Thursday, 26 June 2014

Are MERS cases in Saudi Arabia and the UAE linked to camel imports?

Special Guest writer: @influenza_bio

Looking at the history of MERS coronavirus infections to date, many puzzling questions come to mind.

Evidence of MERS infection has been detected in dromedary camels from Spain to Egypt to the Arabian Peninsula. Why have we seen human cases arise only in a handful of countries in the Arabian Peninsula?

Why have nearly all MERS cases originated in Saudi Arabia (KSA)?

As of June 22, 2014, 718 cases are thought to have been contracted in KSA. The UAE, a distant second, has had 69 cases. Jordan has had 17; Qatar has had 9 (although 1 had a travel history to KSA), Kuwait 3, Oman 2 and Yemen 1. All of these countries have a lot of camels.

  • Has KSA had more cases just because its population is larger? Have other countries had just as many cases, per capita?

Figure 1. MERS cases per capita (2012 data)
  • Is it just an accident that cases have appeared primarily in KSA, followed by the UAE?

    It could be.

  • Is it an artifact of surveillance that most cases have appeared in KSA, followed by the UAE? Are KSA and the UAE just better at detecting MERS cases?

    This would seem unlikely. It's certainly possible that all countries with MERS have missed some to a lot of cases, especially milder ones, and that some countries are better at detecting MERS cases than others. However, the fact that all cases detected outside the Arabian Peninsula also follow the same general pattern of country of origin suggests that we probably have the big picture at least close to right.

  • Have KSA and the UAE had more cases because they have more camels or more camels per person?
    No. First of all, these countries don't have the biggest domestic camel stocks in the region. Second, when you look at the number of MERS cases as a function of the size of domestic camel stocks (not including imported camels, which are typically slaughtered), there is no relationship.

    Figure 2. MERS cases as a function of the size of domestic camel stocks, not including imported camels (2012 FAO data[1])

    Alternatively, when you look at the number of MERS cases per capita as a function of the number of camels per capita (again, considering only domestic camel stocks, not imported camels), there is also no relationship. KSA and the UAE also do not have the greatest numbers of camels per person in the region.

Figure 3. MERS cases per capita (2012 data) as a function of the number of 
camels per capita (domestic camel stocks only; 2012 FAO data[1])

  • Are there more camels with MERS in KSA and, to a lesser extent, the UAE?

    We have no idea, as there has been almost no camel surveillance. But hopefully new surveillance and studies underway in KSA and the UAE will help us to understand the situation in camels in these countries better.

  • Is the large number of cases in KSA due only to nosocomial transmission (i.e., transmission in health care settings)?

    Not really, even though a very large proportion of its cases have been nosocomial. Many of the cases that KSA has had are from April and May 2014 and were in Jeddah, and the WHO has published an estimate that 75% of the cases arising in KSA in April-May 2014 may have resulted from nosocomial transmission.[2] Subsequently, the WHO wrote that, "Approximately one-third of these Jeddah cases are considered to be primary cases, although investigations are currently ongoing to determine whether these patients had contact with another confirmed case."[3] In other words, the WHO may end up revising the 75% nosocomial estimate upwards. Looking at graphs of MERS cases over time[4] suggests that cases may dip a bit in the winter and surge a bit in the spring, although MERS hasn't been around long enough in humans to know this for certain. Depending on the final WHO estimate for the rate of nosocomial transmission in April and May, anywhere from the "expected" number of zoonotic cases (i.e., the average monthly rate before April 2014) to almost double this number (if 75% of cases were nosocomial) may have appeared in April and May. However, even if it turns out that zoonotic cases have appeared in KSA at a more or less steady rate over time, this large amount of nosocomial transmission still wouldn't explain why most zoonotic cases are in KSA. Moreover, it would not seem likely that the effectiveness of infection control and prevention is vastly different across the Arabian Peninsula, and, in fact, there is evidence of nosocomial transmission in KSA, the UAE and Jordan, the countries with the most cases. However, it is puzzling that nosocomial transmission has seemed to peak in the spring of both 2013 and 2014 in many different settings; some environmental factor, analogous to winter for the flu, may be at work here.

