RESULTS
Clinical and Epidemiologic Analysis
From July 26 to October 7, 2014, we identified 69 patients with EVD that was suspected (3 patients), probable (28 patients), or confirmed (38 patients). Of these patients, 33 were male and 36 were female; 80% of the patients were adults between the ages of 21 and 60 years (Figure 2AFIGURE 2
Suspected, Probable, and Confirmed Cases of Ebola Virus Disease (EVD) in the Democratic Republic of Congo.
). Among the patients with suspected disease, a higher proportion of children than adults were found to have non-EVD illnesses. Of the 69 patients, 21 male patients and 28 female patients died, including 3 children under the age of 5 years, resulting in a case fatality rate of 74% among probable and confirmed cases, if all cases and deaths were reported. For 28 paired contacts of patients with EVD, the median (serial) interval between dates of the reported onset of symptoms was 16 days (range, 3 to 27), and the mean (±SD) was 16.1±4.4 days, similar to that in West Africa.2 For 32 of the 49 patients who died, the median time from reported symptom onset to death was 11 days (range, 1 to 30), and the mean was 11.3±6.8; the interval was not measured for the other patients who died. All 8 health workers who were affected — 4 with probable EVD and 4 with confirmed EVD — died. As compared with patients with suspected EVD who were found to have non-EVD illnesses, patients with probable or confirmed EVD were more likely to have fever, headache, diarrhea, vomiting or nausea, fatigue, anorexia, muscle pain, difficulty swallowing, conjunctivitis, and blood in stools or vomit (Table 1TABLE 1
Cases of Probable or Confirmed Ebola Virus Disease, According to Reported Signs and Symptoms, in the Democratic Republic of Congo.
).
This outbreak has been driven by human-to-human transmission, but all cases and contacts appear to have originated from a single index case — in other words, there was no evidence of more than one transfer of infection from an animal reservoir. So far, all cases have been confined to Équateur province, including Boende town (approximately 45,000 inhabitants, with four Health Areas affected), Boende Moke, Bokongo, Lokolia, Lokula, Mondombe, and Watsi Kengo (ranging from 4000 to 12,000 inhabitants), which are situated on or close to road RP314 (Figure 1). The majority of suspected, probable, or confirmed cases of EVD have been reported from Lokolia (39 cases) or Watsi Kengo (18 cases), with only 4 cases reported from Boende town and another 8 from other towns in the district.
The maximum number of cases was reported in the weeks beginning August 17 and 24, according to the date of the onset of symptoms (Figure 2B). It is possible, by plotting the daily case incidence, to discern several generations of cases, as indicated by six peaks in incidence, including the initial index case (Figure 2C). These covered approximately five serial intervals (average, 16.1 days), generating a total of 69 cases in 70 days (Figure 2D).
The rise in case incidence during August was apparently driven by multiple infections acquired from the index case. Of the 29 patients in whom EVD was diagnosed during first 24 days of the outbreak, 21 were reported to be direct contacts of the index case (i.e., they had physical contact or contact with bodily fluids). If all these secondary cases did indeed acquire infection from a single source, this represents a basic case reproduction number2 (R 0) of 21 for this outbreak.
The number of secondary cases arising from each primary case during this outbreak was highly variable. Among other patients with EVD who infected named contacts, 1 patient generated 3 secondary cases, 2 patients generated a further 2 cases each, 30 patients generated a single extra case, and 11 patients generated no further cases. Counting all secondary cases arising among named contacts, including the index case, the average case reproduction number (R) for the whole outbreak was 1.29 (95% confidence interval [CI], −4.71 to 7.29). However, after the exclusion of the 21 cases generated by the index case, the average case reproduction number during the outbreak was 0.84 (95% CI, −0.38 to 2.06), which is below the threshold value for persistent transmission (R>1). This explains the observed decline in the EVD case incidence after mid-August. The last reported patient with EVD became ill on October 4, and no further cases were reported as of October 7.
A total of 1121 contacts of the patients were registered for follow-up. By October 7, a total of 830 had been followed for at least 21 days, which is considered to be the maximum incubation period. The index patient and her contacts had no history of travel to the EVD-affected countries in West Africa (Guinea, Liberia, Nigeria, Senegal, and Sierra Leone) and no history of contact with residents of the affected areas.
