Does bestiality increase your risk of penile cancer?

Aah, the things one learns when awake at 3AM on a Saturday night. Via a few different Tweeps, I ran across this article from Men’s Health magazine, titled “Urgent Warning: Sex with Animals Causes Cancer.”

I probably should have just stopped there.

But no, I read the magazine article, which states:

Brazilian researchers polled nearly 500 men from a dozen cities, and found that–we’re not joking around here–roughly 35 percent of the men had “made it” with an animal. That’s a problem, because screwing a horse, donkey, pig, or any other animal was found to up your likelihood of developing cancers of the penis by 42 percent.

Of course, this meant that now, I had to go dig up the actual journal manuscript. Though nothing is cited by Men’s Health, a quick PubMed search using the terms “sex with animals” and “Brazil” turned up Sex with Animals (SWA): Behavioral Characteristics and Possible Association with Penile Cancer. A Multicenter Study, published last month in The Journal of Sexual Medicine.

Though the MH write-up makes the research sound ridiculous, it’s not a bad paper overall. Starting out with the observation that penile cancer is common in impoverished regions in the world but relatively rare in developed areas, the authors wanted to examine one possible difference in this urban/rural divide: bestiality. So they enrolled 492 individuals who had spent their childhood in rural areas: 118 cases who had penile cancers and 374 controls who were seen at the same clinics for other issues, including check-ups and “cancer prevention” (though it’s not really defined what’s included in that catch-all). All participants were asked a variety of questions about their sexual history, including sex with animals and humans (frequency, number of partners, the usual drill), circumcision status, as well as other factors that might influence cancer outcomes, such as smoking status and history of sexually transmitted diseases and other health conditions.

The authors did find in the univariate analysis (basically, looking at one factor at a time) that there were several statistically significant differences between the cancer group and the control group. These included smoking, a history of sex with prostitutes, the presence of penile premalignant lesions (not surprising) and phimosis (NSFW), a condition where “the foreskin cannot be fully retracted over the glans penis.” As the title suggests, they also found that having sex with animals was significantly higher in the case than the control group (44.0 vs 31.6 percent, p<.008). When they combined risk factors into their multivariate analysis, a few factors still remained in the model. Phimosis was the big one, with an odds ratio of 10.41; SWA was down the list at 2.07 (95% CI: 1.21-3.52, p=0.007). Penile premalignant lesions and smoking also remained, with odds ratios in the middle of the other two. Finally, just because I know many of you out there are curious, they also break down those who have SWA by types of animals they, um, frequent:

The animal types most often cited were mares (N = 80), followed by donkeys (N = 73), mules (N = 57), goats (N = 54), chickens (N = 27), calves (N = 18), cows (N = 13), dogs (N = 10), sheep (N = 10), pigs (N = 6), and other species (N = 3).

Yes, chickens for 27 of them. I don’t even want to know, but I’m sure if I did, I could find out somewhere on the Internets. Please, don’t educate me on that one. They also note that almost a third of the men reported “SWA with a group of men.” I’m leaving that one alone as well (especially as that one wasn’t any different between cases and controls, so it didn’t seem to be an important variable for penile cancer development).

So how do they explain these findings? Their discussion is a bit odd, in my opinion, and narrows in on the SWA finding to the exclusion of their other significant risk factors. Of course, coming from my background, my first thought regarding SWA and cancer jumps to infectious agents. They acknowledge in the introduction that the human papillomavirus (HPV) is associated with about half of penile cancers. Other species of animals can also be infected with papillomaviruses, such as the rabbit of jackalope mythology. A previous study identified five potentially novel papillomaviruses in Australia, just by doing skin swabbing. As such, it’s certainly safe to say that we know very little about the diversity of these viruses that exist in other animal species, much less their cancer-causing potential. It would be fascinating to look at tumor samples from the men in this group who were known to have sex with animals, and see if any novel viruses (papillomas or otherwise) could be identified.

However, they don’t limit their suggestion to only zoonotic infections. That’s when it gets a bit weird to me, as they say things like:

Speculation exists regarding cancer status as an infectious disease in humans [24,25], as studies have suggested that tumor cells can be transmitted from one mammal host to another within the same species [26,27]. PC is frequent in equines [28], but transmission of malignancies between animals and humans has not been reported.Virology does not consider possible viral movement from animals to humans except in cases of zoonosis, such as rabies or pandemic forms of bird or swine flu. However, the hypothesis that the HIV epidemic resulted from simian-human virus transmission has not been fully explored.

