Category Archives: epidemiology

Dogs as Plague Sentinels and Vectors

W020140703377708061490_r75
Marmot fighting a wild dog in northern Tibet (Source: China Tibet Online/ Xinhua)

I’ve been a little obsessed with thinking about dogs and the plague lately. Dogs are often overlooked in historic plague discussions because they usually survive plague and dog-specific fleas are not associated with transmitting plague. Yet, dogs can host many of the fleas common among rodents and others that do transmit the plague including the cat flea (Ctenocephalidis felis) and the human flea (Pulex irritans) (Gage, Montenieri, & Thomas 1994). In a case controlled study of nine US cases of bubonic and septicemic plague in 2006, having dogs in the home and particularly sleeping with a dog was a significant risk factor, probably by flea transfer (Gould et al, 2008).  There is also a growing awareness that dogs can also transmit pneumonic plague directly to humans. Like other aspects of plague biology, there is a lot going on under a veneer of normalcy.

Dogs do readily contract the plague; it’s just not apparent to casual observation. In the American state of New Mexico, 62 domestic dogs were diagnosed with plague just between the years 2003 and 2011 — 97% survived (Nichols et al, 2014).  The dogs were diagnosed by an increase of Yersinia pestis F1 antibody greater than four times greater than the recovered level, by isolation of Yersinia pestis from a body fluid or by direct flourescent antibody assay of a tissue specimen. All of them had some physical sign of infection with fever and lethargy being found in 100% of cases, but buboes or lymphadenopathy (enlarged lymph nodes) were found in only 23% and these were all in the jaw and neck region. The mean time for recovery was two days, although all but one did receive at least one dose of antibiotics. Potential sources of plague exposure are from prairie dogs, ground squirrels, chipmunks, and rabbits. Only three of the dogs had any fleas at all, but as these dogs were pets, most had received anti-flea treatment.

Monitoring plague in working dogs and other carnivores is the most efficient method of doing plague surveillance in the vast semi-arid grasslands that harbor some of the most enduring plague reservoirs. Dogs are especially useful because their immunity only lasts about six months, so a detectable level (titre) of plague antibody indicates recent contact with an infected animal. Gage, Montenieri, and Thomas (1994:6) estimated that  “sampling even a few rodent consuming carnivores, such as coyotes, can be roughly equivalent to sampling hundreds of rodents for evidence of plague infection”. The earliest serologic survey that I have found was done in Navajo lands in 1966-1968. In this same survey  in 1968, “the plague organism was isolated from a pool of fleas (Pulex irritans) taken from the household dogs of a person with plague” (Archibald & Kunitz 1971). Carnivores are now routinely monitored in the US.  Surveying herding dogs in Iran was able to show that the long unmonitored plague foci is still active (Esamaeili et al., 2013). Recent Chinese F1 antibody surveys in the Gansu province are more ominous: in 2012 4.55% of dogs were positive, but it had jumped to 10% of dogs by 2014 (Ge et al, 2014). Another  2014 survey of multiple Yersinia species in dogs found 25% of dogs in Gansu province and 18% of dogs in Qinghai province to be positive for Yersinia pestis F1 antibody, while no plague-free provinces had a single dog that had a positive antibody titre (Wang et al, 2014).

Consumption is the likely primary route of infection for dogs.  The 62 dogs from New Mexico are believed to have been primarily infected by consumption of a plague infected rodent or rabbit (Nichols et al, 2014). In a 2014 case study from China, an infected marmot was taken from a dog, butchered and divided among five dogs. All five dogs developed positive antibody titers for  plague and the shepherd who took the marmot from the dog developed pneumonic plague (but not his brother who butchered the marmot). Aerosol transmission was supported by  the isolation of Y. pestis from sputum and throat samples (Ge et al, 2014). One dog not fed the marmot was negative for the F1 antigen. Three of the 151 human contacts given prophylactic antibiotics developed an antibody titre but did not manifest disease. According to Chinese policy, the five positive dogs were euthanized and the local marmots were depopulated (Ge et al., 2014).

