Thursday, March 30, 2017

Cabin Air Systems as a Mode of Hantavirus Infection

Image of typical cabin air infestation, found online, including
mouse, built directly atop a vehicle's cabin air filter.
Notice how dry the corpse looks. 
Both the wi-fi and the coffee were good and free in the ICU where my wife had collapsed into hantavirus pulmonary syndrome (HPS), so when my reeling subsided a bit I began to do some research. First, since she was still intubated, I found a keen interest in the extubation process. I wanted to understand as much as possible to be able to communicate with the doctors and nurses (who were excellent). More on that in a future post on "ICU-itis."

Then I learned that they had tested her for hantavirus, and I took an interest in that, began to research it, and thought more carefully through our potential exposures. How could she have contracted it? Could it really be her car, which we had worried about? I removed her van's filter and checked for mouse activity. I found it – chewed fiberglass insulation, maple seed shells, and other mouse debris.

When she was first admitted we had answered "yes" to possible rodent exposure. We had previously spotted mouse nest material and other debris atop the cabin air filter in her minivan, a Toyota, Sienna, a couple years before. At that time I cleaned it out, googled it, and found rodents to be a common problem that no one seemed to have a solution for, especially for this common model vehicle.

I thought I had to screened off the entrance, and, given the extreme rarity of hantavirus, and many other things to do, I left it at that, with occasional inspections and clean outs. In retrospect, I didn't take the threat of hantavirus seriously enough.

I had been driving that same car to and from the hospital. Fortunately, I drove with the windows down and the fan at the lowest setting. Still, perhaps it was time to switch to my Honda, Accord. When I checked it for mouse debris atop the filter, it was there as well. Both our vehicles, two of the most common cars on the road, had been invaded by deer mice.

The first pair of positive hantavirus antibody test results came back from the CDC about twelve days after admission. She had been extubated a few days before, only becoming responsive a few days after that, and was slowly regaining the ability to stand up, and to speak, her mind slowly reconnecting with her body after ten days of acute disease and sedation.

She missed our dogs, two miniature Australian shepherds, stating as much in one of her first complete sentences. I found out later that she couldn't remember what they looked like. She wanted to go home, and was very clear about that as well.
Dori and Mick, our dogs from MARS,
the Miniature Australian Rescue Society.

While working on that goal, I studied contraction and incubation time of hantavirus and reviewed our relative, possible exposures in our home, two outbuildings, and vehicles. That discussion will appear in another post, but the similarities and differences in our relative exposures gave me something of a controlled experiment. The largest difference was her much greater exposure to the air and dust from her car's cabin air system.


That led me to wonder about a larger problem. Was it possible that others were contracting sin nombre hantavirus from their cabin air systems?

Although I saw ample speculation in car talk type forums, there appeared to be little to nothing about this exposure mode in the technical literature. The CDC website does not mention it. Neither does the Canadian health agencies' hantavirus information site. I have not found a single mention of mouse infestation of vehicle cabin air systems, from any governmental hantavirus site. Vehicles are mentioned, meaning places like seats an trunks, but the inner passages of the air circulation system are not.

Could my wife's case be unique, even though the mechanism seemed so obvious? Why is there no awareness of other cases of HPS caused in this way.

Which brought me to the question: Is it possible that sin nombre hantavirus infections from vehicle cabin air systems were occurring without being detected and reported?

These questions have several moving parts. Let's break them down to four related questions that should inform them.

1) Is it not only possible  to contract hantavirus through vehicle's cabin air system, but are these systems actually efficient, practical, reasonable mechanisms for conveying biologically active hantavirus virions to human lungs?

2) Are there significant numbers of cabin air infestations by hantavirus carrying deer mice?

3) Could there be a reason that hantavirus infections through cabin air systems are being misdiagnosed, and so be falling through the cracks?

4) Is there some reason that these events could have been statistically camouflaged and so gone unnoticed over the years?  (more on this in upcoming post)

In this post I will show that the answers to all four questions are either certainly or probably yes. In another post I will show that a cabin air system was the probable cause of hantavirus infection in at least one person, my wife. This is not a hypothetical thing. I believe it is probably what nearly killed my wife and I think there is a significant chance that other people may be dying this way.
When I opened the Sienna's filter to photograph for this piece, yesterday,
I found this fresh mouse debris and possible urine stain on the filter.
Sharp metal grating atop filter is a try at a mouse deterrent.

