Physical Address

304 North Cardinal St.
Dorchester Center, MA 02124

Opioid Overdose


History

Opiates have been used for thousands of years, with evidence of opium poppy (known as Hul Gil, “joy plant”) cultivation more than 5000 years ago among the Sumerians in the area of Mesopotamia, near modern day Iraq. Its use for rituals and mystical purposes is evidenced by poppy seed capsules discovered in burial sites in Neuchatel, Switzerland and the Cueva de los Murciélagos , Spain, which have been carbon-14 dated to 4200 BCE. As use spread throughout the Mediterranean, opium’s medicinal properties were recognized.

The Greek physician Galen may have been the first to record an opium overdose in around 140 CE, during which he reportedly treated the patient with an emetic made of sweet wine. In 1656, Christopher Wren experimented with intravenous administration (using animal bladders and goose quills) of opium to dogs and described overdose, and several years later, German scientists J.D. Major and J.S. Elsholtz performed similar experiments on humans with reported toxicity. A century and a half later, Friedrich Sertürner, the German pharmacist who isolated morphine from opium in the early 1800s, reported that he experimented with the alkaloid by administering it to himself, three boys, three dogs, and a mouse. One of the dogs reportedly died, and he described the effect that morphine had on himself and his three young “volunteers” as “near fatal.” The first recorded human fatality from a morphine overdose appears to be from the 1850s when Scottish physician Alexander Wood, one of the first to use and perfect the hypodermic needle and syringe, reportedly performed one of the first injections of morphine on his wife who subsequently died from respiratory depression, although others claim that this story is not true. Overdose is also discussed with regards to laudanum, a mixture of opium and alcohol, associated with overdose in mid-19th century. The Brooklyn (New York) Daily Eagle , January 10, 1861, reported two unrelated instances in a single day. Elizabeth Siddal an artist and wife of Dante Gabriel Rosetti, took laudanum for various illnesses and died of an overdose in February 1862.

Definitions

The term “overdose” is used variably by the lay public, health professionals, and in the medical literature and generally refers to an “excessive amount” of a substance (the noun) or the act of taking such an amount (the verb). This excessive amount is dependent on an individual’s tolerance to the specific substance. Associated signs generally include the “opioid overdose triad” of constricted (“pinpoint”) pupils, altered level of consciousness, and respiratory depression (both rate and effort). Additional signs may include blue or ashen skin, nails and lips; gurgling or snore-like sounds (“death rattle”); decreased blood pressure and heart rate; and pulmonary edema ( Box 54.1 ).

Box 54.1
Signs of Opioid Overdose

  • “Overdose Triad”

    • Constricted (“pinpoint”) pupils

    • Altered level of consciousness

    • Respiratory depression (both rate and effort)

  • Blue or ashen skin, nails, and lips

  • Gurgling or snore-like sounds (“death rattle”)

  • Decreased blood pressure and heart rate

  • “Froth” from mouth (from pulmonary edema)

There are no consistent guidelines or cutoffs to distinguish overdose from intoxication or being under the influence. Some consider an overdose only if the individual has lost consciousness and is unarousable by external stimulation. Number of breaths per minute used to define overdose range from less than 12 per minute to less than 8 per minute. The term “opioid-induced respiratory depression” is sometimes used to describe the side effect of opioid used medically in an attempt to distinguish this effect from the term “overdose” with the associated negative connotation. Another term used to address the complexities of tolerance, co-ingestion, medical comorbidities, and environmental context is “opioid-related overdose.”

The term “overdose” is not recognized in standard medical diagnostic nomenclature such as the International Classification of Diseases and Related Health Problems, Tenth Revision, Clinical Modification (ICD-10-CM) and the Diagnostic and Statistical Manual of Mental Disorders , Fifth Edition (DSM-5). What is commonly thought of as “overdose” can fall under one of two areas: “Poisoning” in the “Injury, Poisoning and Certain Other Consequences of External Causes” section or “Intoxication” in the “Mental, Behavioral and Neurodevelopmental Disorders” section.

International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Clinical Modification (ICD-10-CM)

Injury, Poisoning, and Certain Other Consequences of External Causes

“Poisoning by, adverse effect of & under-dosing of drugs, medications & biological substances” (T26-50).

This category includes adverse effect of correct substance properly administered, poisoning by overdose of substance, poisoning by wrong substance given or taken in error, and underdosing by taking less substance than prescribed or instructed. The determination of poisoning versus adverse effect is based on how the substance was used. If the correct substance was administered as prescribed, the condition is classified as an adverse effect. Using the prescribed medication less frequently than prescribed, in smaller amounts, or not using the medication as instructed by the manufacturer is not coded as poisoning but as underdosing. When the condition is a poisoning, the poisoning code is sequenced first, followed by additional codes for all manifestations. If there is also a diagnosis of abuse or dependence on the substance, the abuse or dependence is also coded. The poisoning code is also used when a condition results from interaction of a therapeutic drug used correctly with a nonprescription drug and/or alcohol. All involved substances should be coded separately.

Poisoning by opioids is designated by the code T40. There are specific codes (first decimal place) for heroin (T40.1), other opioids (T40.2), methadone (T40.3), and other synthetic narcotics (T40.4). In addition, a determination is to be made (third decimal place) as to whether the poisoning was unintentional (.XX1), intentional self-harm (.XX2), assault (.XX3), or undetermined (.XX4). Additional codes specify initial encounter, subsequent encounter, and sequela (fourth decimal place) ( Box 54.2 ).

