“Who cares about that?”
This is what my supportive wife (who also happens to be a doctor) said when I told her I wanted to write about postmortem drug testing. I can understand where she is coming from. Before writing something, it is valid to ask if this is something that my audience is even interested in. This sets in motion a series of questions that go in an uncomfortable trajectory: “Who, even, is my audience?” and “Does anything I do make any difference?”
I have observed that I have several types of readers. Some of you are toxicologists, some of you are non-toxicologist physicians, some of you aren’t in science at all, and some of you are my mom.
It is valid to ask why you should care? What if I put it this way?
It is very important to know why people died.
Consider Zachary Taylor, the 14th US president. His death, long considered a mystery, was attributed first to rotten cherries, and then to arsenic poisoning from his political enemies. Now, we know that he probably died from bacteria that leaked from Washington’s sewers into the “fresh” water supply. It doesn’t matter much to Zachary Taylor now, but living people should know whether to avoid stone fruit or Whig Party members.1 Maybe you don’t care about antebellum US history, selfish reader. But, what if you had an upcoming trip to the Dominican Republic and you wanted to know whether to cancel after someone died at the same resort? You would want to know whether they were poisoned or died from natural causes. Thankfully, autopsies have gotten a lot better since the Zachary Taylor years.
I can explain a bit about the challenges of measuring drug levels after death.
Something to get out of the way: I will mention sticking needles into dead people’s eyeballs.
Who determines how you died? In the US, your death certificate tells us how and why you died. Your final, bespoke legal document contains mostly humdrum information like the date and time of your death but the real meat of the document- the parts most likely to be challenged – are the cause of death (e.g., “blunt force trauma to head”) and manner of death. There are only 5 possible manners in the US (natural, accident, suicide, homicide, and undetermined2) and every US death is classified as one of these. The certificate can be completed by any type of physician, and in most cases, it is completed by whichever doctor was present at the end of life. In the community, if no doctor is nearby, this role is usually served by the patient’s own doctor. In the hospital, the certificate would be completed by the attending physician or their proxy (usually the intern.)
Things become dicier when the death is unexpected or controversial. States have different rules but, in general, when the death involves a child, violence, poisoning or is otherwise unexpected, a death investigation is required. In the US, jurisdictions use one of two types of professionals to lead these investigations and make a conclusion on the cause and manner of death: Coroners or medical examiners (ME). Coroners and MEs perform very similar roles, yet have completely different credentials.
The ME is a licensed physician who completed training in pathology and subspecialty training in forensic pathology. A coroner is an elected or appointed public official. A coroner could be a doctor or an EMT or a person with no scientific background. Reader, you could be a coroner.
These are two very different types of people that have a very similar job. I can’t think of any other parallel examples. It’s almost like we made a rule that operators of passenger planes must either be licensed pilots with 1500 hours of logged flight time or, absent that, any person who likes airplanes.
To be clear, most coroners deliver great service. We simply do not have enough medical examiners in the US to handle the caseload of deaths that require investigations. Coroners tend to work in more rural areas, while medical examiners cluster in cities where there are more controversial deaths. In most cases, the conclusion of the death investigation is straightforward. When a case isn’t simple, a good coroner can have a pathologist conduct an autopsy.
The Role of the Toxicologist
I am not a medical examiner, a pathologist, or even a coroner. The clinical toxicologist can be invited to play a role in the death investigation because many deaths involve questions dealing with medications and other drugs.
Before we even get to the clinical toxicologist, a forensic toxicologist determines drug concentration from patient body fluid (usually blood.) Although both types of professionals are called toxicologists, forensic tox and clinical tox have different training. Forensic toxicologists have specialized Masters or Ph.D. level training. Clinical toxicologists are pharmacists or physicians with fellowship training in managing poisoned patients. (You can also refer to the physician clinical toxicologists as medical toxicologists.)