  • Are there environmental factors in KSA more conducive to infection with MERS, especially around April and May (when we have seen a large number of infections this year, and to a far lesser extent, last year)?

    As mentioned above, there may be something special about spring. It is also possible that something about KSA and the UAE are particularly conducive to human infec
    tion with MERS.

  • Is anything else going on?

    There are probably many other factors at work that we don't know about. For one thing, we still do not know how MERS is transmitted to humans – whether it is transmitted by camel kisses, milk, urine or meat, among other possibilities, for instance. Once we know the routes of transmission and have some estimates of how much each route contributes to the numbers of infections, then we may have a better chance of teasing out even more differences across countries. Yet other factors may be at work, as well.

Is the number of MERS cases in a country related to the number of camels it imports?

At this point I would like to introduce a very speculative hypothesis: that the number of MERS cases in countries of the Arabian Peninsula is related to the number of camels that are imported into those countries. Significant oil wealth in some of these countries has fueled a greater demand for meat over time, and in particular, camel meat. Over recent years, this demand has been met by slaughtering primarily imported camels. 

  • KSA and the UAE are the dominant importers in the region.

    KSA imports the most camels in the region, followed by the UAE. No other country in the region comes close.

Figure 4. Camel imports in 2011 (FAO data[5])

  • Camel imports to KSA and the UAE have been increasing dramatically over the past few years, and the dominant sources of these imports are ports in the Horn of Africa.

    It is difficult to find hard numbers on the numbers of camels imported into KSA or the UAE broken down by country of origin. Also, the most recent FAO import data is from 2011. However, in the absence of access to recent export and import data from a variety of governments, a combination of FAO import data, FAO export data and news reports over the past few years paints a picture of vast and vastly increasing camel exports from northeastern Africa into primarily Saudi Arabia.

    First of all, it should be noted that, according to the FAO,[1] Somalia had 7,000,000 camels in 2012 and Ethiopia had 915,518, while KSA had 260,000. Sudan also had 4,571,000 camels.

    On November 25, 2013, the Financial Times reported in an article entitled, "Somali meat exports to Saudi Arabia soar,"[6] that, 
"Including goats, cattle and camels, total livestock exports from the civil war afflicted territories rose to 4.8m in what is the world’s largest on-the-hoof movement in the live animal trade. 
“There’s no single time that they ever exported such large numbers before,” says Ernest Njoroge, livestock expert at the EU's Somalia Unit. “If the ports in Berbera, Bosaso and Mogadishu become very, very efficient, then that will even increase.
“A lot of Ethiopian livestock is also coming through Berbera. It’s a very big market for us and it’s time for us to increase the capacity of Berbera port,” says Mr Yonis, who hopes planned investments will deliver a terminal dedicated to livestock for export to the Middle East."
A recent article by National Geographic[7] examines the huge and growing livestock trade from, in particular, Somaliland, and notes that, "While livestock—including cattle and camels—are exported year-round from Somaliland, the seasonal Hajj is the busiest time of year." A photo is shown of animals being herded onto a ship bound for Jeddah, KSA. Another photo shows animals in quarantine at the port of Berbera. During both quarantine and transport, animals are packed so closely together that it is hard not to imagine any diseases in the herds spreading like wildfire.
A news article[8] in March, 2011 discussed an announcement by KSA that the Kingdom planned "to increase livestock imports from the Horn of Africa two-fold by 2012," to close to 2 million heads of livestock, composed of sheep, goats, camels and cattle. Perhaps ironically, the article states that, "The decision to increase imports follows after Saudi Arabia's quarantine officials at the Port of Jeddah declared animals from the Horn were disease free, great in quality and strong demand in the local market." The article also notes that, "In October 2009, Saudi Arabia relaxed a eleven-year ban on Somaliland livestock and Somaliland animals have been steadily on high demand in the Saudi Kingdom ever since…. The main markets for livestock from Somaliland are Saudi Arabia, UAE, Oman and Yemen." FAO import data for 2000-2011, however, shows Yemen not importing any camels and Oman importing between 0 and 8,114 camels (which is very few camels).... Perhaps they don't import many camels.
Another news article,[9] in October, 2012, discusses the livestock export trade from Somaliland, and quotes a local official as saying, "Saudi Arabia sells barrels of oil to the world. We make our money selling livestock to Saudi Arabia." This article also mentions animals being shipped to Jeddah, KSA.
Camels from Somalia that are slaughtered for meat have typically been >5 years old.[11]
In addition to importing camels for meat, the UAE may also import some camels from Central Asia for milk production.[10]
Examining the relevant FAO camel import and export statistics for the Horn of Africa and the Arabian Peninsula, it becomes clear that most, and potentially nearly all, camels have been imported into the Arabian Peninsula from the Horn of Africa, meaning largely from Somalia, to an unknown extent from Ethiopia and to a lesser extent from Djibouti; that these imports have been primarily for slaughter, for meat; and that these imports have been increasing rapidly in recent years.