Dogs can transmit plague to humans through fleas that feed on the dog, fleas carried by the dog from the rodent source of the infection,  through bites or scratches, or by aerosols from dogs that develop a systemic infection. While dogs are usually thought of transmitting infected fleas to people, the  number of pneumonic cases linked to dogs is increasing. The first confirmed transmission of pneumonic plague from a dog to a person occurred in China in 2009 (Wang et al, 2015). The index case in turn transmitted pneumonic plague to eleven people. Three of these twelve cases died with the other nine cases confirmed by Y. pestis F1 antibody titres. All of the Y. pestis isolates were later typed to “biovar antiqua” — a reminder that older strains are still very virulent (Wang et al, 2009). In June 2104, in Colorado, a dog transmitted pneumonic plague to three caregivers, one of whom transmitted it to another person. All of four of these cases survived and 88 additional people were given prophylactic antibiotics (Runfola et al, 2015). Three of China’s 2014 plague cases in Gansu province within the Qinghai-Tibet plague focus area  were pneumonic plague in herders.  All three arrived at the medical center too late for effective antibiotic treatment and died (Li et al, 2016). Chinese authorities believe that two of these men may have contracted plague from infected dogs and the third directly from a marmot (Lie et al, 2016).

Dog transmitted plague seems to usually result in family or small settlement size outbreaks. I do wonder about the potential role of dogs in the Bronze Age cases of plague (Rasmussen et al, 2015). Dogs contracting plague by consumption of infected rodents and passing it on to human contacts seems possible with the tools of the Bronze Age strains. It might also be worth investigating the potential role of dogs in the beginning of the Great Manchurian Plague of 1910-1911, which focused on hunters who likely used dogs extensively. Indeed hunters in this region would feed sick marmots to their dogs believing that they could not contract the disease. Outbreaks of 100% lethal plague were not unknown among hunting families in Manchuria (Summers 2012: 122-124). Such a high mortality rate would suggest pneumonic plague.

References:

Archibald, W. S., & Kunitz, S. J. (1971). Detection of plague by testing serums of dogs on the Navajo Reservation. HSMHA Health Reports.

Esamaeili, S., Azadmanesh, K., Naddaf, S. R., Rajerison, M., Carniel, E., & Mostafavi, E. (2013). Serologic Survey of Plague in Animals, Western Iran. Emerging Infectious Diseases, 19(9). http://doi.org/10.3201/eid1909.121829

Gage, K. L., Montenieri, J. A., & Thomas, R. E. (1994). The role of predators in the ecology, epidemiology, and surveillance of plague in the United States, 20.Proceedings of the 16th Vertebrate. Pest Conference (W.S. Halverson& A.C. Crabb, Eds.) Published at Univ. of Calif., Davis. 1994.

Ge P, Xi J, Ding J, Jin F, Zhang H, Guo L, Zhang J, Li J, Gan Z, Wu B, Liang J, Wang X, Wang X, Primary Case of Pneumonic Plague in Marmata himalayana natural focus area Gansu Province, China, International Journal of Infectious Diseases (2014), http://dx.doi.org/10.1016/j.ijid.2014.12.044

Gould, L. H., Pape, J., Ettestad, P., Griffith, K. S., & Mead, P. S. (2008). Dog-Associated Risk Factors for Human Plague. Zoonoses and Public Health, 55(0), 448–454. http://doi.org/10.1111/j.1863-2378.2008.01132.x

Li, Y., Li, D, Shao, H., Li, H and Han, Y. (2016) Plague in China 2014 — All sporadic case report of pneumonic plague. BMC Infectious Disease. 16: 85.