The clinical toxicologist is asked a key question: Did the decedent die from the drug or just die with the drug? The clinical toxicologist is in a unique position to answer this because we specialize in the effects of drugs on (mostly living) humans. In the clinical setting, the toxicologist spends their time with living people trying to keep them from dying from poisoning. Determining why a dead person is dead isn’t an entirely different skillset from keeping a living person alive. When we fail at our attempts to keep poisoned patients alive, we have a close-up look at the events that led to their death. The clinical toxicologist often, but not always, comes to the same conclusion as the medical examiner.3
How Much Drug is Lethal?
On the surface, it may seem straightforward to look at the dose of the drug that the decedent took to determine if they died from an overdose. In general, tiny doses don’t affect you and high doses kill you. For many drugs, we know the LD50 (the dose of drug predicted to kill 50% of the exposed) based on animal experiments and human case reports.The generic dose-response curve shows that that low doses are harmless, big doses are deadly, and in the middle, there is a range of doses with steadily increasing physiologic response (Figure 1). We also see that we are all a little different. A dose that is harmless for one person could be an overdose for another. In one group of patients getting daily methadone, the average therapeutic trough (the steady-state range after initial peak) blood level was 222 ng/mL but ranged from 100 to 254.4 No two of us are identical in our size, body fat, age, tolerance, medical conditions, and other medications that we take. (Your mother was right: you are different and special.)
Drug Levels Change after Death
In the postmortem exam, we don’t have a drug dose, we have a drug level. We may not have a reliable history of how many pills the patient took. We only have laboratory evidence of the drug left in their bodies.
Interpreting this drug level is not straightforward because we don’t know what to compare it to. As it turns out, we can’t automatically compare drug levels in living people to those in dead people. Drugs levels in a body change after death, and this wasn’t recognized until relatively recently. Before the 1970s, it was assumed that drug blood levels before death were similar to drug levels after death and that levels drawn from different locations from the same body would also be similar. After all, in life, the concentration of a drug in your liver blood is the same as in a vein in your leg.
In 1975, investigators reported a series of deaths from digoxin poisoning where different drug concentrations in blood were obtained from three different sites.6 They suspected that the drug had moved into different compartments after death. By 1990, this phenomenon was well-known and recognized as a “toxicological nightmare.”7
For you non-science folks, the main science concepts you’ll need to understand for the next few paragraphs are diffusion and the closely related term, concentration gradient. Diffusion describes how things in solution tend to move from high concentration to low concentration. If you drop sugar in coffee, it diffuses until sweetness is evenly distributed throughout every drop of your delicious life saving, coffee. The concentration gradient describes the difference in the amount of sugar from the high sugar areas to the low sugar area. We would say that sugar diffuses throughout the water, moving down its concentration gradient. The diffusion of drugs, down concentration gradients explains much of what happens after death.
Postmortem Blood Levels
Several postmortem processes explain changes in drug concentration after death. First, drugs and ions may move across no-longer-impermeable barriers after death. Take potassium: During life, cell membranes impervious to the ion divide the body into distinct compartments. Ion-powered pumps use energy from the oxygen you breathe and the food you eat to drive potassium from the blood into cells. As a result, the concentration of potassium in our bloodstream is between 3 and 5 mmol/L, while inside cells it is 140 mmol/L.8 This huge concentration gradient can only be maintained by expending energy and maintaining sealed cellular barriers. After death, metabolism stops, cell walls break down, and the integrity of these barriers is lost. Ions and drugs shift down their concentration gradient, into all the spaces they were kept out of in life. Potassium which was once sequestered in cells leaks out into the blood.