Box 54.2
ICD-10 Multiple Cause-of-Death Codes for Poisoning

  • T40.1 Poisoning by and adverse effect of HEROIN

  • T40.1X1 Poisoning by heroin, accidental (unintentional)

  • T40.1X2 Poisoning by heroin, intentional self-harm

  • T40.1X3 Poisoning by heroin, assault

  • T40.1X4 Poisoning by heroin, undetermined

  • T40.2 Poisoning by, adverse effect of OTHER OPIOIDS (natural and semisynthetic)

  • T40.2X1 Poisoning by other opioids, accidental (unintentional)

  • T40.2X2 Poisoning by other opioids, intentional self-harm

  • T40.2X3 Poisoning by other opioids, assault

  • T40.2X4 Poisoning by other opioids, undetermined

  • T40.2X5 Adverse effect of other opioids

  • T40.3 Poisoning by, adverse effect of METHADONE

  • T40.3X1 Poisoning by methadone, accidental (unintentional)

  • T40.3X2 Poisoning by methadone, intentional self-harm

  • T40.3X3 Poisoning by methadone, assault

  • T40.3X4 Poisoning by methadone, undetermined

  • T40.3X5 Adverse effect of methadone

  • T40.4 Poisoning by, adverse effect of OTHER SYNTHETIC NARCOTICS (other than methadone)

  • T40.4X1 Poisoning by other synthetic narcotics, accidental (unintentional)

  • T40.4X2 Poisoning by other synthetic narcotics, intentional self-harm

  • T40.4X3 Poisoning by other synthetic narcotics, assault

  • T40.4X4 Poisoning by other synthetic narcotics, undetermined

  • T40.4X5 Adverse effect of other synthetic narcotics

Mental, Behavioral and Neurodevelopmental Disorders: Mental and Behavioral Disorders Due to Psychoactive Substance Use

The diagnostic categories listed here are meant to designate the behavioral disorders of substance abuse and dependence. Although not meant to specifically designate the phenomenon of overdose, they are sometimes used interchangeably in studies, reports, and so on. They include Opioid use (F11.92), abuse (F11.12), or dependence (F11.22) with intoxication. Specifiers include uncomplicated (.XX0), with delirium (.XX1), with perceptual disturbance (.XX2), and unspecified (.XX9). There are not separate codes for the different types of opioid but the clinician is encouraged to add the name of the specific opioid (i.e., heroin, oxycodone, methadone) in addition to the numeric code.

Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5)

The other major diagnostic coding system used for behavioral disorders in the United States is the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Substance-Related and Addictive Disorders with Intoxication falls under the “Substance-Induced Disorder” category. With regard to Substance Use Disorders, the DSM-5 is similar to ICD-10-CM except it uses the term Opioid Use Disorder with the severity specifiers of Mild, Moderate, or Severe, whereas ICD-10-CM uses Abuse and Dependence codes (similar to the fourth edition [text revision] of the DSM [DSM-IV-TR]). With the DSM-5, a severity specifier of mild is the equivalent of ICD-10-CM abuse, whereas a severity specifier of moderate or severe is equivalent to ICD-10-CM dependence. The DSM-5 does not have a “Use” diagnostic category, although it does have an intoxication code for “Opioid Intoxication with no comorbid opioid use disorder.” There are four criteria for Opioid Intoxication:

  • A.

    Recent use of an opioid.

  • B.

    Clinically significant problematic behavioral or psychological changes (e.g., initial euphoria followed by apathy, dysphoria, psychomotor agitation or retardation, impaired judgment) that developed during, or shortly after, opioid use.

  • C.

    Pupillary constriction (or pupillary dilation due to anoxia from severe overdose) and one (or more) of the following signs or symptoms developing during, or shortly after, opioid use: (1) drowsiness or coma, (2) slurred speech, (3) impairment in attention or memory.

  • D.

    The signs or symptoms are not attributable to another medical condition and are not better explained by another mental disorder, including intoxication with another substance.

The clinician can add a specifier “with or without perceptual disturbance.”

Unlike DSM-IV-TR, where the intoxication and use disorder (abuse or dependence) were listed individually, DSM-5 uses a code that combines the use disorder and intoxication (in order to distinguish it from intoxication in an individual who does not have a history of a use disorder).

Defining Fatal Overdose

Historically, there has been a great deal of inconsistency in the way that “poisoning” or intoxication-related deaths have been categorized. Because some states have a system based on a centralized medical examiner, whereas others use jurisdictional coroners, there is often a great deal of variability within and between states. Because the numbers reported have significant implications for public health surveillance and resultant prevention strategies, it is extremely important that these deaths be reported in a manner that is consistent as possible. The National Association of Medical Examiners along with American College of Medical Toxicology released guidelines for the investigation, diagnosis, and certification of deaths related to opioid drugs. These recommendations included a complete autopsy, a complete scene investigation (including reconciliation of prescription information and pill counts), and comprehensive toxicological testing (of blood, urine, and vitreous humor) to include opioid and benzodiazepine analytes as well as other depressants, stimulants, and antidepressants. All of these are to be used in conjunction with medical history to determine four key components of the death certificate: Cause of Death, Other Significant Conditions Contributing to Death, Manner of Death, and How Injury Occurred. For Cause of Death, all substances believed to have been responsible for the death, and present in sufficient concentrations, are listed (as opposed to vague statements such as “mixed drug intoxication”). The Other Significant Conditions section lists conditions that might have predisposed a person to death (such as sleep apnea) but were neither necessary nor sufficient to cause the death. The Manner of Death is classified as either accident, suicide, homicide, or undetermined. The recommendations encouraged the use of “accident” for all cases of misuse or abuse without any apparent intent of self-harm and stressed that the “undetermined” designation should be reserved only for those cases for which the evidence exists to support more than one possible determination (as opposed to the more common use when the manner is not absolutely certain). Finally, the How Injury Occurred section should include the information about medical history, route of administration, drug formulation (long-acting, extended release, or immediate-release), and source of drug (prescription, illicit, diverted).

The Substance Abuse and Mental Health Services Administration (SAMHSA) convened a consensus panel that addressed many of the same issues and added some additional case definitions: Drug-Caused Death, which refers to deaths that resulted from exposure to a substance regardless of the intent (accident, suicide, homicide) of the individual; and Drug-Detected Death, which refers to a death in which a drug is detected regardless of the drug’s role in causing the death. A subcategory of Drug-Caused Death is Drug Poisoning Death, which refers specifically to deaths caused by acute exposure. The US National Center for Health Statistics (NCHS), which releases the annual National Vital Statistics Report Deaths, Final Data , has also developed similar categories for drug-induced deaths; however, that classification also includes deaths involving adverse effects of drugs for therapeutic use.