Um, huh? First, the citation they use for the HIV claim is from 1999–indeed, at that point there was still a lot that was unknown about cross-species HIV transmission, but that was 12 years ago! The field has moved on since then. I’m baffled as to what they mean by their first sentence–as far as I know, “Virology” doesn’t consider anything–“Virologists” do, and why would this not be a zoonosis? Though I think direct transmission of cancer cells (like in the case of the Tasmanian devil transmissible cancer) would be unlikely, transmission of microbes which could lead to cancer development is certainly plausible and well within the realm of virology/bacteriology/etc. In my opinion, it’s infinitely more likely than the idea they also suggest of more directly carcinogenic animal secretions.

There were also a number of limitations in the paper. Though they grouped frequency of sex with prostitutes into a “more/less than ten times” dichotomous variable, I don’t see any similar “dose” analysis for the frequency of SWA in their models, even though they did ask the men about this. They make one statement that “long-term SWA (>3 years) was reported by 64% of the PC patients and 46.6% of the controls (P = 0.044).” This difference was statistically significant at the usual cutoff (p< .05), but it doesn't appear that they studied this further--why not? If you have a typical dose-response relationship (the more times the men had sex with animals, the more likely they were to develop cancer in the future), that would strengthen their case for a connection between the two. They also didn't ask about sexual orientation or the nature of the self-reported past STDs. Are any of these participants HIV positive, for example? Anyway, with these limitations in mind, it does appear that Men's Health got it mostly right: don't have sex with animals if you value your penis. But it's unfortunate that they just go for the sensationalism and ignore the more important variables from a public health standpoint, like "don't smoke" and "if you have abnormal penile conditions, you may want to get those checked out, k?" References

Zequi SD, Guimarães GC, da Fonseca FP, Ferreira U, de Matheus WE, Reis LO, Aita GA, Glina S, Fanni VS, Perez MD, Guidoni LR, Ortiz V, Nogueira L, de Almeida Rocha LC, Cuck G, da Costa WH, Moniz RR, Dantas Jr JH, Soares FA, & Lopes A (2011). Sex with Animals (SWA): Behavioral Characteristics and Possible Association with Penile Cancer. A Multicenter Study. The journal of sexual medicine PMID: 22023719

Antonsson and McMillan, 2006. Papillomavirus in healthy skin of Australian animals.

MRSA found in Iowa meat

I’ve blogged previously on a few U.S. studies which investigated methicillin-resistant Staphylococcus aureus in raw meat products (including chicken, beef, turkey, and pork). This isn’t just a casual observation as one who eats food–I follow this area closely as we also have done our own pair of food sampling investigations here in Iowa, and will be doing a much larger, USDA-funded investigation of the issue over the next 5 years.

Let me sum up where the field currently stands. There have been a number of studies looking at S. aureus on raw meat products, carried out both here in North American and in Europe. In a study from the Netherlands, a large percentage of samples were found to harbor MRSA (11.9% overall, but it varied by meat type–35.3% of turkey samples were positive, for example). Most of there were a type called ST398, the “livestock” strain. This was also found in one Canadian study (5.5% MRSA prevalence, and 32% of those were ST398), but no ST398 were found in a second study by the same group.

Here in the US, prevalence has found to be lower than in that Dutch study (from no MRSA found, up to 5% of samples positive). Furthermore, in the previously-published studies, no MRSA ST398 was found in samples of US meat, though this paper did find plenty of methicillin-sensitive S. aureus (MSSA) ST398 strains. Instead, most of the MRSA isolates have been seemingly “human” MRSA types, like USA100 (a common hospital-associated strain) and USA300 (a leading community-acquired strain).

Why am I rehashing all of this? We have a new paper out examining S. aureus in Iowa meats–and did find for the first time MRSA ST398, as well as MRSA USA300 and MSSA strains including both presumptive “human” and “animal” types. This was just a pilot study and numbers are still fairly small, but enough to say that yes, this is here in the heart of flyover country as well as in the other areas already examined.

As I mentioned, this is one of two studies we’ve completed examining MRSA on meat; the other is still under review and much more controversial, but I will share that as soon as I’m able. And with the USDA grant, we’ll be working on better understanding the role that contaminated meats play in the epidemiology and transmission of S. aureus for the next several years, so expect to see more posts on this topic…


Hanson et al. Prevalence of Staphylococcus aureus and methicillin-resistant Staphylococcus aureus (MRSA) on retail meat in Iowa. J Infect Public Health. 2011 Sep;4(4):169-74. Link.

Waters et al. Multidrug-Resistant Staphylococcus aureus in US Meat and Poultry . Clin Infect Dis. 2011 May;52(10):1227-30. Link.

Weese et al. Methicillin-resistant Staphylococcus aureus (MRSA) contamination of retail pork. Can Vet J. 2010 July; 51(7): 749-752. Link.