As a result, potassium blood levels rise after death and postmortem blood potassium concentration is useless as a predictor of antemortem blood potassium. Even if the patient died from potassium overdose, the ME can’t tell from the blood level alone.9
In the same way, drug molecules can shift from the lungs, heart, stomach, or bladder into adjacent tissue and vessels.10,11 Drugs also move into nearby vascular compartments after death. Drugs in the intestine may go into thoracic vessels and drug in the liver may diffuse into the inferior vena cava.12
In life, your blood is an even, liquid mixture of proteins, cells, ions, and whatever drugs you happened to add to that soup. Things tend to be evenly distributed in your blood. This is good because you want your blood to flow easily to and from your hungry cells. Following death, blood clots, and the formerly consistent liquid becomes lumpy and bumpy, causing constituent components to distribute unevenly in the body. The formerly homogenous liquid can separate into cellular and liquid components. Drugs that concentrate in red cells during life may be undersampled from the liquid.12
In life, many drugs stick to proteins like albumin or glycoproteins. These proteins – and the drugs stuck to them – stay in your bloodstream because they are too large to pass into cells. In death, the concentration of proteins in the blood vessels decreases. Proteins are broken down by enzymes or bacteria and diffuse through broken cell membranes, down their concentration gradients and into the spaces around blood vessels. This results in decreased protein available to bind drugs, resulting in less drug stuck to protein and more “free” drug.13 Cyclic antidepressants and beta-blockers are examples of drugs bound to acid glycoproteins that break free after death.14
Although in most cases, metabolism is an energy-dependent metabolic process that stops at death, there are exceptions. Some drugs continue to be metabolized after death. Cocaine is slowly metabolized by residual plasma esterases into ecgonine methyl ester.15 As a result, cocaine is degraded rapidly in the blood after death. After death, we can infer the presence of cocaine from its metabolites, not necessarily cocaine itself.
Spoiler: Your body is going to decay. In a healthy person, blood is free of bacteria. In the postmortem period, gastrointestinal bacteria cross into the bloodstream where they can degrade and metabolize substances. Glucose can be fermented by bacteria and fungi into ethanol. The formation of alcohol from sugar after death can falsely imply that the alcohol was there before death, and death could mistakenly be attributed to intoxication. Medical investigators know this, so the preferred sampling site for determining antemortem alcohol concentration is vitreous fluid.16 Non-medical people, this is eyeball fluid. (I told you we would be discussing needles in dead eyeballs.) Vitreous fluid has a similar alcohol concentration to antemortem blood but doesn’t exhibit much postmortem fermentation.
Which Drugs are Affected by Postmortem Redistribution?
The net result of all these above processes, called “postmortem redistribution”, describes the movement of drugs after death. Some drugs redistribute a lot and others don’t. It’s helpful for death investigators to know how to interpret postmortem drug levels. When drugs don’t shift around, we can compare postmortem drug levels to drug levels in living people.
Ideally, we would have a collection of patients with blood obtained right before and after death and we could compare the pre- and postmortem levels to see if there was postmortem redistribution.17 Tox & Hound was going to put Tom Cruise on this but we had a falling out over antidepressants and now we’re back at the drawing board. . . Using this type of data, authors have assembled lists of drugs that are known to (and not to) demonstrate postmortem redistribution (Tables 1 and 2).
How can we predict when postmortem redistribution will occur?
Unfortunately, these data aren’t available for many drugs. We can make some predictions about postmortem redistribution based on other properties of drugs. Selecting the correct blood sampling site can minimize chances for postmortem redistribution. When possible, blood should be drawn from peripheral or femoral vessels, which are isolated from central organs and are less prone to diffusion from central vessels.
Volume of Distribution
Drugs with large volumes of distribution tend to demonstrate more postmortem distribution. Volume of distribution is an indicator of how much the drug tends to accumulate outside of the bloodstream, so drugs that concentrate outside of blood in life may diffuse back into the blood in death, resulting in increased postmortem levels. Volume of distribution can be related to lipophilicity – fat solubility. Lipophilic drugs, which are pulled to fat during life, may also leak back into blood vessels after death. Basic drugs are also more likely to be pulled into the blood after death when blood becomes more acidic.
Different drug concentrations in different postmortem compartments are a strong clue to postmortem redistribution. The ratio of postmortem drug concentration in cardiac blood compared to peripheral blood (C/P) was the first commonly used predictor of postmortem redistribution. The idea is that heart blood was more susceptible to redistribution, so if plasma and heart levels are similar there is probably no postmortem redistribution. In theory, a high C/P ratio indicates likely redistribution, because the concentration of drug in the blood and periphery should be almost identical, and an excess of a drug in the heart suggests it moved there from the liver or surrounding tissues after death. The measure is imperfect because some drugs that are known not to undergo redistribution have high C/P. CPR may also push drugs out of the heart postmortem, resulting in C:P <1.12
Another calculation may be more useful than C/P. Liver-to-Plasma (L/P) ratio measures the postmortem concentration in the liver compared to the peripheral blood. An L/P of less than 5 L/kg suggests unlikely postmortem redistribution, while an L/P of > 20-30 L/kg predicts the phenomenon.18 In other words, L/P can be plugged into an equation to determine a mathematical factor (called Ft) which would allow the medical examiner to use postmortem peripheral blood to predict antemortem blood concentration.19 The author cautions that this ratio may not hold up after decomposition or long postmortem delay (>48 hr).