Hantavirus Infection Process

First, let's review what steps have to take place for sin nombre hantavirus to infect a human.

The Sienna filter removed. This is just a tiny amount of material.
Most infestations involve far more (see those posts). That may
be because the expanded steel grating we placed atop the filter
(see first pic) succeeded in discouraging nesting, at least. 
Specifically, what has to happen in the process of human infection from mouse-carried hantavirus through inhalation of the virus, which is widely believed to be the most common mode of infection. This process has to involve the delivery of hantavirus virions from an infected mouse to human lungs in sufficient numbers, and in good enough condition, to cause infection.

To become infected with hantavirus it is apparently necessary to inhale a significant number of the virus virions, the portable, particle-like form of the virus. One, or just a few virions will perish or be destroyed by the bodies' defenses. If the disease were that contagious it would be vastly more common.

A virus that has evolved while living in humans for many generations is likely to be well-adapted to working with human anatomy. In that case, the virus probably has a good chance of achieving a new infection in another human with only a small number of virions. Some diseases are believed to be transmitted by as few as, literally, a few virions. Even though the amount of virus discharged into a human by a mosquito is likely small, it can transmit diseases that are well adapted to their new human host.


Hantavirus, however, is transmitted to humans from rodents, its host species, the species that the virus has evolved with. Differences and similarities in hantavirus genes closely match the differences and similarities in their host species. A hantavirus that is endemic to deer mice is likely to be very similar to a hanta that is endemic to the closely-related white footed mouse, another New World mouse. Both of those hantaviruses are likely to be dissimilar to those that have evolved alongside an Old World rodent species.

Both viruses are adapted to living in rodents, not people. When they infect people they have crossed species and will be performing their genetic surgery on cells they are not as familiar with, in an evolutionary sense. People and rodents are very similar, about 99% of our genes are identical with mice. This explains a lot of things, but those things are outside the scope of this post.

The 99% similarity gives a hantavirus a good head start at coopting the genetic mechanisms of our cells, but the 1% difference means that zoonotic viruses tend to do damage and cause severe illness. It also may mean that it is more difficult for hantavirus to infect a human than yet another mouse, after thousands of years of practice. Although I'm not aware of any actual experimental data, it appears that some significant number of hantavirus virions must be inhaled for an infection to occur.

Although it is possible to become infected by a mouse bite, these cases are extremely rare. Deer mice are shy, reclusive folk. It is also possible to become infected by touching something contaminated with mouse urine or feces while cleaning, and then rubbing your eyes or touching your mouth.

But the most probable, or at least the most common mechanism of infection is believed to be inhalation of the hantavirus virions. The lungs are the primary target of the sin nombre genus of hantavirus that causes hantavirus pulmonary syndrome, HPS.

So we have a human exposure requirement that involves inhalation (for this case) of some number of active hantavirus virions.

To be inhaled, those virions have to somehow become airborne. To do that, they must be finely divided into an aerosol, a cloud of material in droplets or particles so fine that that it moves along with the air and circulates freely with it. If the particles are fine enough, they will flow with the air all the way down into the finest passages of the lungs. Smoke, clouds, mist from aerosol spray cans are all example of aerosols.

Viruses and Virions

"Virion" ought to be a common term, but it isn't. Hantavirus virions, the “portable” form of the virus, are roughly the shape of round balls that are about 160 nanometers in diameter. That’s about 6 millionths of an inch or 1/500th the diameter of a typical human hair – far too small to see with light. A typical polyethylene sandwich bag is about 25 µ in thickness, also called “one mil” meaning 1/1000th of an inch. It would take about 150 hantavirus virions to span across that 1 mil of thickness. 

Hantavirus is a single stranded, negative sense, RNA virus, designated –ssRNA in a system called the Baltimore Classification. That means that it is coded by a single strand of RNA, ribonucleic acid, that is twisted in a negative spiral direction in relation to functional groups at its ends. The RNA is coiled up tightly inside an enclosure that surrounds and protects it, and performs other functions, and that is made from proteins that are coded by the RNA. That protein coat is, in turn, surrounded by a membrane comprised of lipids, fatty molecules. 

