Reemerging Sudan Ebola Virus Disease in Uganda, 2011

Two large outbreaks of Ebola hemorrhagic fever occurred in Uganda in 2000 and 2007. In May 2011, we identified a single case of Sudan Ebola virus disease in Luwero District. The establishment of a permanent in-country laboratory and cooperation between international public health entities facilitated rapid outbreak response and control activities.

which was then sealed. The coffi n was released to the girl's relatives for burial, with instructions not to open the coffi n or touch the body before burial.
A blood sample collected at the hospital before the patient's death was transported to the US Centers for Disease Control/Uganda Virus Research Institute (CDC/ UVRI) laboratory in Entebbe for diagnostic testing by reverse transcription PCR (RT-PCR), antigen-detection ELISA, and IgM for fi loviruses as described (1)(2)(3)(4)(5). Evidence of infection with an Ebola virus of the genus and species Ebolavirus Sudan ebolavirus (SEBOV) was detected by RT-PCR and confi rmed by antigen-detection ELISA. Results of ELISA IgM against Ebola viruses and all tests for Marburg virus were negative. SEBOV was also isolated from blood on Vero E6 cells at the Viral Special Pathogens Branch, CDC, Atlanta, GA, USA.
Overlapping PCR fragment copies of the complete virus genome were amplifi ed, and the nucleotide sequence was obtained as described (6). Maximum-likelihood phylogenetic analysis confi rmed SEBOV and demonstrated that the isolate (Nakisamata isolate, JN638998) was closely related (99.3% identical) to the Gulu SEBOV strain obtained from northern Uganda in 2000 ( Figure 1). A postmortem diagnosis indicated Ebola hemorrhagic fever (EHF) caused by SEBOV as the cause of the patient's death.
An investigation team from the Uganda Ministry of Health, CDC Uganda, and UVRI traveled to Bombo Military Hospital and Nakisamata village, the home of the casepatient, on May 13, 2011. The village is located in Luwero District, ≈50 km north of Kampala. The investigation team established that the case-patient reported feeling ill on May 1. She had a mild headache and was given an over-thecounter analgesic. She had a fever with chills on May 4 and began vomiting on May 5. On May 6, she experienced intense fatigue and epistaxis. The patient's grandmother then took her to a local health clinic where she received adrenaline nasal packs for her epistaxis and injections of quinine and vitamin K. The patient's condition continued to worsen, and she experienced hematemesis and vaginal bleeding. She was then transported by motorcycle taxi to Bombo Military Hospital, ≈35 km north of Kampala, by her grandmother and father.
The investigation team identifi ed 25 close contacts of the patient, comprising 13 persons who had physical contact after illness onset at her home and 12 hospital staff members. Four of the hospital contacts were classifi ed as having a high risk for exposure to SEBOV because of possible exposure to the patient's body fl uids: 2 persons who performed tracheal intubation and 2 who handled the body after death. On May 15, a team from CDC Atlanta arrived in Uganda to provide assistance in laboratory diagnostics and epidemiological response. Also on this date, Médecins Sans Frontières, a nonprofi t medical humanitarian Relatives reported that the girl did not travel outside Nakisamata village in the 3 months preceding her illness and did not attend any funerals or have contact with anyone visiting from another town or village before her illness. They recalled no unusual deaths in the area in recent months. They also reported that she had not been exposed to any sick or dead animals in the village or nearby forested area.
During follow-up investigation in Nakisamata village, several species of bats (tentatively identifi ed as belonging to the genera Epomophorus, Hipposideros, Pipistrellus, and Chaerophon) were found roosting in unoccupied houses and several classrooms of the village schoolhouse where the girl attended classes, ≈400 meters from her home. Sixty-four bats were collected, and testing of these bats for Ebola virus (EBOV) is ongoing.
Samples from 4 family members, none of whom reported illness, were obtained and tested for EBOV by RT-PCR, antigen-detection ELISA, and IgM and IgG ELISAs. Test results for 3 of the family members were negative. One juvenile relative had positive IgG test results at a titer of 1,600 but was IgM negative, indicating past infection with EBOV. Since IgM antibodies can persist for as long as 2 months after infection (1,7), this person's infection appears temporally unrelated to the case-patient, who had EHF attributed to SEBOV. No clinical information was available to determine whether the relative's infection was symptomatic. Contact studies and serosurveys suggest that some EBOV infections can go unrecognized (1,(8)(9)(10)11).

Conclusions
This case represents the second documented occurrence of an identifi ed single-case EHF outbreak (12). We were unable to identify an epidemiologic link to any suspected EHF cases before the girl's illness onset, or to conclusively identify a suspected environmental source of infection in and around the village in which she lived. This suggests that her exposure was zoonotic in nature and must have occurred in the vicinity of her residence, since her relatives reported that she did not travel. The fact that an additional family member had serologic evidence of an epidemiologically unrelated EBOV infection further supports the notion that zoonotic exposures have occurred in the vicinity of the case-patient's village.
Rapid laboratory identifi cation in this outbreak supported mobilization of an investigation team 1 day after initial laboratory detection and the rapid establishment of an isolation facility at Bombo Military Hospital. In this instance, the initial high suspicion of EHF by clinical staff, the appropriate use of personal protective equipment and barrier protection by hospital staff, and the rapid laboratory confi rmation of EHF in-country likely contributed to limiting the size of this outbreak.
The timeliness of diagnostic confi rmation and outbreak response was much improved over that during previous EHF outbreaks in Uganda (timeline shown in Figure 2), during which transmission of the virus occurred for multiple months before the outbreaks were detected (13-15). This improvement was possible mainly because of collaboration by the CDC Viral Special Pathogens Branch and the Uganda Virus Research Institute to establish a permanent high-containment laboratory that is capable of performing diagnostic testing for fi loviruses and other causes of VHF in Uganda. The limited extent of this outbreak also demonstrates the powerful utility of a national VHF surveillance system, coupled with the ability to rapidly diagnose and respond to limit the spread of such high-hazard infections in the community and health care facilities. Continued efforts are needed to build and sustain VHF surveillance networks across Africa.