  • The dominant port for animal imports into Saudi Arabia is Jeddah.

    "Jeddah Islamic Port is the largest port in the world of handling of livestock."[12]

    Jeddah is also the second largest city in KSA, but is it a coincidence that the largest number of MERS cases this year have been in Jeddah, just as camel imports have been surging over the past few years?

  • Three camels imported into Egypt from Ethiopia and Sudan were found to be infected with MERS. MERS is in African camel populations.[13]

    One study examined 110 nasal swabs from apparently healthy camels, >6 years of age, in abattoirs in Egypt in June-December, 2013. Serum samples from 52 of these camels were also examined. Forty-eight of the serum samples were positive for MERS antibodies. However, when the nasal swabs were tested for the presence of the MERS virus, it was found that "The animals positive for either MERS-CoV or BCoV-like virus were all imported from Sudan or Ethiopia for slaughter." One of these virus samples was sequenced, and it was found that this virus was of a different lineage than the other MERS virus sequences known to date, from the Arabian Peninsula.

    Some amount of surveillance for human MERS cases is occurring in Sudan. The WHO has stated that, "health authorities are in active search for any suspected MERS-CoV cases by strengthening surveillance in the state general hospital, private clinics and hospitals."[14] I have seen no reports of surveillance in Ethiopia, Djibouti or Somalia; I hope that at least some exists.

  • A fairly close relative of the MERS virus also circulates in bats, and a close bat relative to MERS CoV viruses isolated from humans has been found in South Africa.[15]

    A portion (816 nt) of the genome of a virus from a bat in South Africa was sequenced, and the protein version of this viral fragment was found to differ from human MERS by only 1 amino acid (0.3% difference). The protein versions of two other, shorter fragments of the genome differed by 10.9% and 14%. Other, related viruses have been found in bats from around the world; this virus, from South Africa, was found to be closer to human MERS than any other virus known at the time this work was published. The authors of this study wrote that the "relatedness [between human MERS and the South African bat fragment] was as close as that of SARS-CoV and the most closely related bat coronavirus known." They also write that their finding "enables speculations of an African origin for bat reservoir hosts of MERS-CoV ancestors."

    Subsequently, a short fragment (182 nt) of a virus was isolated from a bat in KSA, and that fragment had 100% identity with the corresponding fragment of human MERS.[16]

    Together, these findings suggest that MERS may have originated in bats in the Arabian Peninsula or Africa. Further sampling in bats in both regions, along with sequencing greater portions of viral genomes, as they become available, would help us to understand the evolution of this virus in bats.