Finally, the surface of the hantavirus virion is studded with two kinds of viral surface glycoproteins, designated Gn and Gc. These are molecular lock-picking systems, in a way, that begin the cell infection process by penetrating cell walls in the species under attack. Working as a team, the two glycoproteins deliver the virion to the cell interior, where its RNA and other parts begin a hostile take-over of the cellular processes. 

These components are manufactured inside the cells of host species after they have been conquered and repurposed by the virus. It is only when the hantavirus virion is at home inside a host cell that it really “comes alive.” While a virion is a purposeful clump of RNA and proteins that cannot reproduce itself and that has no metabolism, a cell that has been coopted by that virion is humming factory of metabolism and of reproduction – of virus components.

These components are ultimately assembled in some fashion by the same cells molecular scaffolding and equipment and at the point when the cells resources have been completely consumed it is them opened up in the end game of the virus and the new family of virions are released. When and where the virions are released and conveyed into the excreta or other shed products from the host species is of significant importance because that is part of the pathway all the virus from a whole species to the human being.

Hantavirus virions continue their journey by being expelled with the urine and feces of the mice. Mouse urine contains the salt, urea, and a surprisingly high concentration of proteins. If a human had such a high concentration they would have proteinuria, and would be very ill. The virions, less than 1/5 of a micron in diameter (160 nm) are suspended in this soup of water, urea, and proteins, and flow along with that fluid.

Electron microscope image of hantavirus virions.
An electron micrograph of hantavirus virions appears at right. The RNA that codes for the virus is coiled and coiled again into a tight mass inside the virion. This is coated with a protein shell, which is coated with a second, protective shell, and studded with 

These particles are so small that it is fairly unlikely that they will become completely separated out from other materials, that is, aerosolized down to individual virion particles. Such a fine division of material is not energetically favorable. The finer the division in a fluid, or any material, the more energy must be added from the bulk form to overcome the surface energy of that material. This is the energy that makes water draw itself into rounded drops, and that sucks liquids up into capillary tubes. 

A molecule of water is more stable, at lower energy, when it is able to stick to other water molecules all around it, on all six sides of the single molecule. This is the state of most molecules in a fluid. The ones that are on the outside surface, however, are only stuck to other molecules on five sides, with the side facing the air only very weakly linked to the gaseous air above it. When you divide bulk water down into finer and finer drops, you have relatively more and more molecules on the surface of the drop compared to those inside of it. 

If you divide the water down into individual, separated molecules, they are no longer stuck to any other molecules, and you have water in gaseous form, steam. The difference in energy between liquid water close to 100ºC (212 ºF) and gaseous water, still at 100 ºC, is enormous. It is called the heat of vaporization of water.  

Nano-divided water droplets are very unstable – they want to stick to anything, anything at all that approaches them. They will stick to oil, metal, Teflon, anything they can approach will lower their energy, so they stick to it, and tight. Even so, at elevated temperatures other molecules blast them or shake them off the surface and they return to a gas form as rapidly as they condensed out.

Something similar has to happen for virus virions to be finely separated. For a water nozzle to make finely divided mist, it has to have a large pressure difference to provide the surface energy. For a soup of virions in mouse urine (water, urea and protein) energy has to come from somewhere. For urine and feces lying in a corner of a garage, this energy simply isn’t there, and the material remains safely undisturbed until it denatures.

If that garage corner is swept with a dry broom, however, very substantial amounts of energy are released in a very localized way where the tips of the broom bristles scrape against the floor surface. This kind of scraping relates to “dry friction” and is a complex phenomenon. It is often characterized by something called “stick slip behavior,” wherein the interface between the bristle and the floor surface and debris alternately sticks, staying locked in place, then breaks free and slips, displacing some distance before enough elastic strain is released so it can stick again. Every slip means that something has broken loose somewhere, and perhaps become floating dust. 

Time is of the Essence

The virions must make the journey from the mouse to human lungs in a limited amount of time. They must still be viable virions when they complete the trip so that they can penetrate a cell wall and take over the cell's machinery to replicate themselves. The lifetimes of the virions vary with the environment they are in. Exposed to the dry air, they decay to 1/10 their number in only two hours. At that rate, 18 hours later they would be reduced to one billionth the original number, and would be unlikely to infect.