De Boer et al. Prevalence of methicillin-resistant Staphylococcus aureus in meat. Int J Food Microbiol. 2009 Aug 31;134(1-2):52-6. Link.

Pu et al. Isolation and characterization of methicillin-resistant Staphylococcus aureus strains from Louisiana retail meats. Appl Environ Microbiol. 2009 Jan;75(1):265-7. Link.

Bhargava et al. Methicillin-resistant Staphylococcus aureus in retail meat, Detroit, Michigan, USA. Emerg Infect Dis. 2011 Jun;17(6):1135-7. Link.

Epidemiology and social media: conference fail

I have written and deleted this post. Twice. But damn it, it needs to be said.

I’m here in charming Montreal for the North American Congress of Epidemiology. It’s a good-sized meeting, as far as epi meetings go. The site notes that it’s a joint effort between four major Epi organizations: The American College of Epidemiology (ACE); The Society for Epidemiologic Research; the Epi section of the American Public Health Association, and The Canadian Society for Epidemiology and Biostatistics. Collectively, those associations represent a lot of epidemiologists.

The conference started off well. The first night kicked off with a movie about bioterrorism preparedness followed by a panel discussion. Great–movies! Engaging public in novel ways! Love.

On to Wednesday, when the first real sessions begin. The opening plenary discussed Science, skepticism, and society. Great again–this is a perfect warm-up. Later that afternoon, there was another session titled “Communicating Epidemiology: The Changing Landscape”. I was happily surprised when the room for this was pretty packed, as these types of meetings tend to be heavy on chronic disease epi and epidemiology methods. However, I was disappointed with the content. While the first talk was to give “a snapshot of how premier science journals experiment with features that blur old distinctions: blogs, data repositories, standard-setting, and advance online publications,” almost none of that was discussed–instead, it focused on how Nature Genetics was doing…something….about datasets. (Unfortunately I don’t have great notes and was at this point still trying to get the wifi to work; more on that later). Either way, it wasn’t anything as advertised in the description I quoted above, and it discussed *only* Nature Genetics–surely there are more “premier science journals” than just NG? (Why only NG discussed? The talk was by Myles Axton, who is the Editor).

Next on tap was Jennifer Loukissas, communications manager at the National Cancer Institute’s Division of Cancer Epidemiology and Genetics, to discuss “When epidemiologists talk to press and public.” However, there really wasn’t any “public” involved–it was a media training session. Period. Use soundbites, stay on target, think about your message, control the interview, call the journalist back in 5 minutes if you need to collect yourself, etc. Good stuff for scientists to know, to be sure, but isn’t there a world out there beyond talking to journalists? More on that later as well.

The third talk was Jonathan Samet of USC, on “Communicating around conflict.” He’s recently worked on the WHO cell phone-and-cancer opinion that was released earlier this month, and essentially extended what Loukissas began as far as what to and not to say to interviewers, particularly in controverisal areas.

This was not exactly my idea of science communication in 2011, especially since everyone agreed at the beginning of the session that scientists were terrible communicators, our messages frequently ended up getting distorted, all the typical canards. Merely telling scientists to stop being so jargon-y and prepare soundbites–while necessary–isn’t going to solve these problems.

During the (very brief) Q&A, I asked about scientists directly communicating with the public–via their websites, blogs, web video, etc., to get their own message out there and not have to worry about journalists messing it up. Loukissas was the main one to answer the question, saying–incredibly–she hadn’t thought about that.

It was all I could do to keep myself from saying “d’oh!,” complete with facepalm.

The reality is that scientists don’t have to be passive any longer, relying only on reporters to translate their work for them in order to send it along to the public. We should have our soundbites, but realize that we can go beyond our manuscripts (I’ve had ones recently trimmed down to 1200 or even 800(!) words). We can write about the research if it’s behind a paywall. We can write about the realities of doing our work as a jumping-off point after a journalist covers your research, and go beyond the dry data that goes into the paper. We can go beyond the press release and talk about what may be interesting to us about our findings, but maybe aren’t the “meat” of the publication, or are secondary to the “main point” that you’ve worked on for your soundbites and want to emphasize to interested journalists. We can elaborate on interesting research done by others, to discuss subtleties that you can’t fit into a 20-minute interview.

And more.

Communication-wise, this meeting has unfortunately been a bit of a letdown. The science is interesting and there have been some great speakers, but I haven’t been able to share much of that because wifi wasn’t arranged for in the conference rooms. I have internet in my room ($14.95/day, of course), but the password from my hotel room isn’t valid downstairs (something it took me almost a day and a half to find out, after getting the run-around from various people), and the organizers either didn’t care, didn’t think, or couldn’t afford to allow attendees to use the wifi network in the hotel conference rooms. So while I was able to take conference notes from the American Society for Microbiology meeting right on Twitter and share them with everyone via the conference hashtag, not so for this meeting. (ASM even had their own mobile app for smartphones).

