation for women does not.⁶⁶ The significance of this calculation is its ability to be tailored to a specific person, rather than the broad extrapolation of the relatively few subjects in the original study. Inclusion of such individual information improves the accuracy of such calculations.⁵⁴

The need for a more standardized approach to calculations involving alcohol in forensic medicine has been known for decades. One author summarized 20 of the most commonly used alcohol-related calculations.⁷ As an example, the following equation is derived to calculate the estimated dose of ethanol necessary to achieve a specific blood alcohol concentration:

where g EtOH is the ingested dose of ethanol in grams; BAC_target is the observed blood alcohol concentration in mg/dL; β_1–n is a range of alcohol elimination rates (typically, 10–20 mg/dL/h); t_s is the time from the start of drinking to the last drink; t_p is the range of times from the last drink to the peak BAC (typically, 30–90 minutes); ΣV_d is the Watson TBW, which is the volume of distribution based on age, weight, height and gender; and Bl_H2O is the approximate percentage of water in whole blood (80.65%).⁷ Noting the inherent variability in terms such as β_1–n and absent specific kinetic data from the individual in question, it can be seen that results of such equations are most reasonably presented as a range, rather than a single discrete calculated value.

Some legal arguments have quite incorrectly suggested that ethanol is odorless. Ethanol has a characteristic pleasant smell, with an odor threshold of approximately 50 ppm.¹⁴ The presence or absence of breath alcohol odor is often used by police officers in the decision to proceed further with sobriety testing. In one study, 20 experienced police officers assessed alcohol odor on 14 subjects with BACs ranging from 0 to 130 mg/dL after drinking beer, wine, bourbon, or vodka.⁴² Assessments were initiated 30 minutes after cessation of drinking. The strength of breath alcohol odor was determined to be an unreliable indicator of BAC. Correct detection of the presence of alcohol was 85% for BACs at or above 80 mg/dL but declined with decreasing BAC or the presence of food. The authors also noted that there were only small differences in odor intensity as a function of beverage type, meaning fusel oils and other constituents of alcoholic drinks are not the primary determinant of detectable odor after absorption of the beverage.

DRAM SHOP

Liquor liability or “dram shop” laws hold the server of alcoholic beverages liable for damages or injuries caused by an individual who was provided alcohol when such service should have been refused. For example, if a an individual who was already intoxicated, or any minor, is served alcohol and crashes his or her vehicle, the person and/or establishment who served the alcohol may be liable for damages or injuries sustained in the crash.

Interestingly, US dram shop laws were first enacted in the mid-1800s, although they were rarely used. The initial intent of these laws was to provide financial support to the families of persons who had become “habitual drunkards” through their patronage of a drinking establishment.⁵² Repeal of Prohibition in the United States in 1933 shifted laws governing alcohol sales from federal to state control, resulting in state-to-state variability in alcohol availability, criminal and administrative laws, and liquor liability. There was also a move from public focus on habitual drunkenness to the damage caused by impaired drivers, as well as a paradigm shift in the standard for irresponsible service away from serving a “drunkard” to serving an individual who was “visibly intoxicated.”⁵² The purpose of liquor liability laws has also changed over time. In the 19th century, laws were enacted to punish tavern owners who contributed to the downfall of patrons. By contrast, current application of the laws is typically as a means to compensate innocent victims injured by an intoxicated patron. This action on behalf of the innocent victim is why liquor liability law is sometimes referred to as “third party” liability.

Beginning in the late 1970s and early 1980s, political activists began to take note of dram shop laws as an effective means for prevention of impaired driving (or DWI). Outreach began to bring attention to irresponsible commercial and social service of alcohol. Various drunk-driving prevention strategies emerged, including “server intervention” (ie, the practice of bar or restaurant workers intervening to prevent an intoxicated patron from driving). Interestingly, although the goal of keeping intoxicated patrons from “getting behind the wheel” is certainly praiseworthy, these programs put notably less emphasis on preventing intoxication in the first place. The 1983 Presidential Commission on Drunk Driving recommended dram shop liability and server intervention as drunk-driving prevention strategies, and federal grant funds for state impaired-driving initiatives also listed such programs as qualifying criteria.¹

Alcoholic Beverage Control agencies in the various states are responsible for reviewing and approving liquor licenses of commercial establishments, collecting taxes, and enforcing criminal and administrative laws prohibiting service to minors and intoxicated persons. Largely as a tool to improve public perception of the industry, the alcoholic beverage producers have largely been responsible for the development of training programs for servers in the retail community. Numerous programs offering Responsible Beverage Service (RBS) training for servers are administered at the state or regional level, or sometimes even sponsored by the server’s employer. Calls for greater organization and standardization of RBS programs in the United States have been answered, at least in part, by the Responsible Hospitality Council, whose work has focused on developing minimum standards for legitimate RBS training.⁵² Although such standards are often welcomed by regulators, some restaurant associations and owners have expressed concern over cost and compliance standards associated with RBS mandates. Nonetheless, the US legal climate seems to dictate that alcoholic beverages be served in a “responsible” manner and that liability for “overservice” falls on the serving establishment. It is also noteworthy that cases placing liability on individuals serving alcohol socially in their home have been successfully argued.

Direct comparison of the efficacy of dram shop liability laws between states is difficult as liability laws and insurance vary widely. Historically, different states and regions emphasized different areas of alcohol service liability, and it was noted that research played only a minor role in determining server training, with the majority of policy formation arising from political influence.⁵² However, even data collected in the 1980s suggested that dram shop laws did result in a statistically significant decline in mortality rates not only for motor vehicle collisions, but also for other alcohol-related causes.⁵⁵ A more recent study concluded that dram shop liability and RBS laws were associated with significant reduction in per capita beer consumption and fatal crash ratios in drivers under age 21.⁵³ This begged the question of whether enhanced enforcement of dram shop laws would result in a decrease in excessive alcohol consumption and related consequences. An evaluation of 11 studies concluded that while dram shop laws were effective in reducing alcohol-related harms, the data supporting the efficacy of enhanced enforcement on reduction of alcohol-related harms were mixed and ultimately judged insufficient.⁴⁷

Specific study of alcohol service liability on younger and underage drinkers has also been investigated. Reduction of the legal age for alcohol purchasing from 20 to 18 in New Zealand in December of 1999 resulted in a statistically significant increase in traffic injuries attributable to male drivers aged 15 to 19 years.³³ Review of annual fatal motor vehicle collision data from the United States showed that the most effective interventions for the prevention of underage alcohol-related fatal crashes were those associated with reducing teen drinking, rather than those restricting teen driving.⁴⁹ An evaluation of 20 minimum legal drinking age 21 (MLDA-21) laws in order to assess which had an effect on fatal underage drinking and driving traffic crashes.¹⁸ Nine of the laws were associated with a decrease in fatal crash ratios: fake identification support for retailers (−11.9%), use alcohol and lose your driver’s license (−7.9%), possession of alcohol (−7.1%), purchase of alcohol (−4.2%), age of the bartender greater than or equal to 21 (−4.1%), responsible beverage service program (−3.8%), zero tolerance 0.02 BAC limits for underage drivers (−2.9%), state dram shop liability (−2.5%), and social host civil liability (−1.7%). Interestingly, 2 laws were associated with a significant increase in fatal underage crash ratios (registration of beer kegs [+9.6%], and prohibition of furnishing alcohol to minors [+7.2%]). The authors concluded that 9 effective MLDA-21 laws were responsible for saving an estimated 1,135 lives annually, though only 5 states have adopted all 9 of these laws.

