S115

Epidemiology of Ebola (Subtype Reston) Virus in the Philippines, 1996

M. E. Miranda, T. G. Ksiazek, T. J. Retuya, Ali S. Khan,
Anthony Sanchez, Charles F. Fulhorst,* Pierre E. Rollin,
A. B. Calaor, D. L. Manalo, M. C. Roces, M. M. Dayrit,
and C. J. Peters

Research Institute for Tropical Medicine and Field Epidemiology

Training Program, Department of Health, Manila, Philippines; Special
Pathogens Branch, Division of Viral and Rickettsial Diseases, Centers
for Disease Control and Prevention, Atlanta, Georgia

Ebola (EBO) (subtype Reston [EBO-R]) virus was originally

discovered in the United States in 1989 in association with
an outbreak of viral hemorrhagic fever among cynomolgus
monkeys (Macaca fascicularis) exported from the Philippines
[1]. Subsequently, 4 other episodes of EBO-R virus infection
among monkeys imported from the Philippines have occurred
in the United States and Italy between 1990 and 1992 [2]. In
each instance, the hallmark of infection was rapid dissemination throughout the facility of a highly fatal infection in monkeys. In contrast, limited studies of humans occupationally
exposed to infected monkeys or their tissues suggested that
infection in humans is infrequent and asymptomatic or, at worst
in these limited infections, associated with only a mild disease.
There have been 4 documented asymptomatic human infections
in the United States [3, 4] and 3 antibody-positive individuals
identified in the Philippines [5]. Following the original episodes
in 1989 1990, the United States modified the procedures used
in the transportation and quarantine of nonhuman primates to
include mandatory disease control requirements [6]. Similarly,
in 1994, the Republic of the Philippines banned the export of
wild-caught monkeys and instituted other procedures (such as
a 45-day quarantine prior to export) to obviate the potential
for export of EBO-R infected monkeys [7].
In March 1996, a monkey imported from the Philippines
died while in quarantine at a commercial facility in Texas; the

Reprints or correspondence: Dr. Thomas Ksiazek, Special Pathogens Branch,

DVRD/NCID, G-14, Centers for Disease Control and Prevention, Atlanta, GA
30333.
* Present affiliation: Department of Pathology, University of Texas Medical
Branch, Galveston, Texas.
The Journal of Infectious Diseases 1999;179(Suppl 1):S1159
q 1999 by the Infectious Diseases Society of America. All rights reserved.
00221899/99/79S10020$02.00

/ 9d49$$se37

12-27-98 05:33:14

jinfa

animal was tested and found positive for EBO antigens [8].
After another animal was confirmed to be EBO virus positive
by antigen virus isolation, and reverse transcription polymerase chain reaction of EBO-R RNA, 50 animals, which constituted a quarantine cohort, of the 100 imported from the Philippines were destroyed [8]. The 8 employees in contact with the
infected monkeys and the other 50 animals from the shipment
were monitored but did not show any signs of illness or evidence of infection [8]. Sequence analysis of the glycoprotein
gene of virus from the first EBO-R infected monkey revealed
a 98.9% nucleotide homology with the original 1989 EBO-R
virus [9].
Following confirmation of the presence of EBO-R virus in
these monkeys, the Philippine Department of Health placed a
moratorium on the export and movement of monkeys within the
Philippines, pending an investigation to explain the origin of
infected animals, despite regulations that stipulate that all exported monkeys are captive-bred and must be quarantined for 45
days prior to export. Herein, we report the results of that investigation and a parallel study of the risk and health significance of
EBO-R infection in occupationally exposed humans.
Methods
Study sites. The Philippines is one of the worlds major sources
of purpose-bred cynomolgus monkeys for biomedical research.
The five licensed breeding and export facilities provide an estimated 5000 captive-bred monkeys every year to the United States,
Europe, and Japan. A small number of animals are also sold to
biomedical research facilities located in the Philippines. All five of
the breeding and export facilities are located in the main Philippine
island group of Luzon (figure 1). Each of these facilities, depending
on their size, houses between 300 and 7500 cynomolgus monkeys
at any given time (table 1). Three of the facilities are actively
replenishing or expanding their breeding stocks with animals col-