The Centers for Disease Control and Prevention (CDC), based on the external-cause-of-injury matrix, recommends defining an overdose when the first-listed E-code is “drug poisoning” and the principal diagnosis is “injury.” Consensus recommendations from the Safe States Alliance’s Injury Surveillance Workgroup on Poisoning (ISW7) recommends two alternative case definitions: one in which an overdose is defined when either the first-listed E-code or principal diagnosis indicates drug poisoning” and the second when an overdose is defined when any E-code or diagnosis lists drug poisoning. One study looking at data from Kentucky found a 50% increase in overdose cases identified when using the second ISW7 definition as compared to the CDC recommended definition. In addition, in 1999, the ICD-10 replaced the previous revision of the ICD (ICD-9) as the classification system used by medical examiners in the determination of cause of death.

Epidemiology

Worldwide

Fatal Overdose

Reliable data on the epidemiology of the fatal overdose are variable and limited by the lack of consistency in case identification, coding, and reporting. The determination of the cause of death varies from country to country, within a given country and, often, within local jurisdictions. Toxicological verification is often not available, extremely limited, or imprecise. Many countries report on drug-related fatalities with some, in addition to drug overdoses, also including deaths due to HIV acquired through injecting drug use, suicide, and unintentional deaths and trauma due to illicit drug use. All of this makes comparisons between countries and regions difficult. With this understanding, the World Health Organization (WHO) estimates that there were 187,100 drug-related deaths in 2013, corresponding to a mortality rate of 40.8 drug-related deaths per million people 15–64 of age. By region, the estimated rates per million are: Africa, 61.9; North America, 136.8; Latin America and the Caribbean, 18.4; Asia, 28.2; West and Central Europe, 22.5; Eastern and Southeastern Europe, 41.5; and Oceania, 82.3. Although most countries report that opioids are the main drugs involved in these deaths, reliable numbers are not available for many countries. Other studies have looked at population-based crude and adjusted mortality rates in various areas and found a range of 0.04 to 46.6 per 100,000 person-years and 0.11–253.8 per 100,000 person-years, respectively. In addition, the largest systematic review of global deaths in people who inject drugs found that overdose is the leading cause in all areas of the world.

The European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) monitors drug-use patterns in the European Union (EU) as well as Norway and Turkey. Despite large variability in case definition and reporting, the EMCDDA represents the best estimate of drug-related morbidity and mortality in Europe. According to their data, most EU countries have seen increases in overdose deaths from 2003 to 2009, at which time there was a leveling or decline for several years in some countries. Increases have been seen in a number of EU countries since 2012, with highest increases in Sweden, Spain, and Turkey as well as Ireland, Lithuania, and the United Kingdom (which saw a doubling of heroin-related overdoses between 2013 and 2015, the highest rates since the early 1990s). For most countries, between 2006 and 2014, overdose rates have increased in older adults and decreased in younger adults. Drug overdoses are estimated to account for 3.5% of all deaths among Europeans 15–39 years of age. In 2014, there were approximately 6800 overdose deaths reported in the EU, 82% of which involved an opioid and 78% in males. The overall rate for the EU is approximately 19.2 overdose fatalities per million population 15–64 years of age (18.3 in the EU plus Turkey and Norway), with individual national rates ranging from 2.4 per million in Romania to 113 per million in Estonia. Other countries with rates over 40 per million include Denmark, Finland, Ireland, Lithuania, Norway, Sweden, and the United Kingdom. Heroin or its metabolites were found in the majority of fatal overdoses, although other substances were often found in combination. Prescription opioids most commonly mentioned in toxicology reports are methadone, fentanyl, tramadol, and buprenorphine. Starting in around 2013, parts of Europe also began to see increases in overdoses related to illicitly manufactured fentanyl and other designer fentanyl analogs. Various areas of Canada have also seen substantial increases in overdose fatalities involving illicitly manufactured opioids.

Nonfatal Overdose

Although fatalities related to opioid use are clearly tragic, nonfatal overdose is more common and exerts a huge cost, both economically and personally. Data on nonfatal overdoses are even more unreliable than data for fatal overdoses due to variability of definition, diagnosis, case identification, coding, and reporting. A number of studies from various parts of the world looking at individuals with histories of opioid misuse suggest that 16%–80% of those interviewed had experienced an overdose in their drug-using career, many of whom had experienced overdoses in the previous 6–12 months. a

a References 43, 46, 68, 140, 160, 178, 184, 261, 262, 299, 363, 436, 544, 589, 723.

Other studies have estimated that 4%–5% of all overdose cases are fatal, with a cumulative risk of death increasing with each successive overdose.

United States

Fatal Overdose

Although not as variable as the majority of other countries, data on overdose within the United States do exhibit a fair amount of variability from state to state. On average, about 81% of all death certificates list the specific drug(s) contributing to the death, with some states reporting the specific drug(s) in less than 50% of their death certificates. In addition, there are issues with the manner in which coded diagnosis are used in gathering data: some are based on the external-cause-of-injury code (E-code) and others are based on the principal diagnosis.

The United States reports one of the highest drug-related mortality rates worldwide (at 4.6 times the global average) and accounts for approximately 20% of drug-related deaths globally. The higher mortality rate in North America likely, in part, reflects better monitoring and reporting. The relative contribution of opioids to all overdose deaths has risen from about 38% in 2004 to more than 63% in 2013. Prescription opioids account for approximately 75% of all prescription drug-related deaths.