INTOXICATION AND ESTIMATION OF BLOOD ALCOHOL CONCENTRATION

Neither the characteristic odor of ethanol nor the measurement of various markers of alcohol exposure (eg, ethyl glucuronide, ethyl sulfate, carbohydrate deficient transferrin, and so on) are useful in estimating BAC or degree of intoxication. The ability of an individual to accurately estimate his or her own BAC is poor.^35,51 Drivers who underestimate their own BAC are more impulsive and riskier drivers than those who overestimate their BAC.³⁶ A lack of accuracy in estimating BAC is also observed in trained medical providers and police officers.^4,13 A large confounding factor in these observations is tolerance. Greater tolerance, as occurs in heavy drinkers and alcoholics, imparts greater difficulty in detecting the clinical effects associated with alcohol intoxication.

Tolerance involves a central adaptation to the intoxicating effects of alcohol. As such, persons with significant tolerance to the effects of ethanol may not manifest signs of intoxication despite having a high BAC. Multiple case reports and case series in the medical literature describe alcoholic individuals with heavy alcohol consumption who have very high BACs but muted or absent clinical effects.^{15,24,29,38,48,58} As a result of tolerance in the chronic heavy drinking population, it is often difficult to detect clinical intoxication even though an individual has a high BAC. Critically, this circumstance does not suggest that, even in the clinically sober chronic drinker, driving abilities are unaffected or that the individual is necessarily capable of safely operating a motor vehicle.⁴ Furthermore, tolerance has no bearing on the potential prosecution of a DUI case on a per se BAC or BrAC basis.

The circumstance is slightly more straightforward in the social drinker. The intensity of the effects of alcohol on the central nervous system (CNS) is generally proportional to the concentration of alcohol in the blood. Dubowski tabulated the stages and effects of acute alcohol intoxication, and these tables are often used in educational programs and some legal proceedings.

However, though useful as a pedagogic tool for explaining the continuum of alcohol intoxication, the table must be used with care as the effects are defined only over a population, thereby making assignment of a specific effect or degree of effect in an individual impossible. The inherent individual variability in the table is apparent in the fact that the BAC ranges for the various stages of alcoholic influence overlap. Furthermore, the population and methods used to compile the table are not described, and the table itself has not been subjected to peer review.

The CNS effects of alcohol intoxication are typically more pronounced on the ascending portion of the blood alcohol kinetic curve than on the descending side due to acute tolerance.²¹ In other words, the clinical effect of intoxication is greater during the absorptive arm of the kinetic curve than on the elimination arm, even though the same blood alcohol concentration is measured in both kinetic phases. This principle is known as acute tolerance, or the Mellanby effect.

In a dram shop case, the ultimate question often becomes, what is the typical BAC at which it is more likely than not that the average nontolerant individual will exhibit signs of intoxication (eg, the odor of alcohol on the breath, clumsiness, difficulty walking or maintaining balance, slurred speech, inappropriate behavior) that are apparent to a bystander or untrained casual observer? A 1986 report by the Council on Scientific Affairs of the AMA reviewed 7 studies spanning 50 years that included more than 6,500 participants for identification of BACs at which individuals appeared “drunk” (ie, clinically intoxicated).² BACs were stratified by increments of 50 mg/dL, beginning at 0.0 to 50 mg/dL and extending to 401 mg/dL. In the lowest BAC group (0.0–50 mg/dL), observer perceptions of subject drunkenness ranged from 0% to 10%, with an average of 4%. The percentage of individuals determined by observers to be drunk increased steadily from a mean of 32% (range, 14%–68%) at a BAC of 51 to 100 mg/dL to a mean of 62% (range, 47%–93%) at a BAC of 101 to 150 mg/dL. In this latter range, 4 of the 7 studies indicated a perception of drunkenness by less than or equal to 50% of observers. However, in the next BAC increment of 151 to 200 mg/dL, observers judged a mean of 89% (range, 83%–97%) of the drinkers to be drunk. In each of the subsequent higher increments, the mean percentage of persons classified as drunk ranged from 95% to 100% with no individual study value less than 90%. Therefore, it can be reasonably concluded that in a nontolerant individual, a BAC of 151 to 200 mg/dL will more likely than not result in observable signs of drunkenness to a casual observer.

Study of the effects of alcohol on driving performance has been the subject of decades of research. Driving simulators are often used in laboratory studies of impaired driving and have the benefit of easy replication and control of experimental conditions. A number of simulation devices have also been developed including “fatal vision goggles,” which provide visual distortion to the wearer in order to simulate the effects of alcohol on visual perception.⁴¹ Additionally, simulators and simulation devices allow study of the effects of impairing substances in populations in which actual drinking is legally or ethically unacceptable (eg, underage prelicensed drivers). However, it is known that driving performance evaluated in a simulator is less sensitive to the effects of alcohol than actual on-the-road driving.³² Additionally, despite frequent use, many driving simulator tests lack validation and head-to-head comparison with naturalistic driving.³²

The validity of a police officer detecting intoxication in drivers involved in motor vehicle crashes also increases with increasing BAC. One report examined a total of 1336 subjects over age 15 who were admitted or died at a level 1 trauma center in Seattle during a 5-year period from 1986 to 1993 and in whom both a recorded BAC and a police assessment of sobriety were conducted.²³ The blood alcohol measurement was conducted in the hospital, and it is not expressly stated if the analytical matrix used was whole blood, serum, or plasma. Four categories of sobriety assessment were used by police: (a) had not been drinking, (b) had been drinking–not impaired, (c) had been drinking–sobriety unknown, and (d) had been drinking–impaired. Officers used a battery of specific criteria to judge whether a driver was intoxicated, including odor on the breath, slurred speech, chemosis, poor coordination of motor function, and the ability to simultaneously perform multiple tasks. The greatest number of drivers were in the 2 extreme categories: had not been drinking (n = 746) and had been drinking–impaired (n = 568). A direct correlation between measured BAC and police officer assessment of sobriety was observed. The mean BAC associated with the “had been drinking–impaired” group was 190 mg/dL with a 95% confidence interval (CI) of 180 to 200 mg/dL. Those in the “had been drinking–sobriety unknown” group had a mean BAC of 130 mg/dL (95% CI, 110–150). Among all drivers, police field assessment of sobriety had a positive predictive value of 85% with sensitivity and specificity of 91% and 90%, respectively, demonstrating recognition of drunk driving by police with a high degree of accuracy, especially in the group with the highest BAC.