Lin, Karen. (2014-07-02) Photo: Himalaya marmot eaten by wild dogs in N. Tibet. China Tibet Online. http://www.vtibet.com/en/news_1746/focus/201407/t20140703_209395.html

Nichols, M. C., Ettestad, P. J., Vinhatton, E. S., Melman, S. D., Onischuk, L., Pierce, E. A., & Aragon, A. S. (2014). Yersinia pestis infection in dogs: 62 cases (2003-2011). Journal of the American Veterinary Medical Association, 244(10), 1176–1180. doi:10.2460/javma.244.10.1176

Rasmussen, S., Allentoft, M. E., Nielsen, K., Orlando, L., Sikora, M., Sjögren, K.-G., et al. (2015). Early Divergent Strains of Yersinia pestis in Eurasia 5,000 Years Ago. Cell, 163(3), 571–582. http://doi.org/10.1016/j.cell.2015.10.009 [Bronze Age cases]

Runfola, J. K., House, J., Miller, L., Coltron, L., Hite, D., Hawley, A., et al. (2015). Outbreak of Human Pneumonic Plague with Dog-to-Human and Possible Human-to-Human Transmission — Colorado, June–July 2014. MMWR. Morbidity and Mortality Weekly Report, 64(16), 429–434.

Salkeld, D. J., & Stapp, P. (2006). Seroprevalence Rates and Transmission of Plague (Yersinia pestis) in Mammalian Carnivores. Vector-Borne and Zoonotic Diseases, 6(3), 231–239. http://doi.org/10.1089/vbz.2006.6.231

Summers, William C. (2012) The Great Manchurian Plague of 1910-1911: The Geopolitics of an Epidemic Disease. Yale University Press.

Wang, H., Cui, Y., Wang, Z., Wang, X., Guo, Z., Yan, Y., et al. (2015). A Dog-Associated Primary Pneumonic Plague in Qinghai Province, China. Clinical Infectious Diseases, 52(2), 185–190. doi:10.1093/cid/ciq107

Wang, X., Liang, J., Xi, J., Yang, J., Wang, M., Tian, K., et al. (2014). Canis lupus familiaris involved in the transmission of pathogenic Yersinia spp. in China. Veterinary Microbiology, 172(1-2), 339–344. doi:10.1016/j.vetmic.2014.04.015

Human Parasites of the Roman Empire

Last week photos of Roman toilets were splashed across the web breaking the news that the Romans were not a healthy as most people seem to have assumed. As with many public health interventions, the real value of a sanitation system is out of view (and out of mind) to most people. Its not the toilet that keeps us healthy; its the water treatment plant. Plumbing just moves waste with its microbes and parasites from one place to another.

Paleoparasitology specialist Piers Mitchell put the Roman public health system to the test by evaluating the evidence for human parasites in archaeological remains from before, during and after the Roman Empire. Comparisons before and after the empire are more difficult in North Africa and the Middle East because these areas had long standing sophisticated civilizations before the Roman empire. There is more clarity between civilizations in Europe since Celtic and Germanic societies did not have anything like Roman infrastructure. Contrary to his expectations, there were just as many parasites and ectoparasites in the Roman era as before or after.  In some cases the empire helped spread parasites across Europe. Relative amounts of parasites across times is difficult to ascertain for a huge variety of reasons. So while the same parasites were present, the degree of infestation would have varied by place and time period, and archaeology can’t reliably predict this.

The Roman achilles’ heel was their use of human waste for fertilizer and fecal contamination of rivers.  Human waste was added to the other manure and redistributed to farm fields and the watershed. What they could not have understood is that human waste is a greater risk for the transmission of human parasites and bacterial diseases. Mitchell also suggests that Roman bath water, that was rarely changed, could have transmitted worm eggs and other parasites. Aquaducts did bring in cleaner water to some of the larger cities but the system could be contaminated and not all Roman sites had access to water from aquaducts. Walter Scheidel (2015:8) has claimed that the city of Rome itself was an example of the”urban graveyard” effect with a very unhealthy population despite having a “heavily subsidized food and water supply”. Scheidel emphasizes the impact of malaria and gastrointestinal disease. We should also keep in mind that a large proportion of gastrointestinal disease would have been bacterial or viral.