Given that data before and after 1999 are not entirely comparable, it still appears that fatalities related to drug overdose have increased steadily since the early 1970s. In 1980, there were an estimated 6100 drug poisoning deaths, a rate of 4.8 per 100,000. During this period, the categorization of drug poisoning deaths did not allow easy distinction between deaths caused by prescription drugs and deaths caused by illicit drugs. Because the category of “opiates” did not distinguish between heroin and prescription opiates, it is difficult to determine how much of the change in opiate-related deaths was attributed to heroin and how much to prescription opioid analgesics.

Beginning in 1999, a new coding protocol was introduced that allowed researchers to better determine which drugs were involved in fatalities by allowing disaggregation of the “narcotics” category into the three largest components: heroin, cocaine, and opioid analgesics. Between 1999 and 2004, the number of annual deaths related to unintentional drug poisoning continued to rise to more than 20,000 in 2004. During this period, the gradual increase in cocaine-related mortality continued, whereas the number of deaths involving heroin stabilized. In contrast, the number of deaths involving prescription opioid analgesics increased from roughly 2900 in 1999 to 7500 in 2004, an increase of 160% in 5 years. By 2004, opioid painkiller deaths numbered more than the total of deaths involving heroin and cocaine combined. Overdose deaths involving prescription opioids quadrupled between 1999 and 2014, from a rate of 1.5 deaths per 100,000 persons to 5.9 deaths per 100,000 persons (more than 165,000 deaths over that period). The group with the highest risk for fatal prescription opioid–related poisoning was white, middle-aged men. As compared to the early 1990s, when overdose rates were lowest in the rural states, the highest mortality rates by 2004 were in New England, the Appalachian states, and the Southwest. In 2015, the four states with the highest drug overdose death rates per 100,000 population were West Virginia (41.5), New Hampshire (34.3), Kentucky (29.9), and Ohio (29.9).

In addition, there was a correlation between state drug poisoning rates and state sales of prescription opioids, with a nearly fourfold difference among states in their use of opioid analgesics. b

b References 97, 98, 436, 493, 521, 521, 666.

The number of prescriptions for opioids written in the United States (which prescribes an estimated 80% of all opioids prescribed in the world) more than tripled between 1991 (76 million prescriptions) and 2012 (259 million prescriptions). The amount of opioid prescribed in the United States has been estimated to be the equivalent of 96 mg of morphine per person in 1997, 700 mg per person in 2007, 782 mg of morphine per person in 2010 (the peak), and decreasing gradually to 640 mg of morphine per person in 2015. Despite the decline between 2010 and 2015, the amount in 2015 is still three times as high as the 1999 number. In addition, there is a great deal of variability across the country, with significantly higher rates in counties with a larger percentage of non-Hispanic whites, higher rates of unemployment and Medicaid enrollment, higher prevalence of diabetes and arthritis, and micropolitan (10,000 to 50,000 population) status.

About 80% of individuals prescribed opioids receive less than 100 mg morphine equivalent per day from a single practitioner with another 10% receiving more than 100 mg morphine equivalents per day from a single provider. The remaining 10% of opioid prescription recipients tend to get higher doses (greater than 100 mg morphine equivalents per day) from multiple prescribers. c

c References 60, 98, 99, 229, 233, 320, 321, 340.

Since 2004, overall rates of fatal overdose have continued to rise, reaching a high of 52,404 in 2015. Of this total, 33,091 fatalities were associated with any opioid, 63% of the total drug overdose deaths in 2015. This represented a 6.5% increase from the age-adjusted rate of 13.8 per 100,000 in 2013 to 16.3 per 100,000 in 2015 ( Fig. 54.1 ). Rates of fatal overdose among non-Hispanic white persons has increased from 6.2 per 100,000 in 1999 to 21.1 per 100,000 in 2015. Increases for Hispanic and non-Hispanic black individuals have been much less dramatic. From 2010 to 2015, 30 states and the District of Columbia saw increased fatality rates.

Fig. 54.1, Age-adjusted rate of drug overdose deaths (A) and drug overdose deaths involving opioids (B)—United States, 2002–2015. Age-adjusted death rates were calculated by applying age-specific death rates to the 2002 US standard population age distribution. Drug overdose deaths involving opioids are identified using the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) underlying cause-of-death codes X40–X44, X60–X64, X85, and Y10–Y14 with a multiple cause code of T40.0, T40.1, T40.2, T40.3, T40.4, or T40.6. Approximately one-fifth of drug overdose deaths lack information on the specific drugs involved. Some of these deaths might involve opioids.

The misuse of prescription opioids has continued to be a major public health problem, with an estimated 91.8 million persons 18 years of age or older (97.5 million including those 12 years of age or older) having used an opioid and 11.5 million (12.5 million including those 12 years of age or older; 4.7% of the US population over the age of 12) reporting nonmedical use in the past year. Of these, 59.9% reported using them without a prescription, with 40.8% reporting that they had obtained them from friends or relatives for free. Of those with a prescription, 22.2% used them in greater amounts than directed on their prescription, 14.6% used them more often than directed, and 13.1% used them longer than directed. Rates were higher in adults who were unemployed, uninsured, and with other behavioral health problems. Although troublingly high, these numbers represent a decrease from a high of 12.65 million in 2006, and a slight increase from 2014, which had seen the lowest number since 2002. Despite this, there were 22,598 fatalities involving opioid analgesics in 2015, up from a previous peak of 16,917 in 2011 ,and following declines in 2012 and 2013. The correlation between increased opioid prescribing and fatal overdose continued to be especially prevalent in rural areas including New England, Appalachia, and the southwest ( Fig. 54.2 ). d

d References 97, 98, 420, 459, 460, 562.

Fig. 54.2, Drug overdose deaths involving opioids by type of opioid—United States, 2000–2014. Age-adjusted death rates were calculated by applying age-specific death rates to the 2000 US standard population age distribution. Drug overdose deaths involving opioids are identified using the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) underlying cause-of-death codes X40–X44, X60–X64, X85, and Y10–Y14 with a multiple cause code of T40.0, T40.1, T40.2, T40.3, T40.4, or T40.6.