The thing is, *epidemiologists need to be plugged into these kinds of things.* So many of the studies reported in the media have to do with epidemiological topics–cell phones and cancer, vaccines and autism, “chronic” lyme, does the internet really give you “popcorn brain”, just to take one current story from CNN. We can’t sit in our towers and just wait for a journalist to call us about those studies anymore–and why should we?

Last year, Craig McClain wrote about why scientists need to use Twitter. That post, while good, focuses mainly on the benefit to the scientist (though he does note that the public can also find information there). We need more of this. We need good, reliable information to be out there on the internet, freely available–and if that’s not possible in your academic publishing model, it’s still possible with a blog, or YouTube video, or basic website that you keep updated with recent news. Epidemiologists are certainly using social media and Google to explore disease; why not give back by wading out there and actually taking part in the conversation?

Epidemiological studies–why don’t people participate?

Maryn McKenna was awesome enough to take some time out of her vacation to blog about our recent ST398 paper, finding “livestock-associated” S. aureus in a daycare worker. She raised one question I didn’t really address previously, regarding our participation by kids and workers at the facility (eight kids out of 168, and 24 out of 60 staff members).

(Staph screening is very non-invasive, by the way; it effectively involves twirling a long-handled Q-tip inside the front of your nostrils. Kinda makes you wonder why families would not have wanted to participate. On the other hand, since Iowa is the pig-growing capital of the U.S., they may have been motivated not to want to know.)

I thought I’d chat a bit about enrollment for this project, since getting people to participate is one of the most difficult parts of these types of studies. First, there really wasn’t any mention of MRSA and swine for this particular study, so I doubt protecting the pig industry was high on anyone’s list for reasons not to participate. However, anytime we do these type of studies, we’re relying on the generosity of individuals in the community–particularly when we didn’t really have participation incentives, as was the case in this project, which was done on a shoestring budget. (We passed out mini hand sanitizer bottles for adults, and had some little toys for the children).

We ran into several challenges for the research which limited our ability to enroll children. Along with a swab, we also had a questionnaire for parents and employees to fill out (as well as a third questionnaire for the director of the facility). For parents and employees, we asked about exposures: did they spend time in hospitals, around animals, at the gym? Had they recently had an infection? etc. For the directors, we asked about cleaning routines at the facility, as well as facility size (number of children and employees). So it wasn’t only the swabs, but also a decent amount of paperwork to fill out when you include the informed consent forms. We also had to do all of this at the facility; because of the way we were sampling, parents didn’t have a chance to take the questionnaire home to fill it out and then return it. So only parents (and employees) who had some spare time during either child drop-off or pick-up really had the chance to participate.

This particular study also started in roughly March 2009–right around the same time as the emergence of novel H1N1. There was a lot of news about the swabs that were taken to test for flu, which are more invasive than regular Staph swabs, so perhaps many potential participants had the mistaken assumption that the swab collection would be more uncomfortable than it really is. (When we were able to swab the child participants, most of them giggled and said that the swab tickled).

Finally, I should note that this facility was one of the larger ones we sampled, and to do this, my grad student returned several times during the day to try and catch parents during common drop-off/pick-up times (and employees who worked different shifts). However, even with this, we certainly missed a number of children and employees, such as those who were part-time and simply didn’t attend or work the day that we were there. We did have higher participation rates at some of the other facilities.

So, I think timing and misinformation–rather than any kind of fear of finding out things they might not want to know–led to our lower participation rate at this facility.

MRSA and pets: should they get tested?

Over at the Worms and Germs blog, Scott Weese has a great post on MRSA testing. He notes the he’s frequently asked by human MRSA patients whether their pet should be tested as well, since several studies have documented transmission of MRSA between humans and their companion animals. His first response is always, “why?”

One big question I ask is ‘why do you want to know and what would you do with the results?’ Sometimes people want to know their pet’s status to see if the pet was the source of their infection. However, MRSA in pets is typically human-associated, and if a pet is carrying MRSA, it probably got it from the owner. Finding MRSA in a pet after someone was diagnosed with an MRSA infection doesn’t mean the pet was the source. More likely, the person got MRSA somewhere else and passed it on to their pet.

Sometimes, people want to know if their pet is at risk of an infection. Carrying MRSA presumably increases the risk of an MRSA infection, but likely only in animals already at risk of an infection because of underlying disease or other risk factors such as surgery. The risk to the average pet from short-term MRSA colonization is probably limited. Also, if the pet was identified as a carrier, we wouldn’t be doing anything to eliminate carriage, since we have no idea if we can do it and it doesn’t seem to be needed (since dogs and cats almost always get rid of it on their own). Therefore, it’s hard to justify screening for this reason. If the animal was getting ready to undergo surgery, then that might change my answer.