Identification and Characterization of the Virus
To identify the causative agent of the Boende EVD outbreak, investigators at Institut National de Recherche Biomédicale in Kinshasa and at Centre International de Recherches Médicales de Franceville in Gabon analyzed eight samples from patients who were suspected of having EDV. On the basis of both a conventional filoviridae-specific RT-PCR assay targeting a conserved region in the L gene and a real-time RT-PCR assay targeting the nucleoprotein gene in EBOV, six of the eight samples tested positive for EBOV. Sequencing of the 346-bp fragment amplified from the L gene in four of the samples revealed the viral sequences. These four sequences were identical and closely related to the EBOV variant that caused an EVD outbreak around Kikwit, Zaire (now part of the DRC), in 1995, indicating that the current outbreak in the DRC was unrelated to the ongoing outbreak in West Africa (Figure 3FIGURE 3
Phylogenetic Tree of Ebola Virus (EBOV) Strains Isolated in Equatorial Africa and West Africa.
). We obtained a coding-complete sequence of the virus from the sample having the highest RNA load and obtained a sequence of 18,953 nucleotides in length, which was called EBOV/H.sap/COD/14/Boe-Lok according to the recently established filovirus variant nomenclature6 (GenBank accession number, KM519951; see also GenBank numbers KM517570.1 and KM517571.1, which were obtained from initial RT-PCR analysis). The sequence showed 99.2% identity (0.8% difference, 145 mutations) with the most closely related variant that was isolated during the 1995 Kikwit outbreak (EBOV/H.sap/COD/95/Kikwit-13709). Among the 145 mutations, 22 were nonsynonymous, but none were expected to induce any important change in viral protein sequences. In particular, we observed 5 nonsynonymous mutations in the NP gene, 1 in the VP35 gene, 1 in the VP40 gene, 8 in the GP gene, none in the VP30 gene, 2 in the VP24 gene, and 5 in the L gene.
We then performed high-throughput sequencing on two additional positive samples after random amplification of extracted total RNA from serum samples. Given the moderate viral load in the serum, we could not obtain the complete sequences but only 2270 bp from one sample and 12,501 bp from the other sample. Analysis of these partial sequences showed an identity of 100% with the complete sequence obtained previously, further suggesting that the current outbreak in the DRC is due to a single introduction of this novel EBOV variant into the human population.
In contrast to the similarity between the Boende variant and other equatorial African variants (especially EBOV/H.sap/COD/95/Kikwit-13709), the gene sequence of the Lokolia isolate showed only 96.8% identity (3.2% difference) with the West African variants, with 601 mutations as compared with a Guinean variant (KJ660347) and 602 mutations as compared with a Sierra Leonean variant (KM233116).
RESULTS
Clinical and Epidemiologic Analysis
From July 26 to October 7, 2014, we identified 69 patients with EVD that was suspected (3 patients), probable (28 patients), or confirmed (38 patients). Of these patients, 33 were male and 36 were female; 80% of the patients were adults between the ages of 21 and 60 years (Figure 2AFIGURE 2
Suspected, Probable, and Confirmed Cases of Ebola Virus Disease (EVD) in the Democratic Republic of Congo.
). Among the patients with suspected disease, a higher proportion of children than adults were found to have non-EVD illnesses. Of the 69 patients, 21 male patients and 28 female patients died, including 3 children under the age of 5 years, resulting in a case fatality rate of 74% among probable and confirmed cases, if all cases and deaths were reported. For 28 paired contacts of patients with EVD, the median (serial) interval between dates of the reported onset of symptoms was 16 days (range, 3 to 27), and the mean (±SD) was 16.1±4.4 days, similar to that in West Africa.2 For 32 of the 49 patients who died, the median time from reported symptom onset to death was 11 days (range, 1 to 30), and the mean was 11.3±6.8; the interval was not measured for the other patients who died. All 8 health workers who were affected — 4 with probable EVD and 4 with confirmed EVD — died. As compared with patients with suspected EVD who were found to have non-EVD illnesses, patients with probable or confirmed EVD were more likely to have fever, headache, diarrhea, vomiting or nausea, fatigue, anorexia, muscle pain, difficulty swallowing, conjunctivitis, and blood in stools or vomit (Table 1TABLE 1
Cases of Probable or Confirmed Ebola Virus Disease, According to Reported Signs and Symptoms, in the Democratic Republic of Congo.
).
This outbreak has been driven by human-to-human transmission, but all cases and contacts appear to have originated from a single index case — in other words, there was no evidence of more than one transfer of infection from an animal reservoir. So far, all cases have been confined to Équateur province, including Boende town (approximately 45,000 inhabitants, with four Health Areas affected), Boende Moke, Bokongo, Lokolia, Lokula, Mondombe, and Watsi Kengo (ranging from 4000 to 12,000 inhabitants), which are situated on or close to road RP314 (Figure 1). The majority of suspected, probable, or confirmed cases of EVD have been reported from Lokolia (39 cases) or Watsi Kengo (18 cases), with only 4 cases reported from Boende town and another 8 from other towns in the district.
