Tox and Hound – Love and Half-Life*

by Sarah Shafer

*Not the video game

I’ve gotten some version of this call many times in my budding career as a medical toxicologist: “Hello, we have a something-year-old patient in the ER who overdosed on ______ (fill in the blank) medication. How long do we need to watch them before we can medically clear them? When I read about the medication, it has a half-life of X hours, so they should be good after that, right?”

A long time ago, I had a beautiful relationship with half-life. Half-life meant everything to me. If someone overdosed, half-life would tell me when they were out of the danger zone. Half-life would tell me that everything would be okay in 6-8 hours. With half-life, I was armed with all the information I needed to anticipate the clinical course of an overdose. Half-life promised me the world.

It was a simpler world back then, during the larval stages of my medical training. Half-life and I were in love. All heart-attack related chest pain would present with crushing chest pain that radiates to the left arm. All appendicitis would present with right lower quadrant abdominal pain. My standardized patients told me exactly what I needed to hear during the history portion of the exam. I looked forward to a career with a perfect work-life balance where I would reach the nirvana of job satisfaction and appreciation for my efforts from my patients and colleagues.

The future always seems so bright in the face of the dim present. As I grew into a fully-fledged physician, it turns out that acute appendicitis can show up as left upper quadrant pain with only an unexplained leukocytosis as a clue. An impending heart attack can show up as a “yucky feeling” when the patient goes up the stairs. Work-life balance has become less of a mantra, and more like an obsessive hunt for a spiteful whale, and in the end, it turns out that half-life is not quite what it seems.

As a toxicologist, I have a lot of conversations about half-life, and it made me realize that a lot of people still see half-life the way that I used to. I wanted to write this post to talk a little more about half-life, and the way that we can apply it towards overdose patients.

What is the half-life, if not a wildly successful video game franchise with a fanbase who has an undying faith in the release of Half-Life 3? In terms of medications, half-life describes the amount of time it takes to eliminate half of the concentration of a particular medication. If you take a medication with a half-life of four hours, it will take four hours for the concentration in the body to drop to half. Every four hours, the concentration halves, until, after five half-lives, you have eliminated 95% of the medication.

If you look up the pharmacokinetic data for a drug, you will see various pieces of information listed, half-life being one of them. This information is usually derived from lab-controlled settings with healthy volunteers taking a therapeutic dose of the drug. It is, at best, an estimate that relies on a set of assumptions that may not exist for your patient. There are some enzymes involved in drug metabolism that can drastically change the way a patient processes medications, such as what happens with tramadol and CYP2D6 polymorphism. Standard pharmacokinetic data doesn’t account for organ disease that may affect the way the drug is metabolized and eliminated from the body. That’s why many medications require dose adjustment in the setting of liver or kidney disease. It also doesn’t account for drug interactions that can affect drug metabolism. Organ disease and drug interactions can both shorten or lengthen half-life, depending on the circumstances. There are certain foods that could interfere with half-life because they are known to affect the enzymes that are involved in drug metabolism. What we know about the half-life of a drug is not always accurate in a real-world setting, and this information becomes even more unreliable in the setting of an overdose because of the way that our interaction with medication changes when it’s taken in supratherapeutic doses.

Most medications have information about a half-life because they are eliminated from our bodies in an organized fashion through first-order elimination. The enzymes available for metabolism outnumber the amount of drug in the body so that the drug interacts with the enzyme in a predictable way. Kind of like a toll booth where the number of available booths outnumbers the cars on the road. Traffic still flows smoothly, so the delays in travel are minimal. First-order elimination is based on drug concentration in plasma, which is how we determine the half-life.

In zero order elimination, there are way more cars on the road than toll booths available. Cars become backed up, so the rate of passage is limited by the number of toll booths available. This means that in zero-order elimination, the rate of elimination is constant and independent of the initial concentration. The drug is eliminated at an “x amount of drug/unit of time” rate, and that rate doesn’t change, so half-life loses its meaning. The most commonly known example of a zero-order substance is ethanol. Most of us are familiar with the idea that our bodies metabolize a drink/hour. No amount of ethanol we drink can change that elimination to a serious degree, which means that if you drink ten shots of alcohol in the hour before you go to bed, you should not be the one driving your friends to the beach six hours later, no matter how “sober” you feel.* If ethanol behaved according to first-order elimination, then it would completely change our approach to ethanol and time to sobriety. For example, if ethanol had a half-life of one hour, it wouldn’t matter how much you drink because, in about five-half lives, you’d be sober, so taking ten shots, six hours before you’re supposed to drive to the beach would be no problem at all.

*Not a true story

In an overdose, drugs that normally follow first-order elimination can switch to zero-order elimination. (We know that this is what happens with some drugs such as aspirin, although there are still many drugs for which we know nothing about elimination kinetics in overdose.) This means the normal half-life goes away as the enzymes get saturated due to the supratherapeutic level of the drug. As the drug gets eliminated and concentration decreases, the enzymes become available again, so it switches back to first-order elimination. We call this Michaelis-Menten elimination or non-linear elimination kinetics. Once it’s back to first-order elimination, we can use half-life to estimate when the drug is fully eliminated, but clinically, we may not know when that switch occurs. This is why a medication like baclofen, with a normal half-life of 3-4 hours, can have a “half-life” up to 34 hours in overdose.^​1​

What further complicates our use of half-life is that many people conflate half-life with other drug characteristics, such as peak concentration, duration of effect, redistribution, and onset to clinical action. The half-life does not always correspond to the onset of the clinical action of a drug. Paroxetine has a half-life of 21 hours, but its onset of action starts around six weeks. Diazepam has a half-life of 1-3 days, but its effects last approximately 4-5 hours. Even if the data we had on half-life was preserved in an overdose setting, it still doesn’t tell us much about what to expect in terms of the timing of drug toxicity, or even how long the toxicity might last.

The point of all this toxicologist rambling goes back to the original phone call at the beginning of this post, and the commonly-held assumption that half-life data can tell us when an overdose patient can be medically cleared from the effects of the overdose. Half-life can be interesting to discuss in regards to drug characteristics, but when it comes to overdose, it often loses any useful clinical application. In the rare case that half-life is preserved in overdose, it still doesn’t correspond to the onset of drug action, duration of effect, or even the timing of toxic overdose effects. If there’s any final statement about half-life that I would make, it’s this: Please don’t ask me about half-life anymore. We’re still friends, but we broke up a long time ago. I’m hanging out with Michaelis-Menten now.