I think Scott makes great points. As he notes later in his post, his advice would be the same whether an animal was found to be colonized or not (since even if the animal is negative, that only means they were *negative at the time of the testing,* and they could become colonized later: practice good hygiene.

I’d add one addendum to his response, though. As he notes, his answer may change if the animal may be facing surgery. I’d add also from the human side that the carriage status of pet animals may be something important to know if there are recurring infections in the family. If this is an issue, then the entire family will need to be looked at as a unit, to see if one of them is a reservoir (potentially re-infecting others in the family)–and the family unit should include pets at this point. As Scott notes, typically pets become colonized from their owners, but once colonized, they do have the potential to pass it back to the other human family members like the world’s worst game of hot potato. As such, they may keep the MRSA transmission chain going, even if their humans are subjected to decolonization measures (as was described in this NEJM paper).

So, in special cases, it may be helpful to have your animals tested for MRSA–but for the vast majority of people, having that knowledge won’t do much one way or the other.

Danger, Will Robinson? Safety in scientific field work

Interesting discussion over at The Spandrel Shop and Cackle of Rad on doing field work in the sciences–and the potential dangers that might be encountered. Now, Prof-like Substance and Cackle of Rad are discussing field work along the lines of biological sample collection, sometimes in the middle of nowhere, which isn’t something I’ve ever done. However, we have our own issues when carrying out our epidemiological field sampling; more after the jump.
Continue reading “Danger, Will Robinson? Safety in scientific field work”

Is There a Viral Cause for Idiopathic Pulmonary Arterial Hypertension?

Student guest post Dayna Groskreutz

Pulmonary hypertension (PH) refers to a condition in which there is high blood pressure in the vessels carrying blood from the heart to the lungs. Pulmonary arterial hypertension (PAH) is a subset of PH referring specifically to an increase in the pressure within the pulmonary arteries (rather than the pulmonary veins or capillaries). The high blood pressure in the vessels causes thickening of these arteries, making it hard for the heart to pump blood to the lungs. Pressure builds up and backs up. Over time, stress on the heart causes it to enlarge, and it becomes more difficult for blood to get to the lungs so that it can get oxygen. Patients become tired, dizzy, and short of breath. Their quality of life is significantly reduced. Data from the National Institute of Health PAH registry in the 1980s concluded that the average survival of untreated PAH is 2.8 years from diagnosis, with 1-year, 3-year, and 5-year survival rates of 68%, 48%, and 34%, respectively. With the development of effective therapy over the last 30 years, survival has improved slightly.

In some cases, PH is caused by an underlying disease, such as sleep apnea, lung disease, or heart disease. A familial form of PH has been described and characterized. PH caused by diet medications like Fen-Phen has been widely publicized. One infectious agent, human immunodeficiency virus (HIV), has been shown to be an independent risk factor for pulmonary hypertension, but neither the virus nor its proteins have been demonstrated in the pulmonary arteries. In many cases, the cause of PH is idiopathic, meaning we do not know why the patient has pulmonary hypertension. This lack of knowledge has led researchers to search for an infectious agent as a cause for idiopathic pulmonary arterial hypertension (IPAH).

Although a causal relationship between IPAH and a viral infection has not been established, a relationship is suspected. Human herpesvirus-8 (HHV-8) is the causative agent of Kaposi’s sarcoma, primary effusion lymphoma, and Castleman’s disease, a rare blood disorder. In 2003, Bull and colleagues reported HHV-8 infection in the lung tissue and in the cells of the pulmonary artery of a patient with PH and Castleman’s disease. They suggested that HHV-8 might be a causative agent for this patient’s PH. The same year this group published a case-control study in the New England Journal of Medicine. The cases consisted of 16 patients with IPAH, and the controls consisted of 14 patients with PH caused by an underlying disease (or secondary PH). They detected HHV-8 infection using both antibody and polymerase chain reaction (PCR)-based techniques. 10 of 16 patients (62%) with IPAH had HHV-8 detected with antibody and PCR techniques, while none of the control (secondary) PH group had HHV-8 detected with antibody, and one patient had PCR evidence of virus. This study provided evidence of HHV-8 infection in the lung and pulmonary arterial cells of patients with IPAH; however, the study did not provide evidence of causation as it was not prospective in design.