Another study of substance abuse treatment–seeking individuals found that, of individuals who had developed an opioid use disorder and reported that they were first exposed to an opioid through a legitimate opioid prescription, approximately 95% reported that they had used another psychoactive substance (including alcohol, 93%; nicotine, 90%; and marijuana, 87%) prior to or coincident with the initial opioid use. In addition, 78% had used licit or illicit stimulants and 60% had used benzodiazepines with respondents reporting an average of four to five different substances used prior to the initial opioid.

Unfortunately, the relative leveling off in prescription opioid misuse has been accompanied by a steady increase in the number of individuals reporting the use of heroin, with 914,000 reporting past year use in 2014 (approximately 0.3% of the population at or above the age of 12), up from a low of 310,000 in 2003 and a significant jump from 680,000 in 2013. Similar increases were found in the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), with 0.33% of the adult population having used heroin in 2001 to 2002 and 1.6% in 2012 to 2013. The current heroin user tends to be slightly older, less urban, more often white, with a more even gender distribution than the typical heroin user of previous decades. The majority of the newer heroin users report being introduced to opioids through medical and nonmedical use of prescription opioids. This increase in heroin use has been accompanied by a fivefold increase in heroin-related fatal overdoses between 2000 (1842 deaths) and 2014 (10,574 deaths) ( Figs. 54.3 and 54.4 ) and an increase of 20.6% just between 2014 and 2015 (12,989 deaths).

Fig. 54.3, Rates of past-year heroin abuse or dependence and heroin-related overdose deaths—United States, 2002–2013. Heroin-related overdose deaths increased by 286% from 2002 to 2013.

Fig. 54.4, Number of overdose deaths from prescription and illicit opioids in the United States, 1999–2015. Although it is sometimes difficult to distinguish illicitly manufactured versus pharmaceutical fentanyl, the numbers generally represent illicitly manufactured fentanyl, fentanyl analogues, and other nonpharmaceutical opioids as well as heroin.

Compared to other causes of unintentional injury deaths, poisoning has continued to increase over the past 15 years. In 2014, unintentional poisoning ranked in the top 10 causes of injury death in all age groups beginning at age 9, and was the leading cause for the age groups between 25- and 65-years-old. Further evidence of the huge impact of overdose (primarily opioid but alcohol and other substances as well) comes from an analysis of all-cause mortality in the United States from 1993 to 2013. The study found that, despite decades of steady decreases in mortality rates (by about 2% per year between 1978 and 1998) for all other age and demographic groups, there was an increase in overall mortality rates (by about 0.5% per year) in white non-Hispanics between 45- and 55-years-old. The majority of this increase was due to poisoning, although deaths from suicide and chronic liver disease also increased. This effect was sufficiently significant to lower the life expectancy for this age group of white non-Hispanics by 0.1 years between 2013 and 2014. A similar rise in mortality rates was not seen in other industrialized countries, including France, Germany, the United Kingdom, Canada, Australia, and Sweden.

Another indication of the growth of the problem related to poisoning fatalities can be seen in the comparison to fatalities related to car crashes. Once the leading cause of accidental fatalities, automobile crash fatalities have gradually declined since the 1970s. As overdose fatalities gradually increased, the numbers of related deaths intersected in 2008, at approximately 38,000 deaths, and have continued to move in opposite directions ever since ( Fig. 54.5 ). Complicating these statistics, however, a six-state study of fatally injured car drivers found that the prevalence of prescription opioids detected in postmortem toxicological testing increased from 1% in 1995 to 7.2% in 2015.

Fig. 54.5, Motor vehicle traffic and drug poisoning death rates—United States, 1980–2008. “Poisoning” also includes intentional poisonings (suicide).

Several specific situations deserve special attention. Overdose involving methadone has had a unique epidemiology, which has been fairly closely tied to the use of methadone for the treatment of pain. Methadone-related overdose deaths increased 22.1% per year between 2002 and 2006, and then declined approximately 6.5% per year between 2007 and 2014 (39% total decline). The percentage of drug overdose deaths involving methadone also decreased from 12% in 2010 to 6% in 2015. This closely paralleled the amount of methadone prescribed, which increased by 25.1% per year between 2002 and 2006 and declined 3.2% per year from 2006 to 2013 ( Figs. 54.2 and 54.6 ) . These rates peaked between 2005 and 2007 in all age groups younger than 55 years, but continued to increase in 55- to 64-year-olds. The declines were seen in all racial/ethnic groups and in males and females, although the rate of decline was slower among females. The decline was also seen despite the fact that more than 100,000 individuals initiated methadone for the treatment of opioid use disorder during that period.

Fig. 54.6, Rates of methadone-involved overdose deaths, methadone distribution, and methadone diversion reports—United States, 2002–2014. The rates shown are for the number of methadone-involved overdose deaths per 100,000 population, number of methadone diversion reports per 100,000 population, and number of grams of methadone distributed per 100 population.

Another unique situation involves illicitly manufactured fentanyl (IMF), also known as nonpharmaceutical fentanyl (NPF), or its analogs (acetyl fentanyl, alpha-methylfentanyl, carfentanil, 3-methylfentanyl, and others) sold as heroin (often branded as more pure heroin under such names as China White, Tango & Cash, and TNT). Earlier epidemics were generally constrained to smaller areas starting in the late 1970s, with an outbreak (primarily in California) with at least 112 related deaths. Over the next decade, sporadic clusters of fentanyl-associated fatalities were encountered in the mid-Atlantic region, with 16 deaths associated with the use of heroin contaminated with 3-methylfentanyl, in Pittsburgh 1986–1988 and 20 deaths associated with illicit fentanyl use in Baltimore in 1992. From 2005 to 2007, a larger outbreak (more than 1000 fatalities) occurred over multiple metropolitan areas (Baltimore, Camden, Chicago, Detroit, Philadelphia, St. Louis, and Wilmington) as well as suburban and rural areas of Delaware, Illinois, Kentucky, Maine, Maryland, Massachusetts, Michigan, New Jersey, New Hampshire, Ohio, Pennsylvania, and Virginia. When the source of that NPF was traced to a single lab in Mexico, which was dismantled, the outbreak came to an end.