Another study designed to determine the ability of police officers, who have more training than the average lay person, to make assessments about alcohol intoxication in drinking target subjects without the aid of special testing normally available to them (eg, standardized field sobriety tests or the ability to smell the odor of alcohol on the subject’s breath).⁸ This study involved 39 police officers who viewed a series of videotaped interviews with 6 volunteer moderate drinkers having targeted BACs in 3 concentration ranges: low (80–90 mg/dL), medium (110–130 mg/dL), and high (150–160 mg/dL). Based on their observations of the taped interviews, the officers answered 3 questions:

1. Has the person been drinking?

2. Was it OK to serve that person one additional drink?

3. Was the person able to drive a car?

Each of the 3 questions could be answered “yes,” “no,” or “not sure.” A fourth question assessed the officers’ confidence in their answers as “not sure,” “little uncertain,” or “positive.” Question 2 was to be answered from the perspective of a social host or bartender, not a police officer. None of the police officers had any formal training in the management or service of alcohol to intoxicated people, such as that offered by commercial seller/server training programs like Techniques of Alcohol Management (TAM) or Training for Intervention Procedures (TIPS). With respect to their answers, the police officers were fairly certain that the subjects had been drinking only in the target groups with the highest BAC (150–160 mg/dL). Officers also answered in the affirmative to question 2 (ie, that it was OK to serve the target subject another drink) the majority of the time in the low- and medium-BAC target groups but not in the high BAC target group. The percentage of officers answering affirmatively to this question was 55%, 75%, and 41% for the low, medium, and high target groups, respectively.

Multiple studies have examined the likelihood of on-premise (bars and restaurants) and off-premise (liquor stores, grocery stores, and convenience stores), as well as outdoor events like festivals, to sell alcohol to obviously intoxicated persons (typically, paid professional actors pretending to be drunk).^19,37,60-62 The results are fairly uniform in that the majority of the time the alcohol was sold or served to the individual. Additionally, it was noted that male servers/clerks who appeared younger than age 31 were more likely to make such sales and the sales were more likely to occur in off-premise establishments.¹⁹

In a typical study, the authors employed 19 actors who were specifically hired based on their ability to feign intoxication. All actors were male and ranged in age from 31 to 59 years, with a mean of 42 years.⁶¹ The actors used a standardized script to attempt to purchase either a single vodka drink after asking what beers were available on tap (on-premise) or a 6-pack of beer (off-premise). Prior to entering the establishment, the actor received specific instructions to demonstrate multiple signs of intoxication, including disheveled hair and clothing, smelling of alcohol, lack of coordination, stumbling, fumbling with money, slurring words, repeating questions, appearing forgetful, and laughing inappropriately. If the actor was asked if he was driving, he responded “no” and if he was asked if he had been drinking, he responded, “I’ve had a few beers.” Attempts at alcohol purchase were made at 223 on-premise and 132 off-premise establishments.

Of the 355 attempts, actors were able to successfully make the alcohol purchase in 280 instances (79%). On-premise establishments served the actors 76% of the time, whereas off-premise establishments allowed the sale 83% of the time. Actors and a nondrinking observer also watched for clues that the server/clerk indicated an awareness or suspicion that the buyer was obviously intoxicated. Verbal indications, including asking the buyer to leave, suggesting a nonalcoholic drink, offering to call a cab, and so on, as well as nonverbal indications, such as staring or rolling of eyes, were recorded. A similar series of observations were made of security staff, other staff/bartenders/cashiers, and other customers. In 51% of the attempted purchases, there was an indication from the server of the recognition of the intoxication of the buyer. Even within the group that recognized signs of apparent intoxication, the alcohol was sold 61% of the time. When the server made no indication of the apparent intoxication, alcohol sales were completed 97% of the time. In 45% of purchase attempts, another staff member or customer made an indication that they believed the buyer was intoxicated. Still, alcohol purchases were completed in 66% of these cases. When no other staff member or customer indicating an awareness of the buyer’s behavior, the actor was served 89% of the time. Details of responsible beverage service (RBS) training for individual servers were not provided, though the authors did comment that such training is an important tool in preventing illegal alcohol sales, especially in younger servers. The authors concluded that sales of alcohol to obviously intoxicated individuals in some US communities is very high and recommended additional study to identify effective training and tools to mitigate such illegal sales.

SUMMARY

• Impaired driving charges may be prosecuted based on BAC and application of a per se statute, as well as on observations of impairment such as standardized field sobriety tests (FSTs).

• Analytical techniques capable of accurately measuring alcohol concentration in a variety of biological matrices have been available for decades.

• The ability to accurately predict a BAC after drinking is poor, not only for the drinking individual but also for trained personnel such as police officers and medical professionals.

• Casual observers begin to reliably detect signs of alcohol intoxication in nontolerant persons at a BAC of 150 mg/dL.

• Clinically apparent signs of alcohol intoxication are difficult to detect in individuals with significant tolerance, increasing the probability of driving when BAC exceeds a per se value, or “overservice” in a bar or restaurant. Tolerance also seriously limits the applicability of tables correlating BAC with specific clinical effects in an individual case.

• The use of standardized FSTs is the norm for prosecution of impaired driving cases, though this battery of tests has also received significant legal and scientific criticism.

• Of the 3 routinely used FSTs, horizontal gaze nystagmus (HGN) has the best correlation with BAC.

REFERENCES

CASE

A newly hired train conductor sustains irreversible brain injury due to a large subdural hematoma from a train crash. Surgical intervention is deemed futile. After 4 days in the hospital, he is declared dead by neurologic criteria (ie, “brain dead”) and removed from life support. An autopsy confirms the brain injury. Postmortem toxicologic samples are obtained from the femoral vein and reveal oxycodone and oxycodone metabolites. As a toxicologist, you are asked to interpret the findings. How do you proceed?

