Just how long does the Ebola virus linger in semen?

The 2013-2016 West African Ebola virus outbreak altered our perception of just what an Ebola outbreak could look like.

While none of the three primary affected countries–Liberia, Sierra Leone, and Guinea-have had a case since April 2016, the outbreak resulted in a total of over 28,000 cases of Ebola virus disease (EVD)–65 times higher than the previous largest EVD outbreak, and more than 15 times the total number of cases of all prior EVD outbreaks combined, from the virus’s discovery in 1976 to a concurrent (but unrelated) outbreak in the Democratic Republic of Congo in 2014.

In March 2016, cases were identified once again in both Liberia and Guinea, just after the outbreak had been declared over. Both countries were declared Ebola-free in June 2016; Guinea for the second time and Liberia for the fourth time. The last series of cases in these countries demonstrated just how different this epidemic was from prior ones, changing what we thought we knew about the virus:

Previous research suggested Ebola could persist in the semen for 40 to 90 days. But that window has been eclipsed in this epidemic by a considerable amount. A probable case of sexual transmission occurred approximately six months after the patient’s initial infection last year in Liberia. Another study found evidence of Ebola in the semen of 25% of surviving men tested seven to nine months after infection. And it takes only a single transmission to kick off a fresh recurrence of the disease.

A recent paper extended this window of virus persistence in the semen even longer–over 500 days. It also explains how the outbreaks began in both countries after being declared Ebola-free–so where did the virus come from?

In a convergence of old-fashioned, “shoe leather” epidemiology/tracing of cases and viral genomics, two converging lines of evidence led to the identification of the same individual: a man who had been confirmed as an EVD case in 2014, and had sexual contact with one of the new cases. Author Nick Loman discussed via email:

The epidemiologists told us independently that they had identified a survivor and we were amazed when we decoded the metadata to find that case was indeed the same person. The sequencing and epidemiology is tightly coordinated via Guinea’s Ministry of Health who ran National Coordination for the Ebola outbreak and the World Health Organisation.

It shows that the genomics and epidemiology works best when working hand-in-hand. If we’d just had the genomics or the epidemiology we’d still have an element of doubt.

The sequencing results also suggested that it was likely that the new viral outbreak was caused by this survivor, and unlikely that the outbreak was due to another “spillover” of the virus from the local animal population, according to author Andrew Rambaut:

If the virus was present in bats and jumped to humans again in 2016, it might be genetically similar to the viruses in the human outbreak but not have any of the mutations that uniquely arose in the human outbreak (it would have its own unique mutations that had arisen in the bat population since the virus that caused human epidemic).

It might be possible that the virus jumped from humans to some animal reservoir in the region and then back to humans in 2016 but because we have the virus sequence from the patients acute disease 15 months earlier we can see that it essentially exactly the same virus. So this makes it certain the virus was persisting in this individual for the period.

So the virus–persisting in the survivor’s semen for at least 531 days–sparked a new wave of cases. Ebola researcher Daniel Bausch noted elsewhere that “The virus does seem to persist longer than we’ve ever recognized before. Sexual transmission still seems to be rare, but the sample size of survivors now is so much larger than we’ve ever had before (maybe 3,000-5,000 sexually active males versus 50-100 for the largest previous outbreak) that we’re picking up rare events.”

And we’re now actively looking for those rare events, too. The Liberia Men’s Health Screening Program already reports detection of Ebola virus in the semen at 565 days following symptoms, suggesting we will need to remain vigilant about survivors in both this and any future EVD epidemics. The challenges are clear–we need to investigate EVD survivors as patients, research participants, and possible viral reservoirs–each of which comes with unique difficulties. By continuing to learn as much as we can from this outbreak, perhaps we can contain future outbreaks more quickly–and prevent others from igniting.

Ebola: Back in the DRC

August, 1976. A new infection was causing panic in Zaire. Hospitals became death zones, as both patients and medical staff succumbed to the disease. Reports of nightmarish symptoms trickled in to scientists in Europe and the US, who sent investigators to determine the cause and stem the epidemic. Concurrently, they would find out, the same thing was happening hundreds of miles to the north in Sudan. In all, 284 would be infected in that country, and another 358 in Zaire–over 600 cases (and almost 500 deaths) due to a mysterious new disease in just a few months’ time.

The new agent was Ebola, but remarkably, the outbreaks were unrelated, at least as far as any direct epidemiological links go. No one had brought the virus from Sudan to Zaire, or vice-versa. Molecular analysis showed that the viruses causing the outbreaks were two distinct subtypes, subsequently named for their countries of origin, Ebola Zaire and Ebola Sudan.

