The idea of Dogs can detect Parkinson's disease by smell. It sounds like sensationalist news, but behind those headlines lies serious science, controlled studies, and years of work by neurologists, chemists, and canine behavior experts. From the story of a nurse who could smell the disease in her husband to dogs trained in laboratories in the UK and the US, the topic has gone from anecdote to high-level research.
In recent years, the following have been published Double-blind trials with detection dogsChemical analyses of skin sebum, pilot projects in several countries, and proposals to transform this prodigious sense of smell into non-invasive diagnostic tests are all underway. However, it's important to distinguish between what has already been proven and what is still promising, because saying that dogs "diagnose Parkinson's years in advance" can lead to misunderstandings if not properly clarified.
From intuition to the laboratory: the smell of Parkinson's
Long before the dogs came into the pictureThe story began with one person: Joy Milne, a Scottish nurse with hyperosmia, a sense of smell far more sensitive than average. In the mid-80s, she began to notice a strange, musky, sticky odor coming from her husband, concentrated mainly on his shoulders and the back of his neck. She thought it was due to his work in the operating room, sweat, or medication, but the smell became increasingly intense and persistent.
Twelve years later, her husband was diagnosed Parkinson's disease at age 45Joy, upon meeting others with the same condition, realized they all shared that characteristic odor. She would later learn that she has hyperosmia and that what she was detecting were metabolic changes linked to the disease, long before the typical motor symptoms appeared.
In 2012, during a Parkinson's conference, Joy felt encouraged to share her story publicly. This caught the biologist's attention. Tilo Kunath, from the University of Edinburgh, a specialist in neurodegeneration. From that moment on, his personal experience became the starting point for several lines of research to demonstrate, with data, that Parkinson's leaves a measurable olfactory signature.
Today we know that the disease causes systemic metabolic disorders These changes manifest on the skin through sebum, an oily secretion produced by the sebaceous glands. These alterations result in a specific profile of volatile organic compounds (VOCs) that form a unique "olfactory signature" of the disease, imperceptible to most human noses, but not to someone like Joy, and certainly not to a well-trained dog.
How to train dogs to detect Parkinson's by smell
Based on this, several research teams in the United Kingdom joined forces with the charity. Medical Detection DogsA pioneer in training dogs to detect diseases. Her goal: to test, under strict laboratory conditions, whether dogs could distinguish between sebum samples from people with Parkinson's and healthy controls.
In one of the most cited studies, published in the Journal of Parkinson's DiseaseThey worked with two dogs: Bumper, a golden retrieverand Peanut, a black Labrador. For approximately 38-53 weeks, these animals were exposed to more than 200 sebum samples from people with a confirmed diagnosis of Parkinson's disease and from individuals without the disease.
The samples were collected with swabs or absorbent discs on the skin, coded, and presented to the dogs in a support system designed for olfactory testingEach time the dog correctly indicated a positive sample (Parkinson) or appropriately ignored a negative one (control), it received a reward. Thus, association after association, they learned which scent profile to indicate.
The methodological key lay in the use of rigorous double-blind testsNeither the dog handlers nor the people placing the samples knew which were positive and which were negative; that information was held only by the computer system that generated the random presentation order. Furthermore, the lines of samples were presented in both forward and reverse order, and those that remained undecided were regrouped into new series until the dog indicated an option.
The results were conclusive: the dogs achieved a sensitivity up to 80% (ability to correctly identify Parkinson's samples) and a specificity of up to 98% (ability to rule out healthy people). They were also able to recognize the Parkinson's smell in samples from patients who had other associated pathologies, which reinforces the specificity of the olfactory signature.
International studies and surprising accuracy figures
The work in Bristol and Manchester is not an isolated case. Independent international studies They have reached very similar conclusions: Parkinson's modifies the profile of lipids and volatile compounds in sebum, and these changes can be detected even in prodromal phases, when the person has not yet received a formal neurological diagnosis.
A study led by the University of Manchester and published in ACS Central Science The researchers chemically analyzed sebum from people with and without Parkinson's disease, identifying approximately 500 substances with clear differences between the two groups. These analyses support the idea that a stable chemical signature exists that could serve as an early biomarker.
