Previous: The Male Biological Clock
Next: Hotter Than Death Valley | Weird Places



View count:1,984
Last sync:2019-11-17 17:20
Go to to try their Casino Probability course. The first 200 subscribers get 20% off an annual Premium subscription.

Blood is a pretty remarkable diagnostic tool! It doesn’t just tell you about blood-related diseases, it can reveal information about everything from your brain to your unborn fetus.

Hosted by: Michael Aranda

SciShow has a spinoff podcast! It's called SciShow Tangents. Check it out at
Support SciShow by becoming a patron on Patreon:
Huge thanks go to the following Patreon supporters for helping us keep SciShow free for everyone forever:

Eric Jensen, Matt Curls, Sam Buck, Christopher R Boucher, Avi Yashchin, Adam Brainard, Greg, Alex Hackman, Sam Lutfi, D.A. Noe, Piya Shedden, Scott Satovsky Jr, Charles Southerland, Patrick D. Ashmore, charles george, Kevin Bealer, Chris Peters
Looking for SciShow elsewhere on the internet?

Preterm Babies


Artery Blockage


Alcohol Use Disorder

This episode of SciShow is sponsored by Brilliant!

Go to to learn more. [♪ INTRO]. It usually isn’t possible to simply glance at a sick person and instantly know exactly what disease they have.

In order to diagnose diseases and assess risks, doctors often rely on biomarkers, a broad category that basically represents anything they can measure to learn something about your health. And one of the easiest things to collect and measure is blood. But blood is fairly exceptional, in that it doesn’t only tell you about blood-related diseases.

It can reveal information about everything from your brain to your unborn fetus. Here are five things your blood can tell you about your health that you might not expect. Doctors are working on ways to predict if your baby will be born up to two months prematurely based on patterns in your blood.

Premature birth is one of the leading causes of infant death worldwide. Around 15 million babies are born before 37 weeks every year. But one way to help this at-risk population is by knowing their gestational age: exactly how long they’ve been developing in the womb.

While being born prematurely carries risks no matter what, the chance of medical complications increases as gestational age decreases. So if a baby is born prematurely, a solid idea of how developed it is will help doctors prepare for those complications. Technologies like ultrasound can help us find gestational age, but they don’t give us as much information as we’d like on how likely a baby is to be born early.

Recently, scientists have turned to things called cell-free RNA transcripts, or cfRNA for short. These are bits of the fetus’s genetic information that make their way into the parent’s blood. Different genes are expressed at different times during pregnancy, so depending on the amount of cfRNA we find from different types of tissue, we can get an idea of the fetus’s gestational age and risk for premature birth.

One 2018 study attempted to determine the genetic patterns in more detail, in order to develop a cfRNA-based blood test to estimate the risk of preterm birth. They saw some things they expected, like increased cfRNA from the placenta and liver as the fetus developed. Then they used those patterns to make a model and estimate when the baby would be born.

While the resulting model was fairly accurate at measuring gestational age, it didn’t do any better than ultrasound. And it could only accurately predict the delivery date for full term births, not preemies. A few different things can trigger a preterm birth, and since we don’t totally understand how they work yet, the research group thinks those unknown factors might be why their estimate isn’t super accurate.

This new blood test isn’t quite ready for the clinic yet. The first experiment only tested a small number of white and African American people at risk of preterm birth. In the future, larger and more diverse samples may be able to help develop a truly usable blood test.

Hitting your head is serious. And if you hit your noggin just right, you might experience memory loss, dizziness, or other mental alterations, all of which are signs of a traumatic brain injury called a concussion. But no two concussions are diagnosed the same, so doctors need some objective markers.

Recently, they’ve used a blood test in combination with imaging like CT scans to help them diagnose head injuries. When someone presents with signs of a concussion, doctors first give them a blood test, and they’re looking for two biomarkers in particular, GFAP and UCH-L1. They’re great biomarkers for head injuries because the amount we can detect increases with the severity of the injury, so how much there is tells us how bad it is.

Plus, GFAP is only expressed in the brain, which makes it a very specific marker for head trauma. So that’s great for concussions. But if someone hits their head and doesn’t show concussion symptoms, we’d still like to know how bad their injury is.

Repeated smaller head injuries can still carry long term health risks, like for athletes who experience repeated minor head trauma. So doctors need a way to track these milder injuries. A study published in 2019 looked at GFAP and UCH-L1 in people with head injuries.

Participants were categorized as either having concussion symptoms, head injury without symptoms, or body trauma only. The researchers tracked these biomarkers at different times after the injuries to see if any patterns emerged. They found that both biomarkers were elevated after head trauma, although the level of UCH-L1 in someone’s blood was higher after even minor head injuries, meaning it wasn’t useful for showing how bad the injury was.

GFAP, on the other hand, scaled up more consistently with the severity of head trauma, even when the person didn’t show symptoms. This means that while both are still good markers for diagnosing concussion,. GFAP was more informative when it came to showing how bad.

While many doctors’ offices already test for these biomarkers, the researchers say that a wider range of measurement devices, like for ambulances, may be available very soon. By now you probably know that heart disease is kind of a big deal. Heart disease is still the biggest killer in the world, and having narrowed or blocked coronary arteries, otherwise known as atherosclerosis, is one of the many risk factors for that.

But as you can imagine, no single test predicts your risk for this disease perfectly. For instance, we already have plenty of biomarkers for cardiovascular disease risk by itself. But if you have narrowed arteries, some of those biomarkers all of a sudden aren’t as accurate.

There are a few ways to measure how blocked those vessels are, though. This includes stuff like echocardiograms and exercise. In early 2019, research out of Duke University Medical Center tried to develop a test that used both exercise tests and biomarkers.