While Uganda is currently battling another outbreak of Ebola Sudan, rumors in the past week have suggested that this virus may have spread to former Zaire (now the Democratic Republic of Congo), where Ebola has reappeared 4 additional times since the first discovery there in 1976. It’s now been confirmed that Ebola is again present in the DRC, with an (unconfirmed) 6 deaths. However, it’s not related to the Uganda outbreak. Reminiscent of 1976, the strain that’s circulating currently in the DRC is the Bundibugyo subtype, which was first identified in Uganda in a 2007-8 outbreak in that country, rather than the Sudan type causing the current Ugandan epidemic. Interestingly, every previous outbreak of Ebola in the DRC has been caused by the Zaire type of Ebola, so the appearance of Bundibugyo is a first–though not altogether surprising given that the outbreak province borders Uganda.

Is this just coincidence that Ebola has twice now broken out in two different places at the same time, but with different viral subtypes? Hard to say. Though we can now say it’s fairly likely that bats are a reservoir host for Ebola and other filoviruses, we can’t say for sure that bats are the *only* reservoir. Indeed, we know that some outbreaks have occurred because the index case was in contact with an infected ape or their meat–were these animals originally infected by a bat, or by another source? How does the ecology of an area affect the chances of an outbreak occurring? Were there reasons that humans might be increasingly exposed to the virus in these different areas–Zaire and Sudan in 1976, DRC and Uganda in 2012–at the same time? Weather conditions? Trade/industry? Host migration or dispersal? We know with another bat-borne virus, Nipah, that changes in farming practices led to increased proximity of fruit bats and farmed pigs–allowing pigs to come into contact with virus-laden bat guano, become infected with Nipah, and subsequently transmit the virus to farmers. Things that may seem completely inconsequential–like the placement of fruit trees–can actually be risk factors for viral emergence. Is there a common factor here, or just bad luck? Only additional hard-won knowledge of filovirus ecology will be able to tell.

Ebola in Uganda: current and past outbreaks

Via H5N1 and other sources, there’s at least one new Ebola case in Uganda:

The rare and deadly Ebola virus has killed a 12-year-old Ugandan girl and health officials said on Saturday they expected more cases.

The girl from Luwero district, 75 km (45 miles) north of the capital Kampala, died on May 6, said Anthony Mbonye, the government’s commissioner for community health, in the first outbreak of the virus in Uganda in four years.

“Laboratory investigations have confirmed Ebola to be the primary cause of the illness and death. So there is one case reported but we expect other cases,” he said.

Though we’ve known about Ebola in Africa since 1976, Ebola wasn’t recognized in Uganda until a bit over 10 years ago. Now, this is the third outbreak in this amount of time. The first occurred in August of 2000; the first case died in Gulu on the 17th of September. Despite an investigation, doctors were unable to determine where or how she had contracted the disease. Her death was followed by the deaths of her husband, two children, and several other family members. This was reported to the Ministry of Health in October of that year, near the peak of the epidemic. An investigation and intervention to control the disease followed, and the epidemic was declared to be over in January of 2001. A total of 425 patients from 3 villages (Gulu, Masindi, and Mbarara) across Uganda were identified based on symptoms and/or laboratory data. 224 of them died, with a resulting mortality rate of 53%; an eerie echo of the 1976 Ebola outbreak in Sudan. Indeed, sequence analysis showed the infecting strain to be the Sudan subtype of Ebola; the first time this type had surfaced since the 1979 outbreak in Sudan. It is hypothesized that Sudanese rebels, who carried out regular attacks around Gulu, may have accidentally introduced the virus in some manner, though this has not been confirmed.

Ebola returned to Uganda in August of 2007, causing 149 illnesses and 37 deaths until the outbreak was declared over in February of 2008. This mortality (36%) was significantly lower than most Ebola outbreaks. Interestingly, when scientists tested this virus, it also reacted strangely with their assays. Therefore, they determined the entire molecular sequence of the virus, and found that it was a whole new strain of Ebola, which they named Ebola Bundibugyo.

I couldn’t find any other details about the current outbreak–how she was infected, if she’s actually the index case or if there were previous deaths that have not yet been confirmed. (The girl died at the hospital–previous deaths may have gone unrecognized if they had died at home). I’m sure more details will be coming in the next days and weeks. What we’re left with now is the knowledge that in 11 years’ time, Uganda is now on its third Ebola outbreak. These have occurred in 3 different areas of the country (Gulu is toward the north, Kampala region in the south near Lake Victoria, and Bundibugyo in the southwest, almost due west of Kampala) and with 2 different strains (thus far). This again feeds my morbid fascination with the virus–what does this mean about Ebola reservoirs in Uganda? Are these cases bat-acquired? Other wildlife? Spillover from other countries, as suggested with the 2000-1 outbreak? As always, Ebola outbreaks tend to raise more questions than they answer.

[UPDATE: via Crawford Killian, CDC says current outbreak is due to Sudan strain.]