Under moist conditions, however, the virions are more durable, lasting about two days for every reduction to one tenth their former numbers. At that rate it may take weeks for them to be completely neutralized.

Most hantavirus infections have occurred under desert-like conditions, such as near the Four Corners region of Arizona, Utah, Colorado, and New Mexico. The dry, low humidity conditions in these places make it relatively easy for the virus particles to separate from the fluid that otherwise binds them into liquid material. Liquids are much more stable, in general, than gases. To make a gas out of the liquid takes a lot of energy. To make a finely divided aerosol out of liquid also takes energy.

That energy goes into creating new surface area material from material that was in the interior of the liquid. The relative amount of surface to interior increases with the fineness of the droplets. The amount of energy depends on what is called surface free energy, or surface tension.

Water has a very high surface free energy, so it is fairly difficult to make into a mist. If you add a surfactant, a detergent to that water, it lowers the surface energy and makes it easier to, say, blow bubbles or form a fine mist. The bile salts in urine act as surfactants, reducing it below that of water alone. There is still a substantial energy barrier to mouse urine aerosol formation. When expelled from the mouse, small amounts may pass into the air, a known hazard in research facilities, but most of it strikes some surface, is adsorbed onto that surface, again, by surface free energy, and then it goes nowhere.

Over time, hours, days, or at most weeks, it will degrade and be fairly harmless, as far as hantavirus is concerned, but it will still be disgusting. When an infected mouse pees in the corner of your garage, you hardly ever come down with HPS because the virus simply degrades in place.

Unless, of course, you sweep it with a broom while it is still "fresh," generating very strong forces at the tips of the bristles and bringing up dust. Dust is often difficult to see. If a bright light like a pen laser is used in a darkened room, it is surprising what little particles it will illuminate.

Or unless vacuum it.  Vacuum cleaners draw in dust and dirt, and filter out most of it, but not all. Virions are far smaller than the pores in any vacuum, even those with HEPPA filters and cyclones. The virus particles are drawn against the filter surface by the very high velocity airflow that exists within these cleaners. Most of the surface of the filter is closed, remember, so the small amount of its surface that is drawing through air is rushing at very high speed.

Both brooms and vacuum cleaners are known mechanisms for causing HPS. The physical mechanism of aerosolizing the material and bringing it to the lungs is clear. Infections caused this way, while believed to be the most common means, remain relatively rare, probably because most virus particles that are vacuumed or swept have been out of the mouse for days or weeks, and are no longer biologically active. If the urine is still moist, the virions are much less likely to be separated into airborne particles. It will simply wet the tips of the broom and smear on the floor a bit.

When most of the water in urine has evaporated, a residue of bile salts, protein bits, and hantavirus virions will be left behind. The salts are likely to make this residue hygroscopic, tending to attract and hold water to some extent.The hygroscopic mix of mostly dried urine will retain some coating of water that is likely to promote longer active life in the virions.

Moist virions are fairly durable, dry virions are more vulnerable. Most known HPS cases have been in dry climates. The Seattle cases stand out as exceptions, having happened in a very wet season in the Puget Sound region. This basic difference in vision environment outside its host suggests that a different mechanism may be favored for transmission in wet-climate hantavirus cases.

But if the air is dry enough, the attractive forces between water and mouse urine residue are overcome and the evaporating water leaves behind relatively dried material. The surfaces of the virions will now be more weakly bonded to other material than they were to the aqueous urine because of the lower surface free energy of the less polar species present. As distinct, spherical, structurally sound units, the virions could separate from the dried urine under the abrasive action of a broom bristle or a strong breeze, becoming airborne and finding their way into lungs.

But the dryness that is required means that time is on your side. The time delay between ejection from the mouse and ingestion to the lungs provides what may be the tightest bottleneck in the infection process. The vast majority of all hantavirus shed by mice is oxidized, destroyed by UV light, or otherwise degraded. Only the tiniest of fractions makes the trip to human lungs in time to do great harm.

A dry environment tends to favor both aerosolization and dust formation that can transport the virus, but, the delivery must be prompt once the material has become exposed to air and dried, or the virions will have degraded.

Cabin Air Systems


Vacuum cleaners, which can draw mouse excreta against a filter, are known mechanisms of hantavirus infection both from specific cases and because of clear, physically realistic reasons. Keep that in mind as we finally look at cabin air systems. Cabin air systems are comprised of the fan, filter, valves, outlets, controls and inlet that provide the air you breath in a vehicle such as an automobile, truck, or tractor.