How Can Death Investigators Use Blood Levels?
The best option for the death investigator is to compare postmortem levels from the decedent with postmortem levels from published cases. For methadone, postmortem levels from patients who died from natural causes averaged 344 ng/mL (range 62 to 1,090), while the mean level in patients dying from overdose was 559 ng/mL (range 114 to 1,939). Higher methadone levels (>1,090 ng/mL) and lower methadone levels (<114 ng/mL) suggest strongly for and against methadone poisoning, respectively. The vast middle range is not conclusive. What is a medical examiner or coroner supposed to do with a level of, say, 400 ng/mL? Did this patient die from methadone or merely die with methadone?
Caution should be taken when the decedent’s postmortem concentrations are compared to antemortem reference levels. (E.g. comparing postmortem drug concentrations with published “therapeutic” drug levels.) This approach should only be used when postmortem reference levels aren’t available and postmortem redistribution is known to be a non-factor.
The answer often lies outside the laboratory. Toxicology is a key part – but not the only part – of this assessment. The investigator doesn’t interpret the postmortem concentration in a vacuum. Even though the “patient” can’t give a history, there are historical clues to put the drug levels into context for the death investigator. What were the circumstances of the death? A modest postmortem fentanyl blood level might not definitively provide an answer to the cause of death. However, if that fentanyl blood level was from a decedent found with an empty syringe containing drops of fentanyl, the cause is clear. Other clues include suicide notes that point to intentional overdose. Medical records and pill bottles may indicate new prescriptions or medications not being taken properly. If a new prescription was filled the day before, and all pills are missing, an overdose is more likely. Sometimes, even after blood levels and history, there is still no cut-and-dry answer. If several drugs may share responsibility for death, the medical examiner can attribute the cause of death to combined drug poisoning. If the intent or reason for poisoning isn’t clear, the manner of death is “undetermined.”
A Tale From Entropy
The Second Law of Thermodynamics says that the entropy in a closed system must increase.20 Life is a constant fight against entropy. The air we breathe and the food we ingest are used to maintain a structure and an order. Ions and molecules and drugs are sequestered into a million tiny places. After the last breath is drawn, that order begins to disintegrate. This is a literal disintegration – a loss of integrity, of connection. Proteins break down, the blood separates into its components, bacteria grow. In this new disorder, once-confined ions and molecules and drugs shift and move, seeking out new spaces following the simple rule of diffusion.
The death investigator is a storyteller, a journalist who must tell a true, objective story. From this growing entropy, the death investigator – coroner or medical examiner – is given incomplete clues and charged to draw a conclusion about the cause and manner of death. Drug concentrations should be evaluated in the proper context. Truly, there are two contexts. There is a context for the person and a context for the drug; a context for inside the body and a context for outside of the body.
The death investigator asks:
What was inside the body?
Is this a high or low drug level? How does it compare to published postmortem cases? Accounting for postmortem redistribution, how does it compare with antemortem therapeutic levels?
What was outside the body?
What evidence was in their medical record, around their house, next to their hand?
From this disintegration, this entropy, a story takes shape.
The dead cannot tell their stories. Voltaire said, “To the living we owe respect, but to the dead, we owe only the truth.”Tombstones by Kevin Wenning
- 1.Actually, Zachary Taylor was a Whig, so he would never have been killed by the Whigs, but the joke is funnier this way. Presented at the: ; 2019.
- 2.“Death by Chocolate” not a legitimate cause or manner of death. Presented at the: ; 2019.