  • The number of MERS cases in each country may be related to the number of camels that are imported.
    The figure below shows that a potentially strong relationship may exist between the number of MERS cases in a country and the number of camels that that country imports. The more camels that are imported into a country, the more people that have been infected with the MERS virus. However, the numbers of data points are small; only KSA and the UAE stand clearly away from zero imports and zero cases. Thus, this graph is what I would call intriguing, but not completely convincing.

Figure 5. MERS cases as a function of number of camels imported (FAO 2011 data[5])

  • The number of MERS cases per capita in each country may be related to the number of camels that are imported per capita.

    If we "normalize" the data examined above to the population size of each country, a relationship between the number of camels imported and the number of MERS cases still holds.

Figure 6. MERS cases per capita (2012 data) as a function of the number of camels imported (2011 FAO data[5]) per person
  • The number of MERS cases in each country is not strongly related to the total number of camels (domestic stocks plus imported camels) in each country.
    If we compare the number of MERS cases in each country to the total number of camels that set hoof in that country in a given year (approximately), there is a possible suggestion of a relationship. Yemen is problematic, though, because it has more camels than any other country in the region but has had only one recorded human MERS case. Thus, there aren't more MERS infections in KSA and the UAE because they have the most camels in the region; they don't have the most camels. 

    Figure 7. MERS cases as a function of the total number of camels (domestic stocks [2012 FAO data] plus imported camels [2011 FAO data[5]])
  • The number of MERS cases per capita in each country is not related to the total number of camels (domestic stocks plus imported camels) per capita in each country.

    The UAE does have the most camels per person in the region, but KSA is very far from having the most camels per person. Thus, there is no apparent relationship between the total number of camels setting hoof in a country per capita and the number of MERS cases per capita in that country.

Figure 8. MERS cases per capita (2012 data) as a function of the total number of 
camels (domestic stocks [2012 FAO data[1]] plus imported camels [2011 
FAO data[5]]) per capita

Taken together, this data supports the hypothesis that the number of human MERS cases in a country may be driven by the number of camels imported into that country, and perhaps that what matters is how many camels are imported from the Horn of Africa. Correlation is not causality, of course; this is only a hypothesis, not a fact.

In May, KSA stopped importing camels from Somalia, Ethiopia and Sudan.[17] So, I'm clearly not the first person to have had this idea. It's not clear whether camels currently being imported into KSA are being tested for MERS, though; one recent Arab News article[18] discusses how "all livestock" in KSA are now going to be tested for MERS, but it was not entirely clear to me whether imported camels would be tested. Subsequent echoes[19] of this news article, however, stated that, "Imported camels would also be tested for MERS and quarantined, Arab News reported him [the Agriculture Minister] as saying;" in reality, the Arab News article made no such direct statement. It would be nice to get total clarity on this issue.

The UAE, on the other hand, has not suspended any camel imports.[20] However, the UAE is now testing camels being imported from Saudi Arabia, Qatar, Oman, Kuwait and Bahrain.[21]

It should also be noted that MERS appears to circulate in domestic camel stocks in at least some countries in the Arabian Peninsula at at least some times; I am not by any means suggesting that MERS is likely to be only in imported camels.

For further information on animal husbandry practices for camels in northeastern Africa, see here.[22] For further information on the Livestock Trade in the Djibouti, Somali and Ethiopian Borderlands, see here.[23]

How can we know whether camel imports are actually a driving force behind human MERS infections?

This part is a little bit complicated. As a first step, it would be important to determine whether MERS is circulating in areas exporting camels to the Arabian Peninsula, and if so, whether the MERS virus in these regions is genetically the "same" as (i.e., extremely similar to) MERS in the Arabian Peninsula.