These systems are distinct from the engine air filter and intake. Cabin air is drawn in through a passage that keeps it separated from the vapors in the engine compartment. Cabin air filters are typically located behind the glove compartment, and are accessed by emptying it, releasing a retaining band, squeezing in the sides and pulling it forward. The filter is removed by squeezing the clips on its sides and sliding it forward. Only do so wearing gloves and a respirator mask. 

People in "car talk" type online forums have speculated for years about the possibility of cabin air hantavirus infection as a possibility and a cause for concern. Many of them have posted images of their own invaded and inhabited cabin air systems, the filter elements having been made the living, eating, and excreting quarters of deer mice and other rodents.

Links to three hantasite.com posts with images of cabin air system rodent infestations are below. These are a small fraction of the online images put up by the fraction people that went through the trouble to photograph their mouse nest covered cabin air filters and post them online in discussion forums. These invasions are somewhere between common, and ubiquitous.

Cabin Air Infestations 1  Cabin Air Infestations 2       Cabin Air Infestations 3


So the answer to "2) Are there significant numbers of cabin air infestations by hantavirus carrying deer mice?" is yes. There must be thousands, at the very least, and possibly tens, or hundreds of thousands of cases of mouse infestations of vehicle cabin air systems. Compiling data to quantitate this is a project for someone out there.

Now consider what happens when an infected mouse builds a nest atop a cabin air filter, and the vehicle's driver enters the vehicle and starts the engine. The mice living within will be startled by the sounds of the vehicle being entered, and terrified when the car is started and the heater fan is actuated. They will be in a virtual tornado of air motion. The natural reaction is a defensive move to make themselves less attractive to predators. They will expel urine and feces. Fresh, hantavirus laced urine and feces.

As the air system operates this urine will be rapidly dried, because the heater will lower the humidity rapidly, or the AC will dehumidify it even more rapidly, and the Defrost setting will dry it fastest of all, dehumidifying the air by operating the AC compressor, and simultaneously operating the heating elements, drawing warm, desiccated air directly through the excrement and filter element.

At some point, potentially aided by holes chewed in the filter element, (see above image) the hantavirus virions will become dried and somewhat separable from the matrix of bile solids left behind by the dried urine, and similarly for virions within the feces. T

There is a clear mechanism for cabin air systems rapidly drying, and rapidly aerosolizing fresh, active, virion material and directly passing it to the vehicle's interior and passenger lungs.

The answer to question (1) above is yes. "1) Is it not only possible  to contract hantavirus through vehicle's cabin air system, but are these systems actually efficient, practical, reasonable mechanisms for conveying biologically active hantavirus virions to human lungs?"

I will get to (3) and (4) in an upcoming post.

Hantavirus infection via cabin air systems is a problem that should be much more seriously investigated and responded to.

The above analysis might be regarded as speculative, or of theoretical nature, a theory constructed to explain something that has not been observed. But we also have a specific case.

The reasons why my wife and I both suspect that my wife contracted hantavirus through exposure from her vehicle's cabin air system will be explained in more detail in a future post on cabin air systems. 

– Mark@hantasite.com

* Throughout this post I may use "hanta" as a shorthand that specifically means the sin nombre strain of hantavirus.

This is one of the strings of comments that is coming to me regarding cabin air infestations. This one was through Facebook. Other cases of this  clearly show that this is a widespread problem in rural King County.

Kathleen   "Mark, thank you for sharing your experience and research. We live in in the woods north of Issaquah and have been battling a deer mouse invasion in our vehicles. After viewing this video and your site, we found contaminated cabin air filters in both our Toyotas which is terribly unnerving. Thanks for trying to sound the alarm."

Mark   "Thank you, Kathleen. I would like to share your comment with the King Co. health folks...."

