Infections do change the way a person smells, and different pathogens produce distinct chemical signatures. The human body constantly releases hundreds of volatile organic compounds from the skin, breath, and waste, and when an infection takes hold, the mix shifts in ways that can sometimes be noticed by other people, by trained dogs, or by specialized sensors. Whether you can reliably identify a specific infection by its smell is a more complicated question, though, because human noses are far less precise than the chemistry involved.
Why Infections Change the Way You Smell
Your body is always off-gassing a blend of chemicals that reflects what is happening inside. These volatile organic compounds come from ordinary metabolism, from the bacteria that normally live on your skin and in your gut, and from the foods you eat. When a pathogen invades, it disrupts that chemistry in multiple ways. The microorganism itself may produce waste products with their own distinctive smell. Your immune system’s inflammatory response also alters which compounds your cells release. And the shift in your resident microbial communities as your body fights the intruder adds yet another layer of change. The overall result is that contracting an infectious or metabolic disease often changes body odor, sometimes in ways that are disease-specific enough to serve as diagnostic markers.1Oxford Academic. The scent of disease: volatile organic compounds of the human body related to disease and disorder
This is not one tidy mechanism. A wound infected with anaerobic bacteria smells different from a lung releasing viral byproducts, which smells different from urine colonized by certain gram-negative organisms. The chemistry varies because the biology varies. But the overarching principle is consistent: infection changes the volatile fingerprint your body puts into the air.
How Well Humans Can Detect Sickness by Smell
People do pick up on the smell of illness, but not with the precision you might hope for. In a controlled study where volunteers sniffed body odor samples from people who were either healthy or acutely sick with a respiratory infection, raters correctly identified the sick person’s sample about 57% of the time when given a paired comparison. That is better than the 50% you would expect from random guessing, and the effect was statistically robust, but it is far from a reliable diagnostic tool.2PubMed Central. Humans can detect axillary odor cues of an acute respiratory infection in others You might sense that something is “off” about a sick person’s smell, but you probably could not name the disease.
There is a reason your brain bothers with this at all. Disgust, especially toward bodily smells, appears to be an ancient system for keeping animals away from sources of infection. Researchers describe it as a “behavioral immune system” present across many species, one that triggers avoidance behavior in response to cues that something nearby might be carrying pathogens.3PubMed Central. Disgust as an adaptive system for disease avoidance behaviour The foul smell of a festering wound or the stench of certain diarrheal illnesses does not just make you recoil for aesthetic reasons. That recoil is a protective reflex shaped by natural selection, pushing you to distance yourself from a potential source of contagion.
The connection goes deeper than behavior. When healthy volunteers were exposed to odors they found disgusting, their bodies mounted a measurable immune response: levels of the pro-inflammatory signaling molecule TNF-alpha increased. The implication is that smelling something that registers as a pathogen threat may prime your immune system before you have even made physical contact with the source.4Evolution, Medicine, and Public Health. Disgusting odors trigger the oral immune system Your nose is not just a warning system. It may be part of the defense.
Bacterial Infections and Their Characteristic Smells
Some bacterial species are notorious for producing distinctive odors, and healthcare workers have informally relied on these smells for generations. Pseudomonas aeruginosa, a bacterium that commonly infects wounds and the lungs of people with cystic fibrosis, produces a sweet, grape-like odor that experienced nurses and doctors learn to recognize. The compound behind that smell has been identified as 2-aminoacetophenone.5PubMed Central. Use of 2-aminoacetophenone production in identification of Pseudomonas aeruginosa It is one of the clearest examples of a single bacterium leaving a detectable olfactory calling card.
Anaerobic bacteria, the kind that thrive in low-oxygen environments like deep wounds and the spaces between teeth, are especially prolific odor producers. Species in the Porphyromonas group, for example, generate volatile sulfur compounds including hydrogen sulfide and methyl mercaptan, the same chemicals responsible for the smell of rotten eggs and decaying matter. These compounds are a major cause of the foul breath associated with periodontal disease, and the same sulfur chemistry explains why infected wounds involving dead tissue can smell overwhelmingly putrid.6PubMed Central. Volatile Sulfur Compounds Produced by the Anaerobic Bacteria Porphyromonas spp. Isolated from the Oral Cavities of Dogs If you have ever been near a gangrenous limb or a severe abscess, you know the smell is unmistakable. That sulfur signature is the anaerobes at work.