Using a combination of tests would give them a better idea of what happens chemically around ischemia, when the heart doesn’t get enough oxygen. Specifically, they wanted to see what happened to those biomarkers over the course of an exercise test and pick out any interesting patterns. The researchers were interested in about 60 chemicals related to the heart’s energy usage.

In particular, a group of chemicals called acylcarnitines. So they recruited participants who were already patients in the emergency department with heart disease symptoms like chest, jaw or shoulder pain. Researchers took blood samples before and after the exercise test and did an echocardiogram to get a picture of how blocked their coronary arteries were.

Their results narrowed it down to four acylcarnitines and one amino acid that might be good at predicting atherosclerosis in the future, and identified which ones change after exercise. The researchers claim their method is a better predictor of atherosclerosis than the blood tests we have now, since it gives a snapshot of conditions before and after ischemia. The catch is, we don’t yet know how much better.

It’s still too early to say when this will be used more regularly. A study with a larger sample size is still needed to validate these initial findings. It may seem strange that a chemical in your blood could be used to diagnose a mental health condition, but it’s totally a thing.

For how widespread depression is, we still don’t really understand the physiology behind it. Having some kind of biomarker for depression would allow doctors to come up with more consistent diagnoses and more targeted treatments. But what chemical are we looking for?

Recently, a chemical called acetyl-L-carnitine, or LAC for short, has been getting some attention thanks to its antidepressant action in rodent studies. LAC is involved in a chemical pathway that ultimately regulates the release of a neurotransmitter called glutamate, one of the chemicals that neurons use to communicate with each other. Now, as far as we can tell in animal models, adding LAC to their diet can work as an antidepressant if that animal had a low level of LAC to start with.

But that doesn’t necessarily mean it works in people, so researchers have started looking into that, too. In a 2018 study, researchers recruited over 100 participants from two different medical centers, gave them all clinical psychiatric exams, and drew some blood. On average, participants with diagnosed depression had lower LAC concentrations than the control group.

And among those participants with moderate to severe depression, things like childhood trauma and a younger age of onset were associated with even lower LAC levels. The research so far only supports LAC for diagnosing depression, not as an oral supplement to treat it. Although that hasn’t stopped companies from trying to sell it for that purpose.

However, researchers were optimistic that with more clinical validation, patients with depression and low LAC might one day benefit from LAC supplements. But it won’t ever be a one size fits all mood boosting pill. Alcohol use disorder, or AUD, is a potentially life-threatening condition, and detecting it is trickier than it seems.

There are different types of drinking behavior associated with AUD. Binge drinking in particular is especially common on college campuses. But researchers aren’t gonna follow college students to the bars on Saturday nights, so they gather data by having their participants fill out questionnaires.

All of these questionnaires ultimately try to get information about how much and how often these folks are drinking. One of them, the Alcohol Use Disorders Identification Test, or AUDIT, is the best at actually identifying binge drinking in college students. But questionnaires are still subjective.

And using a concrete number, like a measurable biomarker, would allow for more objectivity in screening for AUD and getting help for those who need it. We already have urine tests to see if someone abstains from alcohol, but no tests that define alcohol use disorder. One of the best candidates for an AUD test is a biomarker called phosphatidylethanol, or PEth, a molecule that only shows up in the blood when alcohol is present.

But finding a little bit of PEth in someone’s blood doesn’t diagnose AUD, and a single alcoholic beverage doesn’t raise PEth levels any noticable amount. Unfortunately, because each study that measures PEth and alcohol use has been designed differently, it’s hard to compare them side by side and come up with a definitive range of PEth levels. The only thing we can really say is that the more you’ve been drinking lately, the higher your PEth.

So a study published in 2015 tried to double up the diagnostic power and tested the AUDIT questionnaire with PEth levels in the blood. To do this, they recruited young adults from college campuses and had them complete the AUDIT to learn about their alcohol consumption behaviors. The questionnaires put them into one of three groups: binge drinkers, moderate drinkers, and abstainers.

Like in past studies, the researchers found more PEth in the blood of binge drinkers than moderate drinkers and none in the abstainer group. That makes PEth an extremely sensitive and objective measure for alcohol consumption. But again: details.

What that really tells us is that PEth is associated with how much alcohol someone drinks, not how often, what time of day, or their relationship with alcohol. In other words, there’s a bunch of variables that it can’t pin down. This study still depended on self-reported questionnaires, so it isn’t enough evidence to say that PEth levels can identify if someone has AUD.

With all of these biomarkers, it’s important to remember that they’re adding to the pool of information that healthcare professionals use during treatments. One marker isn’t necessarily the end-all, be-all test for diagnosing a condition. But with more in-depth research into each of these chemicals, doctors get a more precise picture of their patients’ status.

And that information can be used to help improve their outcomes. Thanks for watching this episode of SciShow, which was brought to you by If you haven’t had enough of getting smarter today, and really, who has, you might be interested in the courses over on Brilliant.

Brilliant has over 60 courses about science, engineering, computer science and math. And those courses cover everything from the basics to more advanced topics and deep dives. Like, there’s a whole course on casino probability, for example.

It’ll teach you the basics behind games like poker and blackjack. And while it can’t help you beat the house every time, at least you’ll know the odds. These courses are made by dedicated math and science educators and lifelong learners from MIT, Caltech, Duke, the University of Chicago, and more, so you know they know what’s up.

If you’re interested, you can sign up at and get 20% off an annual Premium subscription. And thanks! by checking them out, you’re helping us make more awesome videos. [♪ OUTRO].