INTRODUCTION

Postmortem toxicology is the study of the identification, distribution, and quantification of xenobiotics after death. This information is used to account for physiologic effects of a xenobiotic at the time of death through its quantification and possible redistribution in the body at the time of autopsy. Several variables cause changes in xenobiotic concentrations during the interval between (1) the time of death and subsequent autopsy and (2) the storage interval between the time of sampling and the time of testing. Toxicologists and forensic experts are frequently asked to interpret postmortem xenobiotic concentrations and decide whether the reported values are meaningful and whether these xenobiotics were incidental or contributory to the cause of death.

The development of the field of forensic toxicology and the improvement of laboratory technology now permit more refined identification and quantification of xenobiotics. The interpretation of postmortem xenobiotic concentrations and their significance, however, continues to evolve.

This chapter reviews factors that affect xenobiotic concentrations identified at autopsy and discusses an approach for interpreting postmortem toxicologic reports as they relate to the cause and manner of death.^{39,40,43,45,52-54,61,86,105}

HISTORY AND ROLE OF MEDICAL EXAMINERS

The relationship between antemortem xenobiotic exposures and death has been a subject of investigation for centuries. In 12th-century England, an appointee of the royal court, eventually named the coroner, was designated to record and identify causes of death.⁷⁸ In suspicious circumstances, coroners investigated poisonings, but scientific methods were primitive, and conclusions regarding such deaths were conjecture at best.

By the mid-19th century, however, techniques for detecting certain xenobiotics in postmortem tissue were developed and focused generally on identifying heavy metals as a cause of death in homicides.^{46,78,83,102,108} At that time, coroners were still elected or appointed individuals with little or no medical training. However, with better laboratory techniques and autopsies being performed by trained pathologists, the specialty of forensic medicine continued to develop. In Massachusetts during the late 19th century, these forensically trained pathologists, referred to as medical examiners, ultimately replaced the coroner system and were eventually empowered by the state to investigate and determine the cause and manner of death in certain types of unusual or suspicious cases and the medicolegal autopsies became their primary tools.⁷⁸ Cause and manner of death are defined below.

Currently, a death may prompt a medicolegal investigation if it serves the interests of the state through either the criminal justice or public health systems.⁸⁹ Information from the circumstances of the death are integrated with laboratory data and a medicolegal autopsy when human remains are available. In the United States, the requirements for performing a medicolegal autopsy vary substantially by jurisdiction.^4,50

Some US jurisdictions use elected or appointed personnel with variable or no medical training to determine the cause and manner of death. These are usually termed coroner systems. The coroner is usually not a physician but may have other qualifications. Other jurisdictions require that physicians or other medically trained personnel determine the cause and manner of death and perform a medicolegal autopsy if they deem it necessary. These providers are medical examiners, and their training requirements also vary by jurisdiction. The strictest and most robust regulations specify that autopsies are performed exclusively by forensic pathologists. These are physicians with board certification in forensic pathology. The specific training and performance standards of forensic pathologists are well described elsewhere.^1,89

As of 2016 in the United States, death investigations are the responsibility of a coroner in 14 states, a medical examiner in 20 states, or some combination of coroner and medical examiner in the remainder.⁴ For the purposes of this chapter, the person performing the medicolegal autopsy will be referred to broadly as the medical examiner.

The purpose of the death investigation is to determine the cause and manner of death. The cause of death is the etiologically specific pathophysiologic aberration that is ultimately responsible for death to occur (Table 140–1). For example, the presence of cyanide in the toxicologic evaluation is sufficient to establish cardiorespiratory arrest from cyanide poisoning. The manner of death is an explanation of how the death occurred, or the circumstances surrounding the terminal events. It broadly distinguishes natural from nonnatural (or violent) deaths. Nonnatural deaths, depending on the jurisdiction, can be divided into several categories (Table 140–2). With the identification of cyanide, the manner of death cannot be considered natural because a poisoning is a “chemically traumatic” (violent) event. The medical examiner must make the best determination of the manner of death based on the available evidence.^108,115 An unintentional exposure can be classified as an accident (a legal term for some unintentional nonnatural deaths), and intentional self-injury can be classified as a suicide. If the circumstances indicate an exposure due to the acts of another person, the manner of death can be classified as a homicide.

Determination of the manner of death has important consequences. Homicide necessitates involvement of law enforcement officials for further investigation. Cases deemed suicide not only impact survivors psychologically but also can nullify some life insurance payments. Conversely, a case deemed an “accident” can invoke a double-indemnity insurance clause. Assignment of financial responsibility for workplace deaths can be similarly affected when xenobiotics are identified in the postmortem specimens of involved workers.

Recognition of xenobiotic-related deaths also has significant public health consequences. The findings may identify fatal drug reactions, preventable errors, or rapidly fatal epidemics associated with illicit xenobiotic use or malicious intent activities. In cases of occupational and environmental xenobiotic-related fatalities, interventions can be implemented to prevent subsequent morbidity and mortality. In addition, the gross and microscopic autopsy findings can elucidate mechanisms of xenobiotic toxicity.

Postmortem toxicologic techniques can be used in other types of investigations as well. For example, if carboxyhemoglobin is identified in burned human remains from an airplane crash, a cabin fire before descent is more probable than deaths due to a fire on impact. This type of postmortem analysis is useful in the reconstruction of events leading to the crash.^10,57,66,106

THE TOXICOLOGIC INVESTIGATION

Ordinarily, toxicologic samples are collected as part of an autopsy. In the hospital, when a death is assumed to be from natural causes, the hospital pathologist can perform an autopsy with the consent of the family. In a medicolegal investigation, however, the medical examiner determines the need for an autopsy and has statutory authority to act on that determination independent of familial consent or wishes. A “complete autopsy” is one that “completely” answers the questions needed to determine cause and manner of death; and usually includes gross external and internal inspection of the body and organs and often includes microscopical analyses as well as tissue and fluid sampling for microbiologic and toxicologic testing. Occasionally, only fluid samples are obtained along with an external inspection of the body if a complete autopsy is either unnecessary or the family has valid legal or religious grounds for objecting to an internal examination.

The precise list of xenobiotics screened in postmortem samples varies greatly by jurisdiction and by individual preference of the medical examiner, and is often case specific. Investigations in large cities routinely screen for hundreds of illicit, therapeutic, and environmental xenobiotics. Occasionally, the suspicions of the medical examiner or death scene investigator warrant specific assays that are performed only upon special request.

The sampling of fluid and tissue can be obtained minutes to years after death. The postmortem interval, defined as the time between death and autopsy, is important, but many other factors will influence the extent of bodily decomposition. Some of these factors include environmental conditions such as ambient temperature, humidity, aerobic versus anaerobic surroundings, immersion in water,⁶⁷ type of local flora and fauna, and the body habitus and metabolic state immediately prior to death. Samples for testing can be collected from a body during advanced stages of decomposition, after exhumation from graves, or even after embalming.^{10,39,40,47,85} Knowing the condition of the body at the time of sampling assists in interpreting the toxicologic findings. These postmortem changes are reviewed below.