In parallel, dogs of other breeds have been trained, such as Belgian ShepherdsThe dogs were exposed for 1,5 to 2 years to skin samples from Parkinson's patients (both treated and untreated) and from hundreds of healthy individuals. In one of these studies, the dogs sniffed gauze impregnated with tallow frozen at -30°C and, when they identified the odor consistent with the disease, they lay down on the sample, indicating it to the researchers.
The numbers obtained speak for themselves: a overall sensitivity of 91% and a specificity of 95%In other words, the dogs were correct with a rate far exceeding that of chance, comparable to and even higher than that of certain diagnostic tests we use for other diseases. Furthermore, they were able to differentiate Parkinson's samples from those of other neurological conditions, which supports the idea that they are not reacting to a "generic disease odor," but to a specific pattern.
There are also projects like the Parkinson's Canine Detection ProjectIn the San Juan Islands (Washington, USA), some 20 dogs are being trained to identify Parkinson's disease from the scent of used T-shirts sent in by potential patients. At the end of 2017, accuracy rates of nearly 90% were reported, according to Good News Network, showing that the approach can work outside of a single laboratory.
Can dogs really detect Parkinson's disease years before symptoms appear?
Many media headlines highlight that dogs can detect Parkinson's disease. “years before motor symptoms appear”This statement has a basis in reality, but it needs important nuances to avoid creating false expectations or confusion among patients and their families.
Neuropathological and imaging studies have shown that, when Parkinson's disease is clinically diagnosed, Between 50% and 70% of the dopaminergic neurons in the substantia nigra have already been lostIn other words, the brain undergoes changes for some time before the resting tremor, rigidity, or slowness of movement that we recognize as classic symptoms appear.
Furthermore, there are indications that metabolic alterations in the skin and sebum production change up to 10 years before so that the neurologist can make a diagnosis using current criteria. Hence the idea that Parkinson's "begins at the cellular level up to a decade earlier," as biologist Tilo Kunath explains.
What studies with dogs and with people like Joy Milne show is that there is a early olfactory signature associated with these biochemical changes, and both an extremely sensitive sense of smell and that of a trained dog are capable of detecting it. However, it is not yet fully demonstrated, with large cohorts followed long-term, how many years before clinical diagnosis this signature appears consistently in the entire population.
Therefore, although it is reasonable to think that a dog's nose can anticipate a neurologist's visit by years, We do not yet have a validated early screening system in human medicine based solely on canine olfaction. Currently, these dogs are primarily a powerful research tool for discovering and validating chemical biomarkers, rather than a ready-to-use, mass-market diagnostic method for hospitals.
From dog to laboratory: skin swabs and electronic noses
The success of detection dogs doesn't mean we're going to fill hospitals with them. In fact, Dogs are serving as a biological model so that the technology knows what chemical signals to look for and how to interpret them.
A recent study, again in collaboration with Joy, focused on taking sebum samples using simple skin swabsThese samples are very similar to the swabs used for COVID-19 nasal tests. They are taken to a laboratory and analyzed with a mass spectrometer, an instrument capable of identifying the compounds present and their relative abundance.
The procedure, simply explained, consists of ionizing the sample molecules by bombarding them with electrons, so that they lose electrons and become positively charged. These ions then travel at different speeds depending on their mass and are deflected in a characteristic way when passing through a magnetic field. A detector records these deflections, and from them, the sample is reconstructed. “chemical signature” of the sample.
These types of studies have identified hundreds of different substances in both patients and healthy individuals, narrowing the list down to a set of key compounds that appear to define the smell of Parkinson's disease. Research is currently underway using these compounds. electronic noses and advanced chemical sensors that attempt to mimic the dog's ability to discriminate, but in a reproducible, standardizable device suitable for clinical practice.
The synergy between biology and technology is fundamental here: the dog shows scientists that there is a detectable pattern, helps validate which samples are positive or negative, and guides the search for specific molecules. In turn, artificial sensors will allow, in the future, for the broad application of this information, without depending on the availability of trained dogs or people with exceptional senses of smell.
Why is it so important to detect Parkinson's disease as early as possible?
Parkinson's disease is the second most frequent neurological pathology Alzheimer's disease is expected to soon affect around 10 million people worldwide, and in Spain alone, it is estimated that more than 160.000 people live with it. One of the biggest challenges is, precisely, early diagnosis.