Clostridioides difficile, a gut bacterium that causes severe diarrhea and can be life-threatening in hospitalized patients, has its own reputation for producing an unmistakable stool odor sometimes described as sickly sweet or like horse manure. Many nurses believe they can identify a C. diff infection by smell alone. The evidence, however, suggests this confidence is misplaced. When researchers formally tested whether nurse-identified “classic” C. diff odor could predict a positive lab result, they found a sensitivity of only about 29% and a specificity of roughly 84%. The positive predictive value was just 42%, meaning that more often than not, a stool that smelled like C. diff to a nurse actually was not C. diff.7PubMed Central. The nose knows not: poor predictive value of stool sample odor for detection of Clostridium difficile C. diff does alter stool odor, but other gut infections and antibiotic-disrupted flora produce overlapping smells. The human nose cannot reliably tell them apart.
What Tuberculosis Smells Like on Your Breath
Tuberculosis is one of the infections that has drawn the most interest from breath-analysis researchers, partly because it kills more than a million people a year and rapid, non-invasive screening tools could save many of those lives. People with active TB exhale a distinct cocktail of volatile compounds that healthy people do not produce. A recent study identified eight specific compounds that distinguished TB patients from healthy controls, including chemicals called tridecane, decane, and a compound called o-cymene that had not previously been linked to TB. None of these compounds appeared in breath samples from uninfected people.8PubMed Central. New volatile organic compounds from the exhaled breath of active tuberculosis patients
The same study found a compound that appeared almost exclusively in patients with multidrug-resistant TB, present in 45% of that group, raising the possibility that breath analysis could eventually help distinguish resistant strains from ordinary ones. These are not smells a person would notice in conversation. They exist at trace concentrations that require gas chromatography or similar lab instruments to detect. But they illustrate the principle: the infection leaves a chemical trail in the air you exhale, and that trail carries information about what is happening in your lungs.
Viral Infections Leave a Chemical Trail Too
It is not just bacteria. Viral infections alter the volatile compounds in your breath through a different route. When a virus like rhinovirus infects your airway cells, those cells mount an immune response that changes the metabolic byproducts they release. Researchers identified a signature of long-chain alkane compounds, including decane, that increased in exhaled breath during rhinovirus infection. These compounds correlated with how strongly the immune system was responding, how much virus was present, and how severe the illness became. Crucially, this particular signature was not triggered by bacterial infection, suggesting it may be specific to viral activity in the lungs.9PubMed Central. Virus-induced Volatile Organic Compounds Are Detectable in Exhaled Breath during Pulmonary Infection
COVID-19 prompted a wave of similar research. Multiple studies attempted to identify breath biomarkers that could distinguish SARS-CoV-2 infection from other respiratory illnesses. A systematic review found that several aldehydes, including octanal, nonanal, and heptanal, were elevated in the breath of infected children, while other studies identified additional compounds that discriminated COVID-19 patients from those with other conditions.10PubMed Central. The potential of volatile organic compounds-based breath analysis for COVID-19 screening: a systematic review & meta-analysis If some of those compound names sound familiar from the TB section, that is not a coincidence. Several aldehydes and alkanes appear across multiple infectious diseases, which means the challenge is not just detecting them but interpreting the overall pattern. A single elevated compound is rarely diagnostic on its own. It is the combination and relative concentrations that point toward a specific infection.
Malaria Changes Your Smell to Attract More Mosquitoes
One of the most striking findings in infection-odor research has nothing to do with diagnosis. The malaria parasite Plasmodium appears to manipulate the way its host smells in order to attract more mosquitoes and spread itself further. Studies in both mice and humans have documented this effect. In Kenyan children carrying the parasite, skin odor became more attractive to mosquitoes, and researchers traced the change to increased production of specific aldehydes, particularly heptanal, octanal, and nonanal.11PubMed Central. Plasmodium-associated changes in human odor attract mosquitoes When these compounds were added to a synthetic blend mimicking healthy human odor, mosquitoes were drawn to it more strongly. Even adding heptanal alone to the natural body odor of an uninfected person made that person more attractive to mosquitoes.
Mouse studies showed that this increased attractiveness peaked at a specific stage of infection: after acute symptoms had subsided but while the animal was still highly infectious. The timing is not random. That window is exactly when the parasite benefits most from having a mosquito bite the host, pick up the parasite, and carry it to someone new.12PubMed Central. Malaria-induced changes in host odors enhance mosquito attraction Field studies in Kenya confirmed that patients carrying the transmissible stage of the parasite attracted more mosquitoes than healthy individuals, an effect linked to a specific metabolite that triggers increased release of aldehydes and terpenes from red blood cells.13PubMed Central. Plasmodium falciparum gametocyte-induced volatiles enhance attraction of Anopheles mosquitoes in the field
This is a case where the odor change is not useful for human diagnosis at all. You will not notice that your child smells different after recovering from a malaria fever. But mosquitoes notice, and the parasite exploits that. It is also a vivid illustration of how deeply infection can rewrite the chemical signals a body sends into its environment.