UC: J Infect

Downloaded from http://jid.oxfordjournals.org/ by guest on April 3, 2014

Ebola (subtype Reston [EBO-R]) virus infection was detected in macaques imported into the
United States from the Philippines in March 1996. Studies were initiated in the Philippines to
identify the source of the virus among monkey-breeding and export facilities, to establish surveillance
and testing, and to assess the risk and significance of EBO-R infections in humans who work in
these facilities. Over a 5-month period, acutely infected animals were found at only one facility, as
determined using Ebola antigen detection. Three of 1732 monkeys and 1 of 246 animal handlers
tested had detectable antibodies; all were from the same facility, which was the source of infected
monkeys imported to the United States. Virus transmission, which was facilitated by poor infectioncontrol practices, continued for several months in one facility and was stopped only when the facility
was depopulated. None of the 246 employees of the facilities or 4 contacts of previously antibodypositive individuals reported an Ebola-like illness. This investigation suggests that human EBO-R
infection is rare.

S116

Miranda et al.

JID 1999;179 (Suppl 1)

lected within licensed wildlife trapping areas designated by the

Philippine Department of Environment and Natural Resources on
the island of Mindanao; one facility utilizes only captive-bred
animals for breeding stock, and the other has suspended export
operations but still has limited breeding stock. Holding facilities

Table 1. Description of the five facilities of the Primate Exporters

and Breeders Association of the Philippines, May 1996 (see figure 1
for map of locations).

Facility
A
B
C
D
E

Location
Calamba, Laguna
Paranaque, Metro Manila, and Lian
Puerto Galera, Mindoro
Tanay, Rizal
Tanay, Rizal

Total

/ 9d49$$se37

No. of
workers

Animal
population

21
21
10
19
160
231

1332
652
1330
1579
7500
12,393

12-27-98 05:33:14

jinfa

near wildlife trapping areas on the southern island of Mindanao

serve as prequarantine stations where wild-caught monkeys selected as breeders are held prior to transfer to the breeding facilities
(figure 1).
Surveillance for EBO-R disease in monkeys. In May 1996,
surveillance in monkeys was established at each of the five facilities to determine if any mortality was associated with EBO-R
infection. The veterinary staff of the facilities was responsible for
the collection of liver specimens from all dead monkeys (including
sacrificed moribund and sick ones), regardless of the perceived
cause of illness. To limit dissemination, all monkeys in the same
housing unit as EBO antigenpositive animals were euthanatized.
Liver and blood samples were obtained from the sacrificed monkeys in Facility A. In addition, all blood samples that were obtained
from animals during the cross-sectional antibody survey (see below) in Facility A were also tested for EBO antigen. These samples
were stored frozen until tested for EBO antigen, using the antigencapture ELISA.
Survey for EBO antibodies in monkeys in primate breeding and
export facilities. Blood samples were obtained from a systematic

UC: J Infect

Downloaded from http://jid.oxfordjournals.org/ by guest on April 3, 2014

Figure 1. Map showing location of primate breeding and export facilities (A) and holding facilities near trapping areas (B) in southern island
of Mindanao, Philippines, where wild-caught monkeys selected as breeders are held prior to transfer to breeding facilities.

JID 1999;179 (Suppl 1)

EBO-R in Philippine Primate Facilities

/ 9d49$$se37

12-27-98 05:33:14

jinfa

Table 2. Morbidity and mortality caused by EBO-R virus infection

among nonhuman primates in export facilities in the Philippines,
1996.

Facility
Facility A
Deaths
Sacrificed (sick or moribund)
Sacrificed (healthy)
Live (healthy) (cross-sectional)
All other facilities (no. dead)
B
C
D
E
Others (Mindanao and local laboratory)
Subtotal
Total

No.
tested

%
mortality*

185
168
178
287

13.9

15
3
32
161
22
233
1051

2.3
0.2
2.0
2.1

No.
antigen
positive
(%)

85
28
16
2

(46)
(17)
(9)
(0.7)

0
0
0
0
0
0
131 (12.5)

* Based on initial populations given in table 1.

Total population unknown.

antigen except that an adjusted optical density of 0.2 at 410 nm

was required for positive status for the dilution. Titers of 1:400
were considered antibody positive.