still_life_tor_marancia_vatican
Second century Roman mosaic of foodstuffs

As the mosaic to the left shows, the Romans did change agriculture throughout the empire. They spread Mediterranean preferences for cereals and more fish and other aquatic food sources. Mitchell suggests that the Roman love for fish products, especially the fermented fish sauce garum, probably help spread fish tapeworms found throughout the empire. Many parasites and bacterial spores have evolved to withstand preserving methods like smoking, pickling, and osmotic preservation (like salting or sugaring).  Whipworm was the most common parasite found, but round worms and tape worms were also common. Lancet liver flukes were widespread and indicate the (presumably accidental) consumption of ants.  Antibody based detection (ELISA) has been able to identify Entamoeba histolytica that causes the usually endemic amoebic dysentery (as opposed to the epidemic bacterial dysentery caused by Shigella species). Although not strictly speaking parasites, Mitchell notes an abundance of evidence for flies around cesspits suggesting that they contributed to the spread of diseases associated with fecal contamination. He also notes that schistosomiasis has not been identified in Roman Europe, even though it has been found in medieval European remains.

Turning to ectoparasites, Mitchell found ample evidence of head lice, body lice, public lice, human fleas and bed bugs across the Romanized world. Human fleas (pulex irritans) have been particularly well preserved in Roman, Anglo-Scandinavian and medieval York in Britain. Mitchell notes that human fleas and body lice were present in over 50 archaeological layers at York. He concludes that “the Roman habit of washing in public baths does not seem to have decreased their risk of contracting ectoparasites, compared with Viking and Medieval people who did not use public baths in the same way” (Mitchell 2016: 6). Mitchell suggests that there were enough ectoparasites to support particularly lice transmitted diseases. He notes that Plague of Justinian was transmitted by fleas but is non-committal on the likely specific vector.

In examining the impact of the Roman empire, Mitchell notes that the transition from a wide variety of zoonotic parasites to those primarily associated with human fecal contamination had already occurred before the Roman expansion out of Italy. This shift is paralleled elsewhere and is tied to shift from hunter-gathers to settled agriculture. Whipworm, roundworm and amoebic dysentery were the primary parasites of Roman Europe, while the Romans seem to have made a lesser impact on North Africa and the Middle East where endemic zones of parasites were well established.

Malaria is the one parasitic disease I would have liked to see Mitchell discuss more. Mitchell notes that malarial aDNA has been found in Egypt and anemia possibly caused by malaria in Italy. He overlooks all the malaria work by Robert Sallares including malarial aDNA from Late Roman Italy and better anemia studies correlating with malaria have been done in Italy and Britain by Rebecca Gowland’s group. Yet, malaria is such a big topic that it would be hard to cover along with all the other parasites.

References:

Mitchell, P. D. (2016). Human parasites in the Roman World: health consequences of conquering an empire. Parasitology, 1–11. http://doi.org/10.1017/S0031182015001651

Scheidel, W. (2015). Death and the City: Ancient Rome and Beyond. Available at SSRN 2609651.

See also:

Hall, A., & Kenward, H. (2015). Sewers, Cesspits, and middens: a survey of the evidence of 2000 years of waste disposal in York, UK. In P. D. Mitchell (Ed.), Sanitation, latrines and intestinal parasites in past populations (pp. 99–120).

Changing the Plague-Flea Transmission Paradigm

The old paradigm is dead! Long live the new paradigm!

X. cheopis, the rat flea
Xenopsylla cheopis, the rat flea

Rebecca Eisen, David Dennis, and Kenneth Gage just published an article gathering all the evidence that should put an end to the blocked flea model  as the only significant method of plague transmission.  They summarize the data proving that unblocked fleas can and do transmit Yersinia pestis at levels that readily cause infection in rodents and humans. They call all transmission by unblocked fleas early phase transmission (EPT), even in flea species that never block.