Ideally, we would need to try to find camels with active MERS infections on the verge of being exported to the UAE, and we would need to sequence their MERS genomes. Such sampling would probably best be done in the ports of Mogadishu, Bosaso and Berbera in February, March and April. It would also be great to see sampling done in the UAE in, say, February through May. If the sequences from the camels on the verge of export were to match the sequences seen in camels or people in the UAE, then this would suggest that MERS is circulating between the Horn of Africa and the Arabian Peninsula. It might be possible from such sequence data to determine which direction the virus is traveling, i.e., whether the virus is traveling from Africa to the Arabian Peninsula or from the Arabian Peninsula to Africa. The fact that the animals are traveling primarily, if not exclusively, in one direction, though, would suggest that the virus would be moving primarily from Africa to the Arabian Peninsula.

On the other hand, if MERS sequences from export ports in the Horn of Africa would turn out to be of lineages different from those of MERS sequences seen in the UAE at the same time, then we could safely conclude that MERS is not being imported from Africa.

Is there any other way that imported camels could be driving MERS infections in the Arabian Peninsula besides by bringing more MERS virus with them? Possibly. For instance, it's possible that imported camels provide a large, susceptible pool for new MERS virus infections in their destination country. Travel may make such camels more susceptible than usual, as well. Depending on how long these camels are alive in their destination countries and on how MERS is transmitted to humans, a role for new infections of imported camels in their destination countries could turn out to be more or less relevant. For instance, if it is found that most human MERS infections develop after contact with camel meat, then new MERS infections in imported camels would be important. If no such link of this sort could be found to imported camels, then this apparent correlation would be likely to be spurious, i.e., correlated with MERS infections but for no real, underlying reason.

Suppose that camel imports do drive human MERS infections. What could this tell us? What questions would this raise?

Suppose, for the moment, that some causal relationship does exist between the number of MERS cases in a country and the number of camels imported into that country. This would raise many questions, including the following:

If infected camels are being imported…

  • Are there human MERS cases in the Horn of Africa?
    • How big of a problem are human MERS cases now?
    • How many years back can we trace human infections?
  • Are imported camels infecting people in the Arabian Peninsula?
    • How?
    • Do people becoming infected through eating camel meat?
      • Is meat a more important source of human infections than other forms of camel contact (direct, milk, urine, …)?
    • How can we prevent imported camels from infecting people?
  • Are imported camels infecting camels in the Arabian Peninsula, who then infect humans in the Arabian Peninsula?
    • How are imported camels infecting domestic camel stocks?
    • How are domestic camels infecting people?
    • How can we prevent imported camels from infecting domestic camels?
  • How can we prevent infected camels from being imported?
    • Does the import process increase the incidence of MERS among imported camels?
    • Does contact with other camels at the sites of export and/or import increase the incidence?
    • Does quarantine at the sites of export and/or import increase the incidence?
    • Does shipping increase the incidence?
  • What are the dynamics of MERS transmission in camels in the Horn of Africa and the Arabian Peninsula?
  • Have we been looking for MERS in the light under the lamppost?

If infected camels are not being imported…

  • Are imported camels becoming infected in their destination countries?
    • Are infected imported camels infecting people?
    • Are infected imported camels infecting domestic camels?

And finally, what will happen over the coming year, if KSA continues to ban camel imports from Somalia, Ethiopia and Sudan, while the UAE does not? Will KSA find other sources of camels? New human cases continue to arise in KSA; MERS is clearly still in KSA. But without camels imported from the Horn of Africa, will human cases die down, and will we see no surge next spring? Time will tell.

In the meantime…

We don't know whether imported camels affect MERS transmission in the Arabian Peninsula. We don't know whether any camels exported to the Arabian Peninsula are infected with MERS. We don't know for absolute fact whether camels give MERS to people.

In short, many more studies are needed to discover where people and camels are becoming infected with MERS, to learn how MERS is transmitted to people and to understand local and global MERS transmission dynamics in camels.

It would be good to do some surveillance in Somali ports, though.



NOTE: I (Ian M. Mackay, VDU Editor) did not have a hand in writing this post and thus take no credit for it. This was entirely the work of the Guest Writer.