In current clinical practice, diagnosis is usually based on the detailed neurological evaluation (medical history, physical examination, functional tests) and, in some cases, imaging techniques such as magnetic resonance imaging or specific dopamine studies. Genetic testing can also be performed in families with a history of the disease. However, there is still no single, simple, and completely reliable test that can detect Parkinson's disease in its silent stages.
When tremors, rigidity, or postural changes become evident, the neurodegenerative process has already been progressing for years. This means that, although treatment can improve symptoms and quality of life, It arrives too late to prevent much of the neuronal damage.Hence the enormous interest in finding early biomarkers, whether olfactory, imaging or genetic.
If we can accurately identify who is developing the disease before it manifests, it opens a very valuable therapeutic window. We could implement neuroprotective strategies (pharmacological and non-pharmacological), adjust lifestyle, better plan for the future and facilitate access to clinical trials specifically designed for early phases, where new drugs are more likely to work.
From this perspective, dogs are not the end point of the journey, but a a key step towards more preventive and personalized medicine, in which time ceases to be an enemy and becomes a therapeutic resource to be used.
Advantages, limitations and real role of detection dogs
Studies with dogs show several clear advantages: it is a method non-invasive, inexpensive, and potentially accessibleIt is based on easily obtainable samples (skin sebum, used T-shirts, skin swabs). Furthermore, the sensitivity and specificity rates achieved in controlled trials rival those of some modern diagnostic tests.
However, it's not as simple as "putting a dog in every hospital." Training these animals is a lengthy and specialized process that requires professionals with experience in olfactory detection, protocol design, and animal welfare. It is unrealistic to turn dogs into substitutes for neurologists nor in the only screening tool.
His current role is that of high-precision biological sensors integrated into research projects: they help to confirm that there is a real chemical pattern, serve as a reference to compare emerging technologies and, in certain contexts, could be part of early detection pilot programs combined with other tests.
Furthermore, the widespread use of dogs raises questions of standardization, logistics and ethicsHow do you certify that a dog is sufficiently trained? How do you guarantee reproducibility between different centers? What happens if a dog indicates "positive" for a person without symptoms? How is that result communicated? All of this requires well-thought-out protocols and clear regulatory frameworks.
The researchers themselves insist, as does Claire Guest, director of Medical Detection Dogs, that the priority is to use these findings to develop fast, reliable and cost-effective methods for detection, and not to create a kind of "canine oracle" that replaces regulated medical diagnosis.
Ethical and emotional implications of early olfactory diagnosis
Detecting a neurodegenerative disease such as Parkinson's years before it manifests It opens up important ethical debates. Knowing well in advance that the disease is likely to develop can be a double-edged sword: it provides room for action, but it can also generate anxiety, uncertainty, or fear.
Ethics committees tend to agree that early information, when managed well, empowers the patientThis requires an appropriate context: clear medical advice, psychological support, an explanation of risks and probabilities, and concrete options for follow-up and action. It would be pointless to communicate a “positive feeling” without offering the person a reasonable plan of care and monitoring.
In this context, the dog should not be presented as an "infallible diagnostician," but as a biological sentinel which alerts to the possible presence of early biomarkers. The final word on the diagnosis will still rest with the neurologist, who will integrate the olfactory information (whether canine or instrumented) with the clinical presentation and the rest of the available tests.
The well-being of the animals themselves must also be considered. Working in medical detection involves long and demanding training, so it is essential that their physical and emotional needs are respected, that they have rest, play, and a balanced life, and that they are not treated as mere “diagnostic machines.”
When approached with scientific rigor, ethics, and empathy, early olfactory diagnosis of Parkinson's It can become a powerful tool that benefits both patients and the advancement of medicine, without placing a responsibility on the dogs that does not belong to them.
Everything we know so far suggests that the body "warns" of Parkinson's disease long before tremors appear, and that this warning is released into the air in the form of volatile compounds that some exceptional senses of smell, both human and canine, are able to detect. Converting this ability into useful, safe, and accessible diagnostic tests is the great challenge today. The dogs have already proven that the trail is there.Now it is up to science and technology to transform that discovery into tools that reach the consultation room and allow us to gain time against a disease that, for now, remains incurable.
