I’m sick, and I don’t smell right. I don’t mean that my nose isn’t working—though this cold has me stuffed up. Instead, my own body odor seems somehow different, sour and unfamiliar.
I’m far from the first person to notice this nasty side effect. Scientists have found that dozens of illnesses have a particular smell: Diabetes can make your urine smell like rotten apples, and typhoid turns body odor into the smell of baked bread. Worse, yellow fever apparently makes your skin smell like a butcher’s shop, if you can imagine that.
It’s curious, but not merely a curiosity; some scientists think that if we could identify particular sick smells, we might sniff out diseases that are otherwise hard to detect early, like cancers or brain injuries. Recently, a Scottish woman became famous for her ability to tell whether someone has Parkinson’s disease by smelling their T-shirts.
The star-nosed mole has one of the most sensitive sniffers on the planet. Not only does it use its tentacled nose to feel for prey, but it can also smell underwater by blowing and re-inhaling bubbles.PHOTOGRAPH BY ZIGMUND LESZCZYNSKI, ANIMALS, EARTH SCENES, NATIONAL GEOGRAPHIC CREATIVE
Some sharks can detect the odor of prey in water at one part per 10 billion, equivalent to a drop in an Olympic-size swimming pool.PHOTOGRAPH BY MAURICIO HANDLER, NATIONAL GEOGRAPHIC CREATIVE
An albatross (black-browed albatrosses shown) can pick up the scent of food 12 miles away.PHOTOGRAPH BY PAUL NICKLEN, NATIONAL GEOGRAPHIC CREATIVE
Many animals have a fine sense of smell without having a nose at all. This Atlas moth smells using chemical sensors on its antennae.PHOTOGRAPH BY MATTIAS KLUM, NATIONAL GEOGRAPHIC CREATIVE
Snakes, like this red-bellied black snake, flick forked tongues to capture scent molecules, which are then transferred to two pits at the top of the snake’s mouth that hold a sensory center called the vomeronasal organ.PHOTOGRAPH BY JASON EDWARDS, NATIONAL GEOGRAPHIC CREATIVE
We marvel at such a skill, but anyone with working olfactory senses could probably learn to recognize various “sick smells.” Humans are very good at detecting illness, says Valerie Curtis, a public health researcher at the London School of Hygiene and Tropical Medicine and author of the book Don’t Look, Don’t Touch, Don’t Eat on the science of disgust.
“Signs of sickness are some of the things people find most disgusting,” Curtis says—think mucus, vomit, or pus. Disgust is our way of avoiding things that could harm us, so “it simply makes good evolutionary sense that we use our noses to notice illness.” (Of course, people sometimes like stinky things, too.)
But why would sick people smell differently in the first place? The key is that our bodies are constantly launching volatile substances into the air. They’re carried in our breath and literally ooze from every pore, and they can vary depending on age, diet, and whether an illness has thrown off some cog in our metabolic machinery. Microbes living in our guts and on our skin also contribute to our signature scent, by breaking down our metabolic by-products into smellier ones.
Basically, you’re a walking factory of smells. And if you start paying attention to them, you might notice when something’s off.
THAT’S NOT PATCHOULI
Recently, the case of the woman who can smell Parkinson’s brought attention to the idea of sniffing for disease. Parkinson’s is notoriously tricky to diagnose; by the time most people learn they have it, they’ve already lost half of the dopamine-producing brain cells the disease attacks. But about six years before her husband Les was diagnosed, Joy Milne noticed that he smelled odd.
Les had a “sort of woody, musky odor,” Milne told the Telegraph. Years later, in a room full of Parkinson’s patients, she realized the smell wasn’t unique to Les. All the people with Parkinson’s smelled that way.
She mentioned it to a Parkinson’s researcher in Edinburgh named Tilo Kunath, who mentioned it to his colleague, analytical chemist Perdita Barran. They decided the well-meaning Mrs. Milne may have just noticed the characteristic smell of old people; “We talked ourselves out of it,” Barran says.
That could have been the end of it. But another biochemist encouraged the pair to track Milne down and try a blind T-shirt test: She sniffed six sweaty tees from people diagnosed with Parkinson’s, and six from healthy controls. Milne correctly identified which six had Parkinson’s, but she also tagged one of the control subjects as having the disease.
Despite that error, Barran was intrigued—all the more so eight months later, when the same supposedly healthy control subject Milne had identified was diagnosed with Parkinson’s.
PASSING THE SMELL TEST
The T-shirt test was intriguing, but we have to take it with a scientific grain of salt. After all, there are lots of reasons people might share an odor.
In one notorious dead end, researchers were convinced there was a smell linked to schizophrenia, and a particular compound called TMHA—said to smell like a goat—was identified and described in the prestigious journal Science. There was hope this chemical might even be the cause of schizophrenia, which would open up new avenues for treatment.
But in years of follow-up testing, the results couldn’t be repeated. The TMHA “schizotoxin” went the way of tabletop nuclear fusion.
Barran is now at the Manchester Institute of Biotechnology, where she’s applying the painstaking methods of chemistry to determine whether the Parkinson’s smell is the real deal. She and her colleagues hope to develop a smell test for Parkinson’s, one more rigorous and more practical than having Mrs. Milne smell all our T-shirts.
First, the team is working to chemically identify the molecules involved, which is harder than it looks on CSI. Of the thousands of known volatile compounds, many are not well characterized or data on them exists only within the fragrance industry.
With funding from Parkinson’s UK and the Michael J. Fox Foundation, Barran’s team has already collected more than 800 samples of sebum, an oily substance secreted by skin, swabbed from the backs of volunteers. In preliminary tests they’ve found several molecules that are elevated in people with Parkinson’s disease. Together, the molecules could create a diagnostic fingerprint for the disease.
Next, the team not only has to confirm that these particular molecules are reliably elevated in Parkinson’s patients, but also figure out whether they can detect the smell before symptoms of Parkinson’s appear. Ideally, they would also learn how Parkinson’s triggers the body to produce the molecules.
Barran says she’s up for the challenge—even though her own sense of smell was damaged in an accident and she can’t smell the Parkinson’s odor herself.
“Joy [Milne] has an extremely good sense of smell,” Barran says, “but she isn’t the only person who can smell it. What’s special is how persistent she was in her conviction that it was something that could be used.”
SCENT OF A PATIENT
That brings us back to the question of what you and I can actually smell. While dogs have the most lauded sense of smell and have been tapped to sniff out cancer, research suggests that humans are just as good at detecting many odors.
Judging by the number of neurons in our brains’ olfactory bulbs, people may be better smellers than rats and mice and fall about in the middle of the pack among mammals. Perhaps the biggest barrier to our abilities is that we don’t pay enough attention to smells, and we lack a sophisticated language to describe them.
“We’re less able to rationalize smell,” says Curtis. She recalls using a soap she had brought home from India: “The idea of ‘India’ popped into my head long before I realized it was the smell.”
Likewise, we may not realize when we’re smelling a change in our own or a loved one’s health.
There are hints, though, that we may be decent illness detectors if we pay attention to the task. In a small double-blind study published in the Proceedings of the National Academy of Sciences in 2017, participants could identify sick versus healthy people based on body odor and photographs just a few hours after some of the people’s immune systems were triggered by a toxin that mimicked infection.
So while we don’t yet have a breathalyzer for disease, we might do well to follow our noses.