- 3.Manini A, Nelson L, Olsen D, Vlahov D, Hoffman R. Medical examiner and medical toxicologist agreement on cause of death. Forensic Sci Int. 2011;206(1-3):71-76. doi:10.1016/j.forsciint.2010.06.021
- 4.Jiang H, Hillhouse M, Du J, et al. Dose, Plasma Level, and Treatment Outcome Among Methadone Patients in Shanghai, China. Neurosci Bull. 2016;32(6):538-544. doi:10.1007/s12264-016-0059-0
- 5.Ferner R. Post-mortem clinical pharmacology. Br J Clin Pharmacol. 2008;66(4):430-443. doi:10.1111/j.1365-2125.2008.03231.x
- 6.Holt D, Benstead J. Postmortem assay of digoxin by radioimmunoassay. J Clin Pathol. 1975;28(6):483-486. doi:10.1136/jcp.28.6.483
- 7.Pounder D, Jones G. Post-mortem drug redistribution–a toxicological nightmare. Forensic Sci Int. 1990;45(3):253-263. doi:10.1016/0379-0738(90)90182-x
- 8.Mendelsohn F, Mackie C, Mee M. Measurement of intracellular potassium in dispersed adrenal cortical cells. J Steroid Biochem. 1975;6(3-4):377-382. doi:10.1016/0022-4731(75)90160-0
- 9.Palmiere C, Scarpelli M, Varlet V, Baumann P, Michaud K, Augsburger M. Fatal intravenous injection of potassium: Is postmortem biochemistry useful for the diagnosis? Forensic Sci Int. 2017;274:27-32. doi:10.1016/j.forsciint.2016.11.035
- 10.Pounder D, Smith D. Postmortem diffusion of alcohol from the stomach. Am J Forensic Med Pathol. 1995;16(2):89-96. doi:10.1097/00000433-199506000-00001
- 11.Moriya F, Hashimoto Y. Postmortem diffusion of drugs from the bladder into femoral venous blood. Forensic Sci Int. 2001;123(2-3):248-253. doi:10.1016/s0379-0738(01)00530-8
- 12.Pélissier-Alicot A, Gaulier J, Champsaur P, Marquet P. Mechanisms underlying postmortem redistribution of drugs: a review. J Anal Toxicol. 2003;27(8):533-544. doi:10.1093/jat/27.8.533
- 13.Oehmichen M, Gencic M. Postmortal diffusion of plasma albumin in rat brain. Z Rechtsmed. 1980;84(2):113-123. doi:10.1007/bf02114580
- 14.Konikova A, Vinarskaya A, Nikulin V, Pogossova A, Petukhova L. Protein degradation to low-molecular compounds after death and during reanimation. Virchows Arch B Cell Pathol. 1975;18(4):347-355. doi:10.1007/bf02889261
- 15.Isenschmid D, Levine B, Caplan Y. The role of ecgonine methyl ester in the interpretation of cocaine concentrations in postmortem blood. J Anal Toxicol. 1992;16(5):319-324. doi:10.1093/jat/16.5.319
- 16.Kugelberg F, Jones A. Interpreting results of ethanol analysis in postmortem specimens: a review of the literature. Forensic Sci Int. 2007;165(1):10-29. doi:10.1016/j.forsciint.2006.05.004
- 17.Leikin J, Watson W. Post-mortem toxicology: what the dead can and cannot tell us. J Toxicol Clin Toxicol. 2003;41(1):47-56. doi:10.1081/clt-120018270
- 18.McIntyre I, Sherrard J, Lucas J. Postmortem carisoprodol and meprobamate concentrations in blood and liver: lack of significant redistribution. J Anal Toxicol. 2012;36(3):177-181. doi:10.1093/jat/bks011
- 19.McIntyre I. Analytical data supporting the “theoretical” postmortem redistribution factor (Ft ): a new model to evaluate postmortem redistribution. Forensic Sci Res. 2016;1(1):33-37. doi:10.1080/20961790.2016.1253255
- 20.The first rule of Fight Club is “Don’t talk about Fight Club.” Presented at the: ; 2019.