The largest epidemic of NPF began in late 2013 in Canada and various parts of the United States including the Midwest, Southeast, mid-Atlantic, and New England, and other parts of the world including Eastern Europe and Scandinavia, e

e References 22, 113, 220, 334, 608, 668.

with significant increases in law enforcement confiscation of illicit fentanyl and related overdose fatalities f

f References 226, 276, 293, 484, 530, 608.

( Figs. 54.4 and 54.7 ) . Between 2014 and 2015, the death rate from synthetic opioids other than methadone (primarily fentanyl and its analogues) increased 72%. Unlike the previous epidemics, this one saw both fentanyl-laced heroin as well as counterfeit pharmaceutical pills (oxycodone, Xanax, Norco), g

g References 222, 223, 265, 569, 631, 678.

which contained NPF, often combined with a benzodiazepine, thereby increasing the potential lethality. Also new in this epidemic was the presence of carfentanil, an analogue 100 times more potent than fentanyl, which is used by large animal veterinarians. Concerns over the relative danger of these analogues have lead the US Drug Enforcement Administration (DEA) and many local jurisdictions to implement enhanced precautions for first responders interacting with individuals and crime scenes that may have been exposed to fentanyl analogues. In addition, this epidemic also saw the emergence of some novel opioids (AH-7921, MT-45, U-47700, and others) related to overdose fatalities in Canada, the United States, and Europe. h

h References 17, 34, 209, 225, 236, 359, 374, 463, 472, 561.

Fig. 54.7, Trends in number of drug overdose deaths involving synthetic opioids other than methadone, ∗ ∗ Synthetic opioid–involved (other than methadone) overdose deaths are deaths with an ICD-10, underlying cause-of-death of X40–44 (unintentional), X60–64 (suicide), X85 (homicide), or Y10–Y14 (undetermined intent) and a multiple cause-of-death of T40.4 (poisoning by narcotics and psychodysleptics [hallucinogens]: other synthetic narcotics). number of reported fentanyl submissions to law enforcement, † † Drug products obtained by law enforcement that tested positive for fentanyl are referred to as fentanyl submissions. Reports were supplied by the Drug Enforcement Administration’s National Forensic Laboratory Information System and downloaded July 1, 2016. and rate of fentanyl prescriptions § § National estimates supplied by IMS National Prescription Audit and include short- and long-acting fentanyl prescriptions. —United States, 2010–2014.

Nonfatal Overdose

Numbers for nonfatal overdose are less reliable than those for fatalities. There is much more room for variability in coding. A heroin overdose may be coded as “respiratory depression” or “altered mental status” and “Heroin Use Disorder” and not detected as a “poisoning.” In addition, various databases may capture different components of the issue. The Drug Abuse Warning Network (DAWN), most recently overseen by SAMHSA and the Center for Behavioral Health Statistics and Quality (CBHSQ), collected information on substance-related emergency department (ED) admissions in various areas of the country from 1972, until it was discontinued in 2011. Based on their data, it can sometimes be difficult to ascertain how many visits were actually for an overdose, as they also captured “adverse reaction to a pharmaceutical” and other substance-related admissions. To add to the difficulty, the United States continued to use the ICD-9 for medical coding until October, 2015, when much of the rest of the world had been using ICD-10 for more than a decade.

Despite the limitations in data, unintentional poisoning ranks in the top 10 causes of nonfatal injury treated in an ED for ages 1–4 and all age groups over 15 years old in the United States. Earlier studies showed hospitalizations for heroin overdose increasing 69% between 1993 and 2006. In 2011, DAWN reported roughly 5 million total ED visits related to drugs (1626 visits/100,000 population) a 100% increase since 2004. Overall, medical emergencies related to nonmedical use of pharmaceuticals increased 132% in the period from 2004 to 2011 (1.2 million visits in 2011), with prescription opiate/opioid involvement rising 183% (approximately 348,000 visits in 2011). The specific drugs with the highest number of visits were oxycodone, hydrocodone, and methadone with 151,218, 82,480, and 66,870 visits, respectively. There were 258,482 visits related to heroin use, which constituted 83 visits/100,000 population, not a statistically significant rise from 2004. Studies using other data sources similarly found ED visits involving misuse/abuse of prescription opioids increased 153% between 2004 and 2011, and hospitalizations related to overdose on prescription opioid use increased as well.

Another large, national study used data from the Nationwide Inpatient Sample (NIS) Healthcare Cost Utilization Project (HCUP) in the Agency for Healthcare Research and Quality collected from 1993 to 2009. As these data only captured individuals who were admitted to the hospital, it is limited, in that it looks at only more severe cases. The authors looked at heroin-related overdoses (HODs) and prescription opioid–related overdoses (PODs) and found that PODs increased steadily over the study period. The highest increases were in white, middle-aged women, with rates of increase for whites more than double that for African Americans or Hispanics. Overall HODs increased from 1993 to 1999 but leveled off and decreased until they began to increase again around 2007. Since 1995, rates for African Americans and Hispanics decreased (with the exception of an uptick in 2009), whereas those for whites showed a gradual increase through most of the period with a significant increase beginning in 2007, surpassing African Americans for the first time in 2008. Yet another study using HCUP data found that the rate of hospital stays involving opioid overuse (which included treatment for opioid abuse and dependence in addition to overdose) among adults increased more than 150% between 1993 and 2012 when there were 709,500 opioid-related hospital stays representing a rate of 295.6 stays per 100,000 population.

Looking at hospitalizations related to prescription opioids versus heroin using the NIS between 2000 and 2014, one study found significant geographic variation with PODs highest in the South and lowest in the Northeast and heroin-related overdoses highest in the Northeast and Midwest. Between 2012 and 2014, POD hospitalizations decreased in all areas of the United States except New England, whereas heroin-related hospitalizations increased in all areas of the country.