DECOMPOSITION AND POSTMORTEM BIOCHEMICAL CHANGES

The first stage of decomposition is autolysis, during which endogenous enzymes are released and normal mechanisms maintaining cellular integrity fail.⁶⁰ Chemicals move across compromised and leaky cellular membranes down relative concentration gradients. Glycolysis continues in the red blood cells until intracellular glucose is depleted and lactate is produced. Ultimately, intracellular ions and proteins are released into the blood, and tissue and blood acidemia develops leading to the biochemical changes described in Table 140–3.¹⁰⁸

The next stage of decomposition in most normal environments is putrefaction. This stage involves digestion of tissue by bacterial organisms, which typically colonize the bowel or respiratory system. Later, additional organisms can be introduced by insects or other external sources. As the putrefactive process advances, the colors of the skin and organs change; epithelial blebs may form and separate from the underlying dermis; and gases accumulate, resulting in foul odors and bloating.¹⁰⁸

If death occurs in a very warm, dry climate, such as a desert or comparably arid environment, the body can desiccate so rapidly that putrefactive changes do not occur. This results in mummification and produces a lightweight cadaver with a tight, dry skin enveloping a prominent bony skeleton.¹⁰⁸

If the environment is very cold and devoid of oxygen, such as at great depths under water, putrefaction will be slowed. Anoxic decomposition of fatty tissues occurs, forming a white, cheesy material known as adipocere.

Another phase of decomposition, anthropophagia, occurs in unprotected environments where insects or large animals feed on the remains.¹⁰⁸

Because most postmortem changes result from chemical reactions that are temperature dependent, increased temperatures will accelerate the process and cooler temperatures will retard it. In general, morgue refrigerators achieve low enough temperatures (40°F {4°C}) to prevent further gross decomposition and many of its associated postmortem changes.

Another means of altering natural decomposition is embalming, a process of chemically preserving tissues that can be performed in a variety of ways.^48,49 Typically, blood is drained through large vessel pumps, and embalming fluid is injected intravascularly to perfuse and preserve the face or other tissues. Intracavitary spaces can be injected with the preserving substances, and solid organs are sometimes removed. Some authorities regulate the contents of the embalming fluid specifically to avoid confounding postmortem analysis. Most embalming fluid in the United States consists of formaldehyde, sodium borate, sodium nitrate, glycerin, water, and colorants.

SAMPLES USED FOR TOXICOLOGIC ANALYSIS

Unless the medical examiner has other suspicions about a death, only standard autopsy samples will be obtained from an otherwise intact body.^{26,39,40,55,62,90} These typically include samples of blood, gastric contents, bile, urine, and occasionally solid organs such as the liver or brain. Less commonly, vitreous humor and/or cerebrospinal fluid is obtained for analysis (Table 140–4). Antemortem specimens, if available, can be used for evaluation and comparison. These specimen analyses are reviewed in greater detail below.

Blood

Postmortem cell lysis limits the concept of plasma concentrations, and “blood” concentrations are reported instead. Most commonly, a single site such as femoral or subclavian blood is sampled unless an unusual xenobiotic with nonuniform distribution is suspected of causing the death. In patients with a prolonged postmortem interval and in cases where intravascular blood is coagulated, right heart blood can serve as an alternative sample.

Other sources of blood are sometimes available. These frequently include antemortem samples often from time of hospital admission and occasionally extravasated blood, which is unlikely to undergo extensive metabolism. Intracranial clots, in particular, serve as useful comparative samples in patients with a prolonged survival period after exposure to a xenobiotic.⁶⁸

In advanced states of decomposition, blood from the abdominal or thoracic cavities is less useful for some testing because it can be contaminated by bacteria or other substances that can affect xenobiotic metabolism, recovery, or analysis.

Vitreous Humor

Because of the relatively avascular and acellular nature of the fluid, the vitreous humor is well protected from the early decompositional changes that typically occur in blood.^18,21-24,27 When bodies are immediately refrigerated, creatinine, blood urea nitrogen, and sodium can be reliably approximated from vitreous humor samples for up to 3 or 4 days. Potassium concentrations are less reliable because cell lysis causes intracellular release. When vitreous glucose is elevated, hyperglycemia at the time of death can be assumed. A low vitreous glucose concentration, however, is a less reliable indicator of the antemortem serum glucose concentration, because a low vitreous glucose concentration can be attributable to either antemortem hypoglycemia or postmortem glycolysis even in the relatively avascular vitreous.

The aqueous content of the vitreous is normally higher than that of blood and can affect the partitioning of certain water-soluble xenobiotics, such as ethanol.

Urine

Urine may be available during the autopsy and can reveal renally eliminated xenobiotics or their metabolites. Because the bladder serves as a reservoir in which metabolism is unlikely to occur, the concentrations of xenobiotics obtained during the autopsy reflect antemortem urine concentrations. An isolated urine sample is of limited quantitative value but can be useful when compared with other sample sites.

Gastric Contents

The contents of the stomach are measured and grossly inspected for color, odor, and the presence or absence of pill fragments, food particles, activated charcoal, and other foreign materials.¹⁰⁵ Typically, gastric concentrations of xenobiotics are reported as milligrams of substance per gram of total gastric contents. Xenobiotic-induced pylorospasm, diminished intestinal motility, or decreased splanchnic blood flow may all decrease gastric emptying and affect the quantitative values obtained from sampling different parts of the gastrointestinal tract. Pills and pill fragments from modified release drugs are sometimes identified on autopsy after oral overdose in some cases after surviving several days postingestion.⁷⁵

Solid Organs and Other Sources

Xenobiotic concentrations in solid organs, such as the liver and brain, are usually reported as milligrams of substance per kilogram of tissue. Other tissue samples, including hair and nails, are used for thiol-avid xenobiotics such as metals. Rarely, tracheal aspirates of gases can be analyzed to confirm inhalational exposures when circumstances warrant testing. Pleural fluid analysis of postmortem xenobiotics typically yields qualitative results in decomposed bodies because redistribution of xenobiotics from the stomach and intestines can occur.^34,97 Cerebrospinal fluid (CSF) is inconsistently collected at autopsy and is not relevant in most cases. The ratios of CSF to heart blood are described for a number of xenobiotics, but data are limited antemortem because of the lack of published information, cause of death, or time of last exposure.