Results
Surveillance for monkey disease and mortality caused by
EBO-R virus. Mortality was highest (13.9% based on the
initial population) in Facility A; the average was 1.9% among
all other breeding facilities (table 2). From May to September
1996, samples were obtained from 1051 animals and tested by
antigen detection. Of these animals, 586 (56%) were dead or
were sacrificed because they were sick or moribund. Of the
total samples tested for EBO-R antigen, 818 were from Facility
A animals, including the 287 samples from healthy animals
for the cross-sectional antibody survey. Among the 353 dead
and moribund monkeys collected at this facility, 113 (32%)
were antigen positive, demonstrating that EBO-R virus infection was a significant cause of death. In summary, 131 (16%)
of the 818 animals collected from Facility A had detectable
EBO-R antigen in their blood or liver (or both). The majority
of these antigen-positive monkeys were housed in gang cages.
EBO virus infection among monkeys in this facility was widespread. Over a period of 5 months, 14 of 23 cage units had
antigen-positive animals. All 233 samples from dead monkeys
from the other facilities were negative for EBO-R antigen.
Antibody survey among monkeys. Three of the 301 animals
sampled at Facility A were IgG antibody positive, which is
indicative of previous EBO-R infection. All 3 animals were
resampled a week later and were again found antibody positive.
These animals were resident in the hospital. Two were wild-

UC: J Infect

Downloaded from http://jid.oxfordjournals.org/ by guest on April 3, 2014

sample of animals to determine by antibody testing if past infection

had occurred at each facility. The sampling plan was designed to
detect a minimum antibody prevalence of 5% among defined caging groups and life stages (breeders and captive-bred growers).
Once the total number of animals to be sampled within the defined
groups at a facility was determined, the average number of animals
within a cage or housing unit within this caging group was calculated, and the number of housing units needed to meet this requisite
total sample size was then taken from evenly spaced cages or
housing units. Basic demographic details (age, sex, wild-caught
or captive origin) were collected on all animals in the sample. A
20% over-sampling was done, and all animals awaiting export and
all newly arrived wild-caught animals were sampled at all facilities.
In addition, all animals hospitalized in Facility A, the source of
the infected monkeys for the Texas outbreak, were also sampled.
Pregnant monkeys near term and those 1 year old were not
sampled because of concerns for the trauma and stress of capture
and handling.
Although an embargo on new captures in Mindanao had been
initiated, collection of samples from all newly captured animals in
transit at holding facilities near trapping areas in Mindanao was also
initiated in an attempt to determine the source of the EBO-R.
Risk assessment on occupationally exposed humans. All employees of breeding and export facilities were interviewed. Data
on possible risk factors for infection (i.e., bites or scratches from
monkeys, handling of blood and other tissues) and history of any
EBO-like illness (fever associated with hemorrhage, myalgias)
were gathered. Blood samples were collected and tested for EBOR IgG antibody by ELISA. Family contacts of previously identified
antibody-positive persons (Miranda ME, unpublished data) were
also interviewed, and blood samples were obtained for antibody
testing.
Antigen and antibody assay. EBO antigen detection tests were
done according to a previously published method with only slight
modification [10, 11]. Blood and 10% tissue homogenates were
prepared in a biosafety level 4 laboratory, gamma-irradiated with
a 60Co source, and then tested at dilutions of 1:4, 1:16, 1:64, and
1:256 in both specific and nonspecific (parent myeloma ascites)
coated wells. The antibody used to detect EBO virus antigen captured by monoclonal antibodies had been replaced by serum from
rabbits hyperimmunized with the three known EBO virus subtypes
(Zaire, Sudan, Reston) when the material was made.
Rabbit antibodies that bound to EBO antigens were detected
by goat anti-rabbithorseradish peroxidaseconjugated antibodies
followed by H2O2-ABTS (Kirkegaard & Perry, Gaithersburg, MD)
substrate. Adjusted optical density values (readings at 410 nm
of wells coated with EBO-specific monoclonal antibody less the
corresponding well coated with parent myeloma ascites) 0.1
were assigned positive status. Specimens with EBO antigen titers
1:16 were considered positive.
Antibody was detected by an ELISA test, using an antigen made
from basic buffer-detergentextracted Vero E6 cells infected with
the prototype isolate of EBO-R from the 1989 outbreak in Reston,
Virginia [11, 12]. Sera were tested at dilutions of 1:100, 1:400,
1:1600, and 1:6400 against both the EBO antigen and a negative
antigen made from mock-infected cells prepared in a similar manner. Horseradish peroxidaseconjugated anti-human IgG (Fc specific) was used for humans and primates as a detector antibody.
Antibody titers were calculated in a manner similar to that for

S117

S118

Miranda et al.