Kathleen       "Yes, you can share it. We have our vehicles serviced regularly. We thought we had done a good job of cleaning up by having the interior steam cleaned, wiping surfaces with 10% bleach and placing rodent repellent packs in strategic places. We never thought to check the cabin air filters. We took our cars to the dealership today. They offered us an ozone treatment and sprayed a scent deterrent in vents, but told us there was no way of blocking rodent entrance to the ventilation. They suggested we stay vigilant with trapping and to replace scent repellent regularly. We farmers in the Snoqualmie Valley have noticed a big increase in mice. Despite the fact that we keep all poultry feed in metal cans, have barn cats on duty, and trap regularly, we can't keep up the this years population explosion. I wish there was something I could do to know that the virus is not incubating in my kids right now. I sent a comment already to King County Public Health about our cabin air filter being heavily contaminated."

5 comments:

A Wong said...

Thank you for sharing your experiences. I stumbled upon your site after finding a nest in my toyota FJ earlier in march. I've been at a lose as to how to clean the vents to the point I'd feel comfortable using the ac system again. So far I've had a stainless steel 1/8 in mesh placed near cabin air intake. Only other thing I can think of is decrease the general population size by trapping. However I saw a study that indicated targeting just one area might actually increase exposure risk as the vacancy will promote localized migration from nearby populations.

Anonymous said...

Great post!

John K. said...

Thanks for your posts. My wife and I are in the process of eliminating a longstanding mice infestation from our crawl space. It's been 4.5 weeks since the insulation and vapor barrier were removed, and nobody has become ill. Certainly the three workers who did the work were most at risk, though the dust and musty smell in the house afterwards were of significant concern. There were a few deer mice found underneath, something new for us, though they appeared to be a minority.

Just when I was starting to feel relieved and confident, I went out onto the front porch with my small dog and sat in the wicker chair. I live in northwestern Oregon, and this was during the recent hot spell last week, record high temperatures. I went in after about ten minutes, and only then did I notice 10-15 droppings under where the foot stool was. It also occurred to me that the large yellow spot under the rear corner of the chair may in fact be accumulated mouse urine, though I can't understand why this would be so. So now I'm back to being a bit worried about it, not knowing the chances of dried urine on a sunny afternoon, with a mild breeze blowing, wafting into my lungs in ten minutes time. I didn't disturb the area, other than to sit down. I did spray the feces with chlorine, but not the urine, since it didn't click that this might be from mice. I read this part of your post carefully.

"But if the air is dry enough, the attractive forces between water and mouse urine residue are overcome and the evaporating water leaves behind relatively dried material. The surfaces of the virions will now be more weakly bonded to other material than they were to the aqueous urine because of the lower surface free energy of the less polar species present. As distinct, spherical, structurally sound units, the virions could separate from the dried urine under the abrasive action of a broom bristle or a strong breeze, becoming airborne and finding their way into lungs.

But the dryness that is required means that time is on your side. The time delay between ejection from the mouse and ingestion to the lungs provides what may be the tightest bottleneck in the infection process. The vast majority of all hantavirus shed by mice is oxidized, destroyed by UV light, or otherwise degraded. Only the tiniest of fractions makes the trip to human lungs in time to do great harm."

Mostly looking for some encouraging words. It doesn't seem like I would be at high risk, since it had been sunny for 2-3 days, with the sun hitting the urine spot for at least 2-4 hours each day.

Thanks for your blog. I hope you and your wife are doing well.

Unknown said...

Replying to A. Wong above, screening out the mice is the surest way, and that's something that should be explored better in car forums for each problem model of vehicle. Toyotas are notorious, and difficult to figure out where to put the screen. 1/8" stainless sounds perfect.

Keeping the mouse population down makes sense if you aren't just trying to reduce their numbers in nature, which isn't practical or wise. Barriers and trapping within those barriers, along with frequent checks for mice until certain they are being effectively excluded are approaches that work.

That said, during a population explosion of deer mice, such as what happened in Western Washington in the Fall/Winter of 2016/2017, and there is an infestation problem that covers many thousands of square mile, not just your garage. The next time that happens, I will give notice in advance based on maple seed production. Watching that already.

Unknown said...

Replying to John K. above. If you didn't raise dust and if time had passed between mouse emission and encountering their material then the odds are pretty much zero that there is any risk. It takes some direct exposure of the active, fresh virus to contract hantavirus.

I will post on hantavirus cleanup soon, as I've recently learned that neither health officials, exterminators, or hazmat folks have clear ideas about how to proceed with that. I've developed some fairly efficient cleaning techniques, and procedures that minimize risks without making things ridiculous to do.