OTHER SAMPLING SOURCES

In an embalmed body, either the organs or tissues that remain relatively unembalmed, such as deep leg muscle,⁷⁰ or the embalming fluid itself can be used for analysis. When a body is disinterred, soil samples are usually obtained from above and below the coffin to validate identification of xenobiotics that leeched into, or out from the body.⁸⁵

On rare occasions, cremated remains, often referred to as cremains, are the only source of sampling available. Most battery-containing metallic implants, such as pacemakers, are removed before cremation to prevent explosions or environmental hazards.¹⁰⁷ Other metallic implants such as joint prostheses are removed in whole or in part after cremation depending on the particular alloy. Only dental remains, particulate matter, and occasionally calcified blood vessels are available for analysis.^5,114 In most cremations performed in the United States, the incineration process is followed by mechanical grinding to yield a fine particulate matter.¹¹⁴ The ability to extract xenobiotics from cremains is markedly limited at best, and little published data are available on the subject. A technique to identify heavy metals such as lead from cremains was described, but is not routinely used.^5,114

ENTOMOTOXICOLOGY

A variety of anthropophagic insect species can demonstrate the presence of xenobiotics.^2,44,64,65 This process of analysis is termed entomotoxicology. In putrefied bodies or bodies that have undergone anthropophagy, fluids and even insect parts can be analyzed. Forensic entomologists collect samples of these insects from the remains. After considering the stage of insect life, environmental conditions, and the season, the approximate time of death can be extrapolated. The family of flies Calliphoridae, such as the common blow fly, is attracted to unprotected remains by a very fine scent that develops in the cadaver within hours of death. The adult fly lays eggs on mucosal surfaces or in open wounds. After the eggs hatch, the larvae feed on the decomposing tissue. Larval samples can then be examined for the presence of xenobiotics. To achieve accurate analysis, these samples must be preserved immediately after collection because living larvae can continue to metabolize certain xenobiotics. In the next phase of their life cycle, the larvae undergo pupation, secreting a substance that encloses them into pupal casings until they hatch as adults. These pupal casings are often found in the soil beneath the body or in the body cavities. Some xenobiotics are identified in the pupal casings long after the adult fly has emerged and flown away (Table 140–5).⁹¹

INTERPRETATION OF POSTMORTEM TOXICOLOGY RESULTS

After fluid and tissue samples are collected and analyzed for the presence of xenobiotics, the process of interpreting these results begins. This complex task attempts to account for the clinical effects of a xenobiotic at the time of death by integrating medical history, autopsy, death-scene findings, and toxicologic reports. Multiple confounding variables can affect the final concentrations of xenobiotics from the time of death to the time of testing after the autopsy. Variables include the nature, metabolism, and distribution of the xenobiotic; the body habitus and state of health of the decedent prior to death; physical and environmental variables during the postmortem interval; the techniques of sampling and analysis; and other variables of the autopsy (Tables 140–6 and 140–7).

Variables Relating to the Xenobiotic

Postmortem Redistribution

The xenobiotic concentration in blood can be higher during the time of sampling at autopsy than at the actual time death occurred if significant postmortem redistribution occurs.^{56,88,113-120} Redistribution typically occurs with xenobiotics that have large volumes of distribution and when postmortem changes result in release of intracellular xenobiotic into the extracellular compartment.⁹⁴ For example, amitriptyline can be released from tissue stores into the blood as autolysis progresses, resulting in a significantly higher blood concentration at the autopsy than at the actual time of death. If postmortem redistribution is not considered, xenobiotic concentrations obtained during the autopsy can be misinterpreted as being supratherapeutic or even toxic, and the cause of death can be inappropriately attributed to this xenobiotic.

Postmortem Metabolism

Less commonly, xenobiotic concentration can decrease secondary to postmortem metabolism. For example, cocaine continues to be degraded after death by endogenous enzymes such as cholinesterases in the blood, which continue to function in postmortem tissue and in vitro. Unless blood is collected immediately after death and placed in tubes containing enzyme inhibitors such as sodium fluoride, the concentration of cocaine will decrease, and the analysis will not accurately reflect the concentration of the drug at the time of death.^{61,76,106,111} All available information regarding postmortem redistribution or metabolism of a specific xenobiotic must be evaluated for the proper interpretation of the toxicologic results.

State of Absorption and Distribution

Both in the living and deceased, the state of absorption, distribution, and other toxicokinetic principles affect the apparent concentration of a sampled xenobiotic. For a xenobiotic with minimal postmortem metabolism or redistribution, the phase of absorption is suggested by the relative quantity of the xenobiotic in different fluids and solid organs. For example, a high concentration of xenobiotic in the gastric contents, with progressively lower concentrations in the liver, blood, vitreous, and brain, suggests an early phase of absorption at the time of death. When a xenobiotic is orally administered and the tissue concentration is highest in the liver, the relationship suggests a postabsorption phase but a predistribution concentration. A concentration found to be highest in the urine suggests that the xenobiotic was in an elimination phase at the time of death. Although this approach has limitations, it is important for correlating the state of absorption and the expected clinical course of the xenobiotic. Unfortunately, multiple samples are not always available at the time of autopsy or the interpretation of reports, and opportunities for subsequent sampling are often limited.

Xenobiotic Properties

Xenobiotic chemical composition affects partitioning into fat or water compartments and total body burden is poorly captured when sampling the blood compartment. The significance of this and partitioning of blood to serum is uncertain.⁷⁴ Xenobiotic stability refers to the ability of a xenobiotic to maintain its molecular integrity despite postmortem changes such as decomposition of the body, adverse storage conditions, or the lack of preservatives.^{7,15,17,58,59,64,74,100,109,116-119}

Postmortem xenobiotic stability was assessed in homogenized liver tissue infused with various concentrations of xenobiotics.¹⁰⁹ The samples were allowed to putrefy outdoors, and sequential sampling of xenobiotic concentrations was performed. The xenobiotics that decreased in concentration as putrefaction progressed were considered labile, and samples with a constant concentration were considered stable. The authors proposed that the chemical characteristics of a xenobiotic determine its stability. For example, labile xenobiotics share the molecular configuration of an oxygen–nitrogen bond, thiono groups, or aminophenols. Conversely, chemical structures that enhance stability include single-bonded sulfur groups, carbon–oxygen and carbon–nitrogen bonds, and sulfur–oxygen and hydrogen–nitrogen bonds. Although not explicitly studied in otherwise intact but putrefying bodies, logically, a less stable xenobiotic can be recovered in a lower concentration than the actual concentration at the time of death. This must be considered when information regarding stability is available and the body of a decedent is in an advanced stage of decomposition.