The preponderance of ED substance-related visits for children 5 years of age or younger involve accidental ingestions (which were recorded separately by DAWN). On average, more than 3200 children younger than 5 years of age are seen in the ED each year due to accidental opioid overdose. In 2011, of a total of 113,634 visits, over 77,000 involved children in this age range. Pain relievers (including acetaminophen, aspirin, and nonsteroidal anti-inflammatories) were the most common class of drugs involved in accidental ingestion among children 5 years of age or younger, with 6.7% of visits related to opioids. Although overall visits for accidental ingestion by patients 5 years of age or younger were stable from 2004 to 2011, visits involving narcotic pain relievers increased 225% (from 1596 to 5187 per year).

Pathophysiology

As mentioned above, the hallmark of opioid overdose is constricted (“pinpoint”) pupils, altered level of consciousness, and respiratory depression. This triad has been found to have 92% sensitivity and 76% specificity for heroin overdose. The amount of a given opioid needed to produce these effects is dependent on tolerance, which is a complex interaction of single-cell and neuronal network-level alterations. Tolerance develops at different rates and to varied degrees to various clinical aspects of opioid intoxication, with that to respiratory depression developing slower and to a less complete extent than some other effects. Tolerance is also affected by nonpharmacological factors such as the environment in which the opioid is used. Animal studies and epidemiological data support the finding that fatal overdose is more likely when the opioid is used in a novel setting as compared to a setting in which the drug has previously been used.

The pupillary constriction, miosis, appears to be caused by parasympathetic excitation. This may not occur with overdose on meperidine, propoxyphene, and pentazocine. Altered mental status may vary from mild sedation to stupor and coma. Gag reflex may also be suppressed. In addition to the typical, centrally mediated respiratory depression (discussed below), severe opioid overdose can also produce noncardiogenic pulmonary edema (NCPE) and bronchospasm. This typically presents with frothy, pink secretions and rales. In nontolerant individuals, opioid use can also cause nausea and vomiting. Rarely, overdose may also produce a centrally mediated muscle rigidity of the chest and abdominal wall. Rhabdomyolysis may be seen as a result of prolonged motionlessness and compression of muscle, typically in the context of coma. Hypotension, acidosis and dehydration may further increase the risk of this. There are also reports of rhabdomyolysis occurring in the absence of coma suggesting that it may, rarely, be a direct toxic effect or allergic reaction to heroin or an adulterant.

Respiratory depression (decreased respiratory rate and effort, hypoxia, and hypercarbia) is the main cause of death due to opioid overdose, although cardiac arrest and arrhythmia induced by anoxia can also occur. The mechanisms through which μ-opioid agonists suppress respiration are complex and not yet fully understood in humans. Opioid receptors are found in various central and peripheral areas including the pons and ventrolateral medulla in the brainstem; the insula, thalamus, and anterior cingulate cortex higher in the brain; the carotid bodies and vagus nerve; as well as in the epithelial, submucosal, and muscular layers of the airways themselves. It appears that the strongest effect on respiration by μ-opioid receptor agonists is mediated through a decrease in ventilatory response to CO 2 through action on the respiratory centers in the brainstem. This results in a disruption of the respiratory pattern with prolongation of inspiration and, at higher opioid doses, changes in tidal volume.

The mechanism by which benzodiazepines might worsen respiratory depression associated with opioid use is not well described. Benzodiazepines bind to the γ2 subunit of the γ-aminobutyric acid A receptor (GABA A ) and have no activity at the μ-opioid receptor. There is some evidence that the GABAergic and opioidergic systems are coexpressed in several areas of the rat brain and that there is anatomical overlap in receptors in various brain regions. It has been postulated that benzodiazepines may also have effects on signal transduction and second messenger systems involved with μ-opioid receptor regulation. It is generally felt that any potential effect of benzodiazepines on the opioid system is not primarily pharmacodynamic in nature.

Certain opioids have specific physiological effects that may contribute to the clinical picture in overdoses. Methadone and the related l -alpha-acetylmethadol (LAAM; not currently commercially available) are fairly unique among opioids in their potential to cause QTc prolongation and resultant torsade de pointes. The true prevalence of this QTc prolongation and torsade de pointes associated with methadone and the implication for methadone-related overdose are not clear.

Seizures have rarely been reported with morphine, fentanyl, sufentanil, alfentanil, and meperidine. Most of these have been reported only in conjunction with general anesthesia and were not confirmed with simultaneous electroencephalography (EEG) readings. It is possible that, what appears to be seizure activity may actually be rigidity, myoclonic jerks, and other nonepileptic movements. i

i References 63, 89, 392, 462, 549, 570, 644.

Meperidine, however, has been associated with seizures, primarily due to the metabolite normeperidine. The risk of this is increased in the presence of renal impairment (which reduces the clearance of normeperidine) or concomitant use of medications that induce microsomal liver enzymes (through increased conversion of meperidine to normeperidine). j

j References 294, 318, 338, 361, 437, 636.

Fentanyl and its analogs (e.g., acetylfentanyl, alfentanil, sufentanil) appear to be able to cause skeletal muscle rigidity (especially chest wall, sometimes referred to as “wooden chest”). It appears to be more common with rapid intravenous administration, although it is not dose dependent. Onset may be rapid and produce an inability to ventilate. It may be related to noradrenergic activation of spinal tracts associated with the locus coeruleus. Because the chest wall rigidity is not discernible after death, the role of this phenomenon in fatal overdose is not clear.

Buprenorphine, a partial μ-opioid agonist that has been associated with a significantly lower risk of overdose, needs special consideration when discussing opioid-induced respiratory effects. At lower doses, buprenorphine produces CNS and respiratory depression in a dose-dependent manner. However, as the dose is increased, a plateau is reached at which these effects no longer continue to increase, making respiratory depression less likely than with other opioids. The ceiling effect on respiratory depression may be more evident in habitual users than in opioid naïve individuals, in whom respiratory depression may occur before reaching the ceiling. k

k References 155, 164, 311, 445, 720, 721.