A subsequent study by Laney et al compared 19 patients with IPAH, 29 patients with secondary PH, and 150 controls, and looked for evidence of HHV-8 in their blood using serologic tests. The rate of HHV-8 in the blood of IPAH was 0%, controls 0.7%, and secondary PH 10.3%. Two of the three secondary PH patients with HHV-8 in their blood had HIV-associated PH, and the association of HIV and HHV-8 is well documented. The authors concluded that HHV-8 does not have a role in IPAH or non-HIV-associated PH.

Nicastri et al next retrospectively analyzed data from 75 patients referred to their institution for lung transplant. 16 had IPAH, 17 had secondary PH, 7 had PH due to repetitive blood clots in the lung, and the remaining 10 had PH associated with miscellaneous other diseases including autoimmune disease and HIV. The 42 patients without PH consisted of patients with cystic fibrosis and other lung diseases. They performed antibody tests to detect HHV-8 in the blood. Of the patients with PH, 3% had HHV-8 detected in their blood, while 19% of patients without PH had HHV-8 detected. The authors concluded there was no direct relationship between HHV-8 infection and PH.

Finally, a German study by Henke-Gendo et al examined lung tissue from 26 patients who underwent lung transplant for IPAH from 1993-2003. Using an antibody test, they detected HHV-8 protein in the diseased lungs removed at the time of transplant in 61.5% of the cases; however, they were unable to confirm HHV-8 infection by PCR in all cases. They concluded that HHV-8 is unlikely to play a role in the pathogenesis of IPAH.

In recent years, there has been a search for a causative infectious agent for idiopathic pulmonary arterial hypertension. Two papers published by the same group at the University of Colorado provided some evidence that an association might exist, but these findings have not been confirmed in three subsequent studies by other investigators. The original authors at the University of Colorado recently published a cell-based study showing that HHV-8 can infect pulmonary endothelial cells, or the cells that make up the pulmonary arteries, lending further plausibility to the association

However, in absence of further evidence at this time, HHV-8 and PH appears to be an inconclusively proven association.


1. D’Alonzo, G. E., Barst, R. J., Ayres, S. M., Bergofsky, E. H., Brundage, B. H., Detre, K. M., Fishman, A. P., Goldring, R. M., Groves, B. M., Kernis, J. T., and et al. (1991) Ann Intern Med 115, 343-349

2. Keogh, A., McNeil, K., Williams, T. J., Gabbay, E., Proudman, S., Weintraub, R. G., Wlodarczyk, J., and Dalton, B. (2009) Intern Med J

3. Bull, T. M., Cool, C. D., Serls, A. E., Rai, P. R., Parr, J., Neid, J. M., Geraci, M. W., Campbell, T. B., Voelkel, N. F., and Badesch, D. B. (2003) Eur Respir J 22, 403-407

4. Cool, C. D., Rai, P. R., Yeager, M. E., Hernandez-Saavedra, D., Serls, A. E., Bull, T. M., Geraci, M. W., Brown, K. K., Routes, J. M., Tuder, R. M., and Voelkel, N. F. (2003) N Engl J Med 349, 1113-1122

5. Laney, A. S., De Marco, T., Peters, J. S., Malloy, M., Teehankee, C., Moore, P. S., and Chang, Y. (2005) Chest 127, 762-767

6. Nicastri, E., Vizza, C. D., Carletti, F., Cicalini, S., Badagliacca, R., Poscia, R., Ippolito, G., Fedele, F., and Petrosillo, N. (2005) Emerg Infect Dis 11, 1480-1482

7. Henke-Gendo, C., Mengel, M., Hoeper, M. M., Alkharsah, K., and Schulz, T. F. (2005) Am J Respir Crit Care Med 172, 1581-1585

8. Bull, T. M., Meadows, C. A., Coldren, C. D., Moore, M., Sotto-Santiago, S. M., Nana-Sinkam, S. P., Campbell, T. B., and Geraci, M. W. (2008) Am J Respir Cell Mol Biol 39, 706-716

Can your pet dog make you sick? Multiple Sclerosis and Canine Distemper Virus

Student guest post by Raj Nair.

Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease that affects the central nervous system (CNS) consisting of the brain and the spinal cord [1]. It is thought to be an autoimmune disease since individual’s immune system attacks their own healthy tissues [1]. However, studies to ascertain triggering factors such as genetic, environmental, and infectious causes are still in progress [2]. So one wonders “Who is more susceptible to develop MS” Literature reveals that typically people between 20 and 50 years of age are commonly diagnosed with MS, affects more women than men, and Caucasians of Northern European ancestry are more prone to develop this disease [1, 2]. Knowledge on the pathophysiology of this disease is that immune system attacks myelin, which forms a protective coat surrounding the nerve fibers of the brain and the spinal cord [2]. The myelin sheath can be compared to insulation around an electrical wire. Loss of this protective layer impedes transmission of nerve signals [1]. Consequences of this damaged connection are the spectrum of symptoms seen in MS. Some of these symptoms are blurred vision, loss of balance, poor coordination, extreme fatigue, tremors, loss of sensation or odd sensations (pin and needle sensations), slurred speech, blindness, difficulty concentrating, poor memory and judgment, and in severe cases paralysis [1]. However, every person is wired differently and so these symptoms are not consistently seen in all patients with MS [1]. Considering all of the above facts my guess is symptoms may vary depending on a person’s immune system and the external or other internal factors governing their immune system. The disease is rarely fatal and most of the people are only mildly affected. Moreover, most of the affected people remit spontaneously [3].