Xenobiotic Chemical Reactions

An artifact can result from a chemical reaction with a xenobiotic added during the postmortem interval, such as embalming fluid.⁴¹ In a study of xenobiotic-spiked blood and formalin in test tubes, amitriptyline was formed by the methylation of nortriptyline.^29,119 Identification of amitriptyline, which was not present at the time of death, could confuse the interpretation of toxicologic analyses.

Expected Clinical Effects of the Xenobiotic

For a fatality to be attributed to a xenobiotic, the expected clinical course from the exposure should be consistent with the autopsy findings. For example, what are the implications of a person found dead minutes after ingesting pills, if a large concentration of acetaminophen (APAP) is identified in both the gastric contents and blood but not in other tissues at autopsy?⁹⁹ Although suicidal intent can be supported by this finding, the onset of death within minutes is inconsistent with a fatality due to an APAP overdose. Thus, another cause of death must be sought. Interpretation of postmortem toxicology must also incorporate clinically relevant consequences of xenobiotic interactions. For example, the combined ingestion of phenobarbital and ethanol can cause fatal respiratory depression. Although neither are necessarily fatal alone, their additive effects must be acknowledged during toxicologic interpretation.

Variables Related to the Decedent

Comorbid Conditions

The clinical response to a xenobiotic can be affected by acquired and inherited physiologic conditions that are not always identified or identifiable on autopsy. A thorough medical history is important and can assist in interpreting the clinical effects of a xenobiotic exposure. Similarly, certain clinical conditions can produce substances that interfere with postmortem laboratory assays. For example, an individual with a critical illness can produce digoxinlike immunoreactive substances (DLIS), which may cross-react with the postmortem digoxin assay.⁸ Without knowledge of DLIS production, the results can confound toxicologic analysis (Chap. 62).

Tolerance

Tolerance is an acquired condition in which increasingly higher xenobiotic concentrations are required to produce a given clinical effect. It is an important consideration for deaths in the presence of opioids, ethanol, and sedative–hypnotics. For example, respiratory depression and death from methadone is easily diagnosed in an opioid-naive individual with a history of methadone exposure and methadone-positive postmortem samples. However, the same methadone concentrations in a patient on chronic methadone maintenance therapy will not produce the same outcome. Unfortunately, no autopsy markers are available to indicate tolerance, and no biochemical or histologic markers are available during autopsy that can be used to predict clinically dangerous xenobiotic concentrations in a tolerant individual.³⁰ Complex postmortem assays analyzing opioid receptors are not routinely used.³⁸ Postmortem assessment of tolerance ultimately depends on knowledge of the patient, pharmacokinetics of the xenobiotic, and the best judgment of the investigator.

Pharmacogenetics

There is genetic variability in the expression of certain metabolic enzymes. For example, pharmacogenetic differences in metabolic enzymes such as CYP2D6 can affect the metabolism of codeine to morphine. Individuals with duplicate alleles for CYP2D6 may be ultra-rapid metabolizers, rendering them susceptible to morphine toxicity despite generally acceptable dosing of the parent compound codeine. Deaths in young children are reported from bioaccumulation of morphine. In such cases, postmortem blood analysis can reveal an elevated concentration of morphine, the codeine metabolite, and raise suspicion for a malicious overdose. Postmortem genotyping can provide an alternate explanation. Such distinctions are not routinely identifiable on autopsy.^19,31,32 Postmortem evaluation of frozen tissue or blood can also identify genetic predisposition of long-QT syndromes that can be consequential in both the diagnosis of sudden death that can be considered along with toxicologic data, but such testing is not routine across jurisdictions.⁷¹

Variables Relating to the Autopsy

State of Decomposition

In bodies with advanced stages of decomposition, xenobiotics diffuse from depot compartments such as the stomach urinary or gall bladders into adjacent tissues and blood vessels and secondarily affect sample concentrations near sites of origin of diffusion.^{25,39,68,81,82,93-95}

During putrefaction, bacteria cause fermentation of endogenous carbohydrates, resulting in ethanol formation. In decedents without gross evidence of putrefaction, especially those in cool, dry environments, endogenous ethanol production is minimal.^24,30,84 With a longer postmortem interval or in an environment that is more conducive to ethanol production, the distinction between endogenous and exogenous sources of ethanol becomes more difficult. Sampling from multiple sites is often useful in making the distinction, but not definitive.¹¹⁰ The pattern of ethanol detected in the blood and absent from the urine and vitreous samples is more suggestive of postmortem ethanol production. However, postmortem ethanol production may occur in the vitreous in decedents with hyperglycemia at the time of death and significant putrefaction at autopsy. A comprehensive review of interpreting postmortem ethanol concentrations is available elsewhere.^68,84

Handling of the Body or Samples

Inappropriate handling of the body can result in artifacts.^99,101 In one reported case, methanol was detected in the vitreous humor of a decedent after embalming.¹⁴ The methanol was subsequently traced to a spray cleanser that likely settled on the surface of open eyes during washing of the body.

In addition, inappropriate handling of samples can affect xenobiotic concentrations. In one study, autopsy blood was obtained from a diabetic man who died of bronchopneumonia. The samples were stored at 40°F (4°C) and tested at 2 and 5 days postmortem. The blood ethanol increased from 0.4 to 3.5 g/L due to an inadequate addition of fluoride preservative to the samples. The combination of inadequate preservation, hyperglycemia (vitreous glucose of 996 mg/dL), and bacterial sepsis created an ideal environment for ethanol production by fermentation.⁶⁸

In the United States, heavy metals are currently banned for use in embalming because they can contaminate subsequent evaluation for metal poisoning. Formalin can also affect stability or quantitative identification of some xenobiotics. When necessary, an analysis of the embalming fluid used in the body or soil samples from around disinterred bodies may facilitate the toxicologic investigation.^{21,36,116,118,119}

Autopsy Findings

In many xenobiotic-related deaths, the anatomic findings are nonspecific,¹¹⁶ but in some cases, the autopsy does reveal confirmatory or supportive findings. A large quantity of undigested pills in the stomach is consistent with an intentional overdose, and suicide should be considered as the manner of death. Centrilobular hepatic necrosis can be found in decedents with a history of acetaminophen overdose. The autopsy can also reveal other findings such as coronary artery narrowing, chronic hypertensive changes, renal abnormalities, or a clinically silent myocardial injury. Such information is useful to assess the potential effects of a xenobiotic in a patient with previously undiagnosed conditions. In other cases, the absence of a chronic condition can be strongly suggestive of a xenobiotic-related death. For example, a decedent with an autopsy finding of aortic dissection in the absence of chronic hypertensive findings or other predisposing conditions suggests a xenobiotic-induced hypertensive crisis as can occur from the use of cocaine or other sympathomimetics.