Although the exact mechanism for this is not entirely clear, it appears that buprenorphine alone generally leads to mild decreases in PaO 2 with minimal increases in PaCO 2 as compared to the significant increase in PaCO 2 generally observed with full opioid agonists.

Increased respiratory depression associated with the combination of buprenorphine and benzodiazepines has been reported in the anesthesia and animal literature since the mid-1980s. This effect is generally found to be most pronounced for flunitrazepam, as compared to other benzodiazepines, with a relatively more pronounced effect on respiratory depression being observed when flunitrazepam is combined with buprenorphine as compared to some other opioids. In addition, buprenorphine’s main active N -dealkylated metabolite, norbuprenorphine, appears to produce considerably more potent respiratory depression than buprenorphine, and buprenorphine itself may act as a protectant against norbuprenorphine’s effects in rodents.

Risk Factors

Over the past several decades, multiple studies have attempted to determine risk factors for overdose, both nonfatal and fatal. l

l References 24, 171, 174, 175, 178, 179, 566, 731.

Although there is some variation over time and across different areas of the world, there is a fair amount of consistency in many of these factors ( Box 54.3 ).

Box 54.3
Risk Factors for Opioid Overdose

Substance Usage Factors

  • Prior nonfatal overdose

  • Recent reduction of tolerance

  • Concomitant use of opioids with CNS depressant substances

  • Use in a novel environment or alone

  • Higher or variable heroin purity

  • Intravenous use

Medical History Factors

  • Significant medical problems (especially hepatic and respiratory)

  • Significant mental health problems (especially depression and anxiety)

  • History of other substance use disorder

  • Prescription of opioids for pain (especially higher doses and longer-acting formulations)

Demographic Factors

  • Male

  • Middle-aged

  • Economically disadvantaged

  • Undereducated

  • Race (variable)

Risk Related to Substance Use Factors

A history of prior overdose m

m References 46, 59, 93, 139, 145, 174, 177, 178, 253, 254, 261, 375, 594, 619, 723.

is one of the strongest predictors of subsequent overdose. In addition, the risk of fatality appears to increase with each prior nonfatal overdose.

A history of opioid dependence with reduced tolerance following medical detoxification, n

n References 323, 422, 423, 450, 506, 541, 550, 584, 623, 637, 704, 707, 724, 731.

release from incarceration (especially in the first 1–2 weeks), o

o References 15, 49, 50, 53, 52, 54, 139, 196, 252, 253, 302, 350, 364, 384, 450, 464, 559, 586, 597, 617, 714.

cessation of treatment (especially in the first month), p

p References 136, 151, 152, 154, 176, 192, 197, 248, 300, 301.

or self-imposed abstinence is also a very strong predictor of overdose. One study found that a total of 36 of 276 patients died after discharge from residential substance abuse treatment during the 8-year study period. Two-thirds of these deaths were classified as opiate overdose deaths, with six of these occurring within the first 4 weeks following discharge from the program, yielding an unadjusted excess mortality of 15.7 (rate ratio) in this period. There was no significant association between time in treatment and mortality after discharge, and no baseline characteristics correlated significantly with elevated mortality shortly after discharge. Another study looking at more than 32,000 patients in California seeking methadone maintenance treatment over a 5-year period, found that the highest mortality risk occurred during the 2 weeks immediately following discontinuation of treatment, with a rate approximately 30 times that expected in the general population. Yet another study found that victims of fatal overdose had significantly lower concentrations of morphine in hair samples compared to active heroin users, although their morphine levels were not significantly different than those in former heroin users who had been abstaining for several months prior to the study.

Another highly correlated factor is the concomitant use of opioids with CNS depressant substances , especially alcohol, benzodiazepines, and barbiturates. q

q References 15, 33, 52, 53,116, 117, 118, 139, 149, 228, 305, 320, 322, 327, 329, 358, 422, 424, 490, 526, 565, 614, 653, 688, 690, 691, 692, 711, 731, 728, 734.

The relationship with alcohol use is not entirely straight forward, with some studies of fatalities showing higher alcohol levels correlated with lower morphine and 6-monoacetylmorphine (6-MAM) blood levels but others finding no significant correlation. One study actually found that complete abstinence from alcohol was a risk factor for opioid overdose fatality, possibly because individuals in that study who were not using alcohol tended to have higher rates of daily heroin use. Studies looking at blood levels of sedatives and opioids in fatal overdoses have not demonstrated a clear relationship.

Another important factor is use in a novel environment, r

r References 172, 178, 289, 313, 329, 600, 602, 603.

likely related to a decrease in tolerance, which appears to be classically conditioned through environmental cues. This has also been reported in non–substance-abusing individuals who use opioids. Related to this, being homeless and using alone in secluded areas, often abandoned buildings, increases the risk of overdose and fatal overdose. At least part of the increased risk can be explained by the lack of others to directly administer or seek medical assistance. Possibly related, several studies have shown that rushing the use of heroin, because of fear of being caught, may also raise the risk for overdose.

Higher or variable heroin purity or “cut” is another important risk factor. Darke et al. measured the range and average heroin purity over a 2-year period and found that both were independent predictors of fatal overdoses and accounted for approximately 40% of the total variance. Related to the purity of heroin, the cost of a gram of pure heroin (which, in the United States has decreased from $2690 in 1982 to $1237 in 1992, and to $552 in 2002 and $465 in 2012) has also been shown to be correlated with overdose, with a 2.9% increase in the number of hospitalizations for overdose for every $100 decrease in the per gram price. In addition, heroin adulterated/substituted with other opioids , especially fentanyl (or its analogs: acetyl fentanyl, alpha-methylfentanyl, carfentanil, 3-methylfentanyl, and others) also appears to raise the risk for nonfatal and fatal overdose. s

s References 4, 57, 111, 113, 220, 257, 336, 385, 561, 562, 712.

Some studies have shown a higher risk for overdose with intravenous route of administration of opioids, whereas others have not found a significant correlation.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here