Why should anyone care…. Because MS is unpredictable [1,3], there is no universal cure for the disease [3], can be a chronic condition [1], possibility of disease recurrence [4], and the most important being ‘a single cause’ for the disease has not been identified. As in all diseases with multiple interacting causes, in MS too there is no single pathogen or environment to complete its disease triad. Evidence has it that the disease is more common in Northern America and Canada demonstrating a north-south gradient [5]. Migration studies have established that risk for acquiring MS remains unchanged for those who move from a high prevalence area after age 15, while risk decreased for those who moved at an earlier age [5]. In addition, the genetic angle has been studied by conducting twin studies and studies on specific types of genes. Results yielded prove that genetics can lead to an increase in MS susceptibility but probably not cause MS [5]. In order to make more sense of all the above susceptibility factors and with my interest in infectious causes of diseases, I decided to probe into existing infectious perspectives on MS.

History has it that in 1868, Jean Marie Charcot described the first human demyelinating disease, Multiple Sclerosis. It was postulated then that the disease was a result of exposure to dampness or injuries or emotional stress. However, in the era of microbiological advances, one of Charcot’s students postulated an infectious etiology for MS [7]. Moreover, the CNS pathology and presence of IgG antibodies and oligoclonal bands are known to be consistent with an infectious or immune mediated neurological disorder [6]. Several infectious agents such as Epstein Barr virus, Canine Distemper virus (CDV), measles virus, Chlamydia pneumoniae, Varicella, Human Herpesvirus-6 (HHV-6), and mumps virus have been associated with MS. Viruses win hands down against bacteria in having a strong association with MS. Studies have a tilt towards a viral cause of MS due to the following reasons: low concordance of MS in monozygotic twins similar to what was seen in paralytic poliomyelitis (also a viral infection), spontaneous viral models of CNS demyelination, and increased titers of viral antibodies in MS patients (particularly measles virus). However, these associations can only be strengthened using criteria such as consistency of association across studies, biological plausibility, temporal association, specificity and dose -response relation (epidemiologists know these are the Bradford-Hills criteria!) I will briefly attempt to establish the causal role of CDV in development of MS. Reason I chose this virus? I lost my pet dog to Old Dog Encephalitis (ODE) due to chronic CDV infection. Now I am left thinking ‘Am I or any of my family member’s ideal candidates for developing MS later in life?”

CDV is an RNA virus belonging to the family of Paramyxoviridiae, is closely related to the measles virus in humans and is the most neurotropic form of morbillivirus. As observed in the measles virus, CDV can jump species [5] and causes fatal CNS demyelination in animals including primates [9]. However, the catch-22 is that there has not been one virus (measles or CDV) consistently detected in samples from MS patients to prove its causal role. To make things worse, there is a possibility of cross-reaction in testing for CDV and measles virus using molecular techniques in samples obtained from MS patients [10, 13]. Neutralization assay used to identify viral antibodies in patients have shown considerable variation in the CDV/measles antibodies ratio [10, 11]. This implies that there is a potential for CDV to produce undiagnosed or subclinical human infections [10]. To explain MS on the basis of owning dogs per se, several studies have observed that significantly higher proportion of dogs were kept indoors in the colder northern United States as compared to the southern and western region [12]. This may explain the north-south gradient noted in the prevalence of MS. So logically, greater exposure to dogs before onset of neurological symptoms was expected. However, this phenomenon could not be studied well using case-control studies owing to the higher exposure of humans to dogs in Western countries particularly the United States [6].

An interesting aspect studied was exposure to CDV infected dogs. Most of these studies yielded significant exposure to dogs with distemper-like illness at least 5-10 years before development of MS [12, 14]. Historically other studies have shown significant increase in MS incidence rates preceded by a CDV epidemic in locations such as Newfoundland [16], Key West [15], Sitka [17], and the Faroe Islands. One of the most interesting readings was a study conducted to determine environmental changes implemented that may have lead to a reduction in MS incidence in Key West [15]. An animal shelter on the island which was used to dump euthanized dogs was shut down. This change was said to have attributed to reduction in the MS incidence on the island.