Trained Dogs as Living Diagnostic Instruments
If the human nose picks up infection odors only slightly better than chance, the canine nose operates in a different league. Dogs have been trained to detect malaria, urinary tract infections, C. diff, and COVID-19 with accuracy that often meets or exceeds standard laboratory thresholds. In one study, dogs sniffing nylon socks previously worn by children detected malaria infection with a sensitivity of 72% and specificity of 91%, including in children who had no symptoms. That performance exceeded the diagnostic threshold set by the World Health Organization for malaria screening tools.14PubMed Central. Canine olfactory detection and its relevance to medical detection Dogs trained to detect bacterial urinary tract infections achieved sensitivity near 100% and specificity above 90% across multiple pathogens.
During the COVID-19 pandemic, several groups trained dogs to detect SARS-CoV-2 from sweat samples. In one large study using 21 dogs, over two-thirds achieved an overall sensitivity of 90% or higher when identifying positive samples, even when the analysis accounted for confounding factors like diabetes, fever, and other illnesses that might alter a person’s smell.15Open Access Journal of Veterinary Science & Research. Use of Canine Olfactory Detection for Covid-19 Testing Study on U.A.E. Trained Detection Dog Sensitivity The practical bottleneck is not accuracy but logistics. Training a detection dog takes months, each dog can only work for limited periods before tiring, and scaling a dog-based screening program to serve airports or hospitals is challenging. Dogs are proof of concept more than a deployable solution for most settings.
Electronic Noses and Sensor-Based Diagnosis
The gap between what a trained dog’s nose can do and what a human nose can do has motivated decades of work on artificial smell detection. Electronic nose devices use arrays of chemical sensors to build a “breathprint” or “odorprint” that can be compared against patterns associated with specific diseases. In hospital settings, these devices have shown real promise for infections. One study found that an electronic nose could detect ventilator-associated pneumonia with 88% sensitivity and 66% specificity by analyzing exhaled breath. In ear, nose, and throat infections, a preliminary study with 90 patients correctly identified bacterial infections in about 88% of cases, and the technology could even distinguish between different strains of Staphylococcus aureus, including the drug-resistant form MRSA.16PubMed Central. The electronic nose technology in clinical diagnosis: A systematic review
Electronic noses have also been tested against invasive fungal infections and late-onset sepsis in premature infants, with varying degrees of success. The technology works on breath, but also on gases released from stool and from wound surfaces. The core limitation is standardization. Results vary between devices, between research groups, and between patient populations. No electronic nose has yet become a routine clinical tool for infection diagnosis, though several are in advanced development.
Wearable Sensors That Monitor Wound Infections
One of the most practical near-term applications of infection smell science involves wound care. A recently developed wearable device sits over a wound without touching it and measures the volatile compounds and water vapor rising from the wound surface. Unlike smart bandage prototypes that need direct contact with the fragile wound bed, this sensor creates a small sealed air pocket above the wound and monitors the gases diffusing through it. The device is wireless, miniature, and designed for repeated transient measurements, allowing clinicians to track whether a wound is healing normally or becoming infected without disturbing the tissue.17PubMed Central. Healing cascades and infections in wounds monitored using a wearable sensor of gaseous flux
This approach is appealing because wound infections are common, costly, and often caught late. A sensor that could flag a shift in volatile chemistry hours or days before visible signs of infection appear would give clinicians a head start on treatment. The device is still in its research phase, but it represents the direction the field is heading: continuous, passive, non-invasive monitoring of the chemical signals that infections broadcast into the air around them.
When Plants Get Sick, They Smell Different Too
The connection between infection and altered volatile chemistry is not unique to animals. When plants are attacked by pathogens, they release their own volatile organic compounds, including monoterpenes and other molecules that serve as internal alarm signals. These plant volatiles can trigger defensive responses not only within the infected plant but also in neighboring plants that detect the compounds in the air.18Portland Press (Essays in Biochemistry). New molecules in plant defence against pathogens Some of these compounds are produced by the invading microbes themselves, adding another layer to the volatile conversation between host and pathogen.
This cross-kingdom parallel suggests something fundamental about life’s relationship with infection. Whether you are a human with a Pseudomonas wound, a child carrying malaria parasites, or an oak tree fighting off a fungal pathogen, infection changes the chemical message your body puts into the environment. The specific compounds differ, but the principle is ancient and universal. For organisms with the sensory apparatus to detect these signals, infected individuals smell different from healthy ones, and that information, whether it triggers disgust in a human, attraction in a mosquito, or a defensive gene cascade in a plant, shapes the outcome of the encounter.