The maximum number of cases was reported in the weeks beginning August 17 and 24, according to the date of the onset of symptoms (Figure 2B). It is possible, by plotting the daily case incidence, to discern several generations of cases, as indicated by six peaks in incidence, including the initial index case (Figure 2C). These covered approximately five serial intervals (average, 16.1 days), generating a total of 69 cases in 70 days (Figure 2D).
The rise in case incidence during August was apparently driven by multiple infections acquired from the index case. Of the 29 patients in whom EVD was diagnosed during first 24 days of the outbreak, 21 were reported to be direct contacts of the index case (i.e., they had physical contact or contact with bodily fluids). If all these secondary cases did indeed acquire infection from a single source, this represents a basic case reproduction number2 (R 0) of 21 for this outbreak.
The number of secondary cases arising from each primary case during this outbreak was highly variable. Among other patients with EVD who infected named contacts, 1 patient generated 3 secondary cases, 2 patients generated a further 2 cases each, 30 patients generated a single extra case, and 11 patients generated no further cases. Counting all secondary cases arising among named contacts, including the index case, the average case reproduction number (R) for the whole outbreak was 1.29 (95% confidence interval [CI], −4.71 to 7.29). However, after the exclusion of the 21 cases generated by the index case, the average case reproduction number during the outbreak was 0.84 (95% CI, −0.38 to 2.06), which is below the threshold value for persistent transmission (R>1). This explains the observed decline in the EVD case incidence after mid-August. The last reported patient with EVD became ill on October 4, and no further cases were reported as of October 7.
A total of 1121 contacts of the patients were registered for follow-up. By October 7, a total of 830 had been followed for at least 21 days, which is considered to be the maximum incubation period. The index patient and her contacts had no history of travel to the EVD-affected countries in West Africa (Guinea, Liberia, Nigeria, Senegal, and Sierra Leone) and no history of contact with residents of the affected areas.
Identification and Characterization of the Virus
To identify the causative agent of the Boende EVD outbreak, investigators at Institut National de Recherche Biomédicale in Kinshasa and at Centre International de Recherches Médicales de Franceville in Gabon analyzed eight samples from patients who were suspected of having EDV. On the basis of both a conventional filoviridae-specific RT-PCR assay targeting a conserved region in the L gene and a real-time RT-PCR assay targeting the nucleoprotein gene in EBOV, six of the eight samples tested positive for EBOV. Sequencing of the 346-bp fragment amplified from the L gene in four of the samples revealed the viral sequences. These four sequences were identical and closely related to the EBOV variant that caused an EVD outbreak around Kikwit, Zaire (now part of the DRC), in 1995, indicating that the current outbreak in the DRC was unrelated to the ongoing outbreak in West Africa (Figure 3FIGURE 3
Phylogenetic Tree of Ebola Virus (EBOV) Strains Isolated in Equatorial Africa and West Africa.
). We obtained a coding-complete sequence of the virus from the sample having the highest RNA load and obtained a sequence of 18,953 nucleotides in length, which was called EBOV/H.sap/COD/14/Boe-Lok according to the recently established filovirus variant nomenclature6 (GenBank accession number, KM519951; see also GenBank numbers KM517570.1 and KM517571.1, which were obtained from initial RT-PCR analysis). The sequence showed 99.2% identity (0.8% difference, 145 mutations) with the most closely related variant that was isolated during the 1995 Kikwit outbreak (EBOV/H.sap/COD/95/Kikwit-13709). Among the 145 mutations, 22 were nonsynonymous, but none were expected to induce any important change in viral protein sequences. In particular, we observed 5 nonsynonymous mutations in the NP gene, 1 in the VP35 gene, 1 in the VP40 gene, 8 in the GP gene, none in the VP30 gene, 2 in the VP24 gene, and 5 in the L gene.
We then performed high-throughput sequencing on two additional positive samples after random amplification of extracted total RNA from serum samples. Given the moderate viral load in the serum, we could not obtain the complete sequences but only 2270 bp from one sample and 12,501 bp from the other sample. Analysis of these partial sequences showed an identity of 100% with the complete sequence obtained previously, further suggesting that the current outbreak in the DRC is due to a single introduction of this novel EBOV variant into the human population.
In contrast to the similarity between the Boende variant and other equatorial African variants (especially EBOV/H.sap/COD/95/Kikwit-13709), the gene sequence of the Lokolia isolate showed only 96.8% identity (3.2% difference) with the West African variants, with 601 mutations as compared with a Guinean variant (KJ660347) and 602 mutations as compared with a Sierra Leonean variant (KM233116).
การแปล กรุณารอสักครู่..