Important findings summarized:

  • The blocked flea model –  that only a biofilm blocked Xenopsylla cheopis which can not eat so it tries to aggressively feed and  regurgitates high numbers of Yersinia pestis into the bite site – is insufficient to account for either epizootics or large human outbreaks. Blocked fleas do transmit the plague but are simply insufficient to account for the speed and volume of epizootic and epidemic transmission.
  • Transmission can occur as quickly as the very next blood meal taken by the flea, at times within 1-2 hours. Y. pestis does not need to replicate in the flea for transmission to occur. This makes it much more likely that the flea will survive long enough to transmit the infection.
  • Early phase transmission has been experimentally observed to cause infections after exposure to a single Oropsylla montana flea. Therefore, exposure to large numbers of unblocked infected fleas is not required for transmission. Epidemiologic findings suggest that most US cases come from bites from a single or at most a few fleas, and this is consistent with findings around the world where fleas that do not block are primary vectors.
  • Many reservoirs of plague are maintained by fleas that never block. Prairie dog reservoirs in the western US and great gerbil reservoirs in central Asia are both maintained by fleas that are never blocked by a biofilm.
  • “In short, EPT was observed in all flea species evaluated at varying temperatures. Transmission occasionally occurred as early as 3 h post-infection but usually was observed over 1-4 dpi [days post infection]. Although all flea species tested were capable of EPT, efficiency in these studies varied among species, suggesting that some fleas are likely to be more important than others in the rapid spread of plague in nature, especially those that are both efficient transmitters and abundant on susceptible hosts.” (p. 3)

  • Strains of Y. pestis that can not form a biofilm transmit as effectively by EPT as biofilm competent strains. Virulence factors that are necessary for biofilm production are not necessary for EPT.
  • EPT – compared to a contaminated “dirty needle” – is a mechanical form of transmission that “requires no modification or multiplication of the pathogen in the vector for transmission to occur” (p. 4).
  • In the pre-antibiotic era, many human bubonic plague infections and all septicemic and pneumonic cases would have produced bacteremia levels sufficient to infect fleas for EPT. In the 71 fatal plague cases recorded by the CDC between 1956 and 2013, 86.8% were either primary or secondary septicemic cases.
  •  “Epidemic support in favor of interhuman flea borne transmission comes from records of limited bubonic plague outbreaks in isolated rural communities under exceptional circumstances of heavy human flea infestations, high familial attack rates, and a lack of evidence for concurrent rat-flea borne plague. … [studies in Africa, the Middle East and the Andes mentioned]… Based on epidemiologic pattern of person to person spread, especially the high attack rates among contacts of the sick, an absence of domestic rats, and an unusual abundance of P. irritans infesting villages and their homes, investigators concluded that the outbreak resulted from infective bites by P. irritans.” (p. 7-8)

  • They note that interhuman plague transmission by pulex irritans has been documented early in the 20th century and supported by laboratory experiments. As covered here a year ago, infected Pulex irritans were recovered from the homes of plague patients in Madagascar in 2013. They end with a call for more work on P. irritans to evaluate its role in modern and historical human epidemics.

It is worth noting here that throughout the article they cite many studies using many different fleas. EPT studies have also been demonstrated  for mouse fleas (Aetheca wagneri Baker)  and cat fleas (Ctenocephalides felis).  I’ve never really understood why studies of historic plagues often overlook mice as a source of fleas.

I also have to add that mechanical transmission by the flea makes a lot of evolutionary sense.  It gives evolution a place to start tinkering. ‘Good enough’ is the stuff of evolution! Optimization only occurs after a very long evolutionary process, and may never be achieved. The fact that X. choepis evolved a method (via bioflim blockage and regurgitation [LPT]) to keep transmission going longer does suggest that the rat flea has been historically important to Y. pestis evolution. Obviously mechanical transmission has also allowed Y. pestis to expand into areas and exploit new opportunities where a more complicated, required transmission system would have been an obstacle.

Experiments proving that EPT is possible have been scattered over the last 50 years! And, yet the old paradigm still reigned. Why? Obviously there has been a lack of communication within science and between science and the humanities. It would really be helpful for a historian of 20th century science to look into how this could have happened.

Reference:

Eisen, R. J., Dennis, D. T., & Gage, K. L. (2015). The Role of Early-Phase Transmission in the Spread of Yersinia pestis. Journal of Medical Entomology, tjv128–10. http://doi.org/10.1093/jme/tjv128. Advanced access, Aug. 19, 2015. (Open access)