With all of the above evidence and in context with the Hills criteria, I will conclude that there is biological plausibility of CDV playing a causal role in development of MS owing to the demyelinating nature of illness caused by this virus. In addition, this virus still causes disease in dogs despite the widespread use of vaccines [18]. This reinforces the possibility of contracting the virus via exposure to infected dogs (zoonotic disease). However, temporal association could not be established between the virus and occurrence of MS. This may be due to the fact that there are other viruses too causing demyelinating diseases such as measles and HIV, which have a well established role in human diseases. Similar symptoms of demyelinating diseases caused due to viruses other than CDV may have resulted in incorrect estimation of MS prevalence or incidence. There is some consistency among case-control studies which demonstrate exposure to CDV or dogs before the development of MS. However, owing to chronic nature of the CDV these studies do not really make a concrete argument for the role of this virus in MS causation. There does not seem to be any study conducted to examine a dose-response relation of the virus with respect to development of MS. In lieu of the above evidence, a criterion of specificity has been wasted and is best overlooked.

So one real conclusion from observations made so far is that CDV may be responsible for the causation of MS. However, it is definitely not the only factor in the causal pathway. This implies that CDV may be a necessary factor in the development of MS as could be other infectious agents (bacterial or viral). However, an individual’s environment, genetics and immune system are other sufficient factors crucial in disease causation. Also I take this opportunity to highlight the cause ‘Take good care of your pet dog and yourself’. It is the rule of nature, “What goes around, comes around”.


1. National Multiple Sclerosis Society. (n.d.). What is multiple sclerosis? Retrieved April 12, 2010.

2. The Journal of the American Medical Association. (2006). Multiple Sclerosis. Retrieved April 12, 2010.

3. National Institute of Neurological Disorders and Stroke. (2010). NINDS Multiple Sclerosis Information page. Retrieved April 12, 2010.

4. The Multiple Sclerosis Information Trust. (2008). All about multiple sclerosis. Retrieved April 12, 2010.

5. Cook, S.D. (1996). Epidemiology of multiple sclerosis: Clues to the etiology of a mysterious disease. Neuroscientist, 2, 172-80. Retrieved April 12, 2010.

6. Cook, S D, Rohowsky-Kochan, C, Bansil, S, et al. (1995). Evidence for multiple sclerosis as an infectious disease. Acta neurologica Scandinavica. Supplementum, 161, 34-42. Retrieved April 12, 2010

7. Johnson, R T. (1994). The virology of demyelinating diseases. Annals of neurology, 36 Suppl, S54-S60. Retrieved April 12, 2010.

8. Giovannoni, G, Cutter, G R, Lunemann, Jan, et al. (2006). Infectious causes of multiple sclerosis. Lancet Neurology, The, 5(10), 887-894.

9. Yoshikawa, Y, Ochikubo, F, Matsubara, Y, et al. (1989). Natural infection with canine distemper virus in a japanese monkey (macaca fuscata). Veterinary microbiology, 20(3), 193-205.

10. Hughes, R A, Russell, W C, Froude, J R, et al. (1980). Pet ownership, distemper antibodies and multiple sclerosis. Journal of the Neurological Sciences, 47(3), 429-432.

11. Rohowsky-Kochan, C, Dowling, P C, & Cook, S D. (1995). Canine distemper virus-specific antibodies in multiple sclerosis. Neurology, 45(8), 1554-1560.

12. Norman, J E, Cook, S D, & Dowling, P C. (1983). Household pets among veterans with multiple sclerosis and age-matched controls. pilot survey. Archives of neurology, 40(4), 213-214.

13. Haile, R, Smith, P, Read, D, et al. (1982). A study of measles virus and canine distemper virus antibodies, and of childhood infections in multiple sclerosis patients and controls. Journal of the Neurological Sciences, 56(1), 1-10.

14. Cook, S D, Natelson, B H, Levin, B E, et al. (1978). Further evidence of a possible association between house dogs and multiple sclerosis. Annals of neurology, 3(2), 141-143.

15. Macgregor, H S, & Latiwonk, Q I. (1992). Search for the origin of multiple sclerosis by first identifying the vector. Medical hypotheses, 37(2), 67-73.

16. Pryse-Phillips, W E. (1986). The incidence and prevalence of multiple sclerosis in newfoundland and labrador, 1960-1984. Annals of neurology, 20(3), 323-328.

17. Cook, S D, & Dowling, P C. (1982). Distemper and multiple sclerosis in sitka, alaska. Annals of neurology, 11(2), 192-194.

18. Cook, S D. (1987). Man, dogs, and hydatid disease. The Lancet, 1(8523), 21-22.