Artifacts Related to Sampling Sites

Site-specific differences in postmortem xenobiotic blood concentrations are common.³⁷ For example, blood obtained from femoral vessels can have low glucose concentrations because of postmortem glycolysis, but the blood glucose concentration removed from the right heart chambers can be high as a result of perimortem release of liver glycogen stores. Concentrations between heart, femoral, and subclavian samples vary significantly.⁸⁰ As noted above, hyperglycemic conditions are more reliably assessed from sampling the vitreous humor because it is an acellular fluid and in an environment that is relatively protected from diffusion artifacts in the early postmortem interval. An elevated vitreous glucose concentration strongly suggests antemortem hyperglycemia. The individual interpreting the toxicologic report must know the exact site from which the sample was taken.^22,62

Ideally, samples from more than one site would be available for comparison; fortunately, multiple site sampling is now a standard procedure required for accreditation. In decendents with minimal putrefaction or decomposition, comparison of concentrations from different sites can reveal important information regarding the extent of xenobiotic absorption at the time of death and acute versus chronic exposure.^{11-13,28,32,52-54,77,87,90,93,96-98,103,110,112}

Other Considerations

Published therapeutic, toxic, and fatal xenobiotic concentrations from autopsy samples as well as postmortem redistribution data are available to aid in the interpretation of these specimens.^6,73 However, the conditions associated with reported concentrations (“average” range) do not necessarily permit comparisons with the concentrations of the particular case under investigation. Thus, these resources although valuable, should not be accepted as absolute values that specifically define either toxic or therapeutic concentrations. Similarly, formulas available for assessing xenobiotic doses or concentrations in the living are not usually applicable when analyzing postmortem samples.

Other Limitations

Although there are generalized standards of practice in forensic investigations, specimen collection and laboratory methodologies may vary.^3,6 Some xenobiotic concentrations may be falsely elevated or depressed depending on the chosen methodology.^33,77 Descriptions of specific toxicology laboratory techniques are beyond the scope of this chapter, but these variables must also be considered in the interpretations of results. Other limitations may include a lack of information relating to the circumstances of death, or the possible compromise in specimen handling to maintain proper chain-of-custody for evidence that is often of paramount importance in medicolegal autopsies.^39,40

Case Discussion

Interpreting the oxycodone concentrations obtained from postmortem femoral samples requires information about the medical and behavioral health of the decendent before death, medication history, interviews with family, coworkers, scene investigation, and thorough review of all medications administered during the hospital stay. A subsequent evaluation revealed a previously undisclosed history of opioid misuse syndrome and recent discharge from an opioid rehabilitation program. A review of hospital records revealed that no oxycodone was administered in the hospital. The scene investigation and evaluation of the train and rails revealed no clear mechanical failure. The concern is for illicit oxycodone use. Samples of blood from the subdural hematoma revealed high concentrations of oxycodone. Although there may be postmortem drug redistribution, extravasated blood is generally protected from these changes and likely more closely related to antemortem concentrations. Ideally, a sample of blood from admission or recovered from the blood bank should be sent for forensic toxicologic studies. A family member subsequently came forward, having found a bottle of pills with the markings of oxycodone in the decedent’s home. The family member remarked that he seemed unusually drowsy prior to leaving for work and they were suspicious for relapsed opioid misuse. A coworker confirmed that he seemed lethargic and asked if he was “okay.” An analysis of the tablets revealed that the pills were oxycodone. The cause of death was deemed traumatic brain injury due to blunt force injury while intoxicated by oxycodone. The manner of death was deemed an accident, the legal term for a nonnatural, unintentional death.

SUMMARY

• Accurate interpretation of postmortem toxicology reports requires an understanding of the potential biochemical changes that affect these samples.

• It is exceptionally difficult to make an absolute correlation between the xenobiotic concentrations of postmortem blood and tissue to the concentrations in comparable samples from the actual time of death.

• Postmortem toxicology is an evolving discipline that permits identification of the most likely cause and manner of xenobiotic-related deaths.

• Progress in this field will depend on the continued collaboration between the treating physicians, medical and forensic toxicologists, and medical examiners.

REFERENCES

INTRODUCTION

Xenobiotics cause brain death as a result of the unique vulnerabilities of the central nervous system. With supportive care, however, many such patients are suitable candidates for organ donation.^11,12,38,47

Early identification of potential donors is critical because the viability of transplantable tissue diminishes as duration from the time of brain or cardiac death prolongs.³⁰

Donor identification by clinicians is one barrier. Among a group of toxicologists given scenarios of brain death from poisoning due to cocaine or carbon monoxide, toxicologists had variable perceptions of donor and organ suitability.⁴⁵ Such inconsistency can affect the potential pool of donors in patients with clinical signs of brain death.

Some xenobiotic poisonings can confound clinical assessments of brain stem function and mimic brain death (Table SC12–1). Published brain death protocols require exclusion of poisoning,^12,43,49 but limited guidance is provided.⁴⁴ Obtaining and interpreting xenobiotic concentrations is complex and, in many circumstances, not feasible. Few xenobiotics have timely and readily available assays. This is further complicated when the exposure history is unknown. Determining brain death from novel psychoactive substances (NPS) and illicit synthetic opioids is especially challenging. Routine testing for many NPS is often unavailable or uninterpretable, further complicating decisions regarding donor suitability.

Other methods of establishing brain death are utilized such as cerebral blood flow. The effects of xenobiotics on these studies are not well described.³⁴

Once brain death is established, organ procurement personnel assist in obtaining familial consent, deciding which organs are most suitable for transplant, and maximizing physiological support and perfusion until organ procurement occurs.⁴⁷ The protocols following controlled or uncontrolled donation after cardiac arrest are socially complicated and practiced extensively in Europe.⁴²

Successful transplantation of organs is reported from poisoned donors associated with a variety of xenobiotics (Tables SC12-2 and SC12–3).^{2,3,6,8,11,12,18,26,29,36,38-41,46} Although some xenobiotics are highly toxic, such as cyanide and carbon monoxide (CO), transfer of clinical poisoning to the organ recipient is not reported. This is likely caused by several factors, including xenobiotic metabolism, tissue redistribution or binding prior to procurement, as well as the means of handling organs during the transplantation process. For example, some xenobiotic clearance occurs in the myocardium during organ rinsing and cardiopulmonary bypass.³⁵ Furthermore, individual organs do not uniformly manifest toxicity following specific xenobiotic insults. For example, the heart of a CO-poisoned donor was examined after a transplantation failed for technical reasons. The myocardium did not demonstrate histologic signs of CO poisoning.⁴¹ A comprehensive review with organ-specific procurement recommendations post-CO poisoning is presented elsewhere.⁴