Lindsay marvels at the scent’s staying power. “It’s extraordinary that you can give nylons socks to a child, pick them up the following day, wrap the socks in aluminum foil, pop it in a plastic bag, freeze them for 12 to 15 months, then start training the dogs.”

Lexi and Sally were first given a crash course in sniffing on a subset of the socks. For several months, the pair underwent six sessions of intensive schooling each week at the Medical Detection Dogs charity in the United Kingdom, gradually learning to recognize the scent of infection from parasitically perfumed nylons. By the end of their training, both eagerly gravitated towards the socks that smelled of disease.

Then came the real test. When faced with a full battery of 145 uninfected and 30 infected socks, the dogs correctly identified 90% of the malaria-negative and 70% of the malaria-positive samples. Lindsay is confident that the numbers will only improve as Lexi and Sally practice with different samples and populations. And in the days since the first trials, a third dog named Freya has been trained for the same set of tasks.

“This is a very interesting and original idea,” says Marije K. Bomers, an infectious disease specialist at the VU University Medical Center in Amsterdam who has studied dogs’ ability to detect C. difficile infections in human stool. “The results are far from perfect, but I think it’s a really good start. It’s hopeful and positive.”

Freya, a Springer Spaniel, puts her nose to the test with sock samples from children in The Gambia—some of which may carry a whiff of malaria. Photo Credit: Durham University/Medical Detection Dogs/London School of Hygiene & Tropical Medicine

For now, Lindsay and his team have set their sights on a couple specific targets. The dogs could operate at ports of entry—where there is often already infrastructure to support canine professionals—into countries where malaria prevalence is low or diminishing, preventing the reentry of parasites that might spark a new wave of infection. In the United States, for example, malaria remains uncommon in clinics but often manages to invade U.S. ports of entry. About 1,700 cases are diagnosed each year within U.S. borders, mostly in travelers returning from countries where the disease runs rampant.

Additionally, dogs could act as scouts in the field, nosing out the last infected members of communities where malaria has almost been eradicated. These cases have proven especially troubling to detect, especially when the infections are asymptomatic. But even in the absence of active illness, canines may be able to sniff out the telltale bouquet of infected blood.

Researchers aren’t completely sure why mosquitoes and mutts are so sensitive to malarial musk, or where the scent originates. It may be that the parasites themselves are particularly pungent. When it comes to mosquitoes, at least, such a strategy could be beneficial to Plasmodium: It ups the parasites’ chances of hailing a cab to their next victim—an olfactory version of a hitchhiker’s thumb. Alternatively, the distinctive odor could arise from the human body’s response to infection. Both explanations could be true—or the answer could be something else entirely.

Receive emails about upcoming NOVA programs and related content, as well as featured reporting about current events through a science lens.

There’s also no guarantee that the aroma that’s attracting mosquitoes is also what is piquing the pooch proboscis. “We don’t know what the dogs might be detecting,” Lindsay says, “but at this stage, we just want to know whether they’re detecting or not.”

If the team’s first trials are any indication, it seems the dogs are detecting. But the researchers have a long road ahead of them before this breed of diagnostic leaves the lab. A scene of smelly socks lined up in a research facility doesn’t necessarily translate to patients speeding through clinics or tourists navigating chaotic hubs of travel. Additionally, the months-long training regimen could pose some significant logistical hurdles to implementing this strategy on a grander scale, says Audrey Odom John, a parasitologist who is investigating the utility of malarial odors for medical devices at Washington University in St. Louis.

As the researchers try to implement canine-based detection systems, they’ll have cultural barriers to consider as well. In some countries, canines aren’t always seen as companions. As Lindsay and his team move their efforts forward, he explains they will prioritize introducing the dogs and their role in disease prevention to locals before proceeding with their work. But acceptance may not be universal.

Claire Guest, Chief Executive Officer of Medical Detection Dogs, with a cohort of canines at the charity's headquarters in the United Kingdom. Dogs in these teams have been trained to detect certain cancers, bouts of hypoglycemia in diabetics, and, most recently, malaria. Photo Credit: Durham University/Medical Detection Dogs/London School of Hygiene & Tropical Medicine

Because of these drawbacks, Odom John explains, it’s unlikely dogs will be a hospital staple in the near future. But other technologies are beginning to surface in the meantime, leveraging the power of medical devices also sensitive to scent. Odom John’s lab is currently investigating the use of breath detection technology in malaria diagnostics. Others, such as James Covington, an engineer at the University of Warwick in the United Kingdom, are developing electronic noses, or e-noses, to pick up subtle changes in a person’s aroma in the presence of disease—and the tech is already showing promise for tuberculosis and cancer.

These little machines certainly have their perks. For one, unlike dogs, they’re mass-producible and require no training, Bomers says, making them a potential gold standard in the clinic. But they have their own set of pitfalls: To get results, researchers need to obtain a sample, insert it into the device, and wait. Dogs, on the other hand, are proactive and spontaneous. If they catch a whiff of something, they’ll make sure you know it. And, of course, there’s still no device that beats that canine sense of smell.

“The e-noses are pretty good—about as good as our noses—but not as good as dogs’ noses,” Lindsay explains. “And dogs don’t need to be plugged in.”

Luckily, Odom John thinks that if medical technology and sniffer dogs both continue to progress, they’ll likely serve complementary purposes—one in the clinic, perhaps, and one in the field—rather than vying for diagnostic space. With nearly half the world’s population at risk of contracting malaria, there’s certainly plenty of work to go around.

“There’s such an urgent need for new diagnostics,” Odom John says. “So we need a lot of shots on goal.”

Share this article

Major funding for NOVA is provided by the David H. Koch Fund for Science, the NOVA Science Trust, the Corporation for Public Broadcasting, and PBS viewers.