Table 3. Results of cross-sectional EBO-R IgG antibody survey in

the nonhuman primate export facilities in the Philippines, May 1996.

Facility
A
B
C
D
E
Holding facilities, Mindanao
Total

Total no.
tested

No. antibody
positive (%)

301
198
288
317
584
44
1732

3 (1)
0
0
0
0
0
3 (0.2)

Discussion
This investigation documents active serial transmission of
EBO-R within a primate export facility in the Philippines. This
facility was the source of the 1996 importation of EBO-R
infected monkeys into Texas, of infected monkeys from 3 of
the 4 EBO-R outbreaks in the United States in 1989 1990,
and of infected monkeys in Siena, Italy, in 1992 [2, 13]. The
only other known source of EBO-R infected monkeys from
the Philippines involved an illegal animal exporter in 1989 [6].
Results of antibody and antigen detection tests suggest that
virus transmission was present only in Facility A. The constant
detection of antigen-positive animals from the hospital and
breeding cages confirms serial virus transmission in the facility.
The serial transmission of virus was attributed to the facilitys
poor husbandry and infection-control practices, including reuse
of needles, unrestricted movement of personnel into and out
of infected housing units, damaged cages, which permitted
some animals to escape and roam freely, and the high incidence
of animal fighting in gang cages (due to directly connected
cages housing at least 8 animals). This cage configuration was
not observed in the other facilities. These same practices undoubtedly contributed to the higher overall mortality seen in
Facility A, even when laboratory-confirmed EBO virus mortality is removed.

/ 9d49$$se37

12-27-98 05:33:14

jinfa

This study and the epidemiologic study of the first monkey

outbreak in 1989 showed that gang-cage housing increased the
risk of monkey-to-monkey transmission [13]. Close physical
contact also provided the opportunity for fighting among the
animals. In addition, the high viremias observed during the
latter clinical stages in infected animals [14] would be sufficient
to allow animal-to-animal transmission by arthropods by purely
mechanical or flying needle means.
Following the 1989 1990 episodes, the Centers for Disease
Control and Prevention modified the requirements for procedures used in the transportation and quarantine of nonhuman
primates, including mandatory disease-control efforts, and they
included EBO virus testing requirements [15]. This was done
to strengthen infection control within quarantined primates and
to reduce exposure risks to humans occupationally in contact
with the primates. The containment of transmission to a single
animal cohort during the 1996 outbreak in the Texas quarantine
facility [8] strongly supports the effectiveness of these regulations. These recent importations of EBO-R infected monkeys
were detected while the animals were in quarantine, and, thus,
the potential for transmission was minimal to facility employees and other animals in the facility. Since August 1994, the
Philippines has banned the export of wild-caught monkeys and
introduced additional safety measures, including a 45-day quarantine prior to export [7]. This episode suggests that there was
a lapse in procedure, since the quarantine in Facility A failed
to pick up infected monkeys and did not prevent their export.
Five Filipino animal handlers have been found with EBOR antibody. During the 19891990 study, using the indirect
fluorescent antibody test with sera considered positive if the antibody titer was 1:256, 3 individuals were determined to be
antibody positive [5]. In 1993, 2 new EBO-R antibodypositive
individuals were detected, by use of the IgG antibody ELISA
(Miranda ME and Ksiazek TG, unpublished data). All of these
persons worked in Facility A, the source of the infected monkeys
in 19891990 and 1992. A sample from 1 of the 3 antibodypositive persons from the 19891990 study was retested in the
current study and found to be negative. The other positive persons
were no longer working in the animal facility during the current
study. In the present study, the IgG antibody ELISA was used.
One of the 2 people who were EBO-R IgGpositive during the
1992 study remained positive, while the other person was no
longer antibody positive. This indicates that, as determined by
our current ELISA technology, some antibody-positive persons
may revert to undetectable antibody status through time.
Results of this and other investigations in the Philippines
and United States show that humans can be infected with the
EBO-R virus [3 5]. However, transmission to humans has
been infrequent and did not result in any reported illness. This
suggests that infection results in a very mild or inapparent
illness and that EBO-R does not pose the same public health
threat as the African EBO virus subtypes.
Although we have confirmed the presence and transmission
of EBO-R within Facility A, the source of the virus was not

UC: J Infect

Downloaded from http://jid.oxfordjournals.org/ by guest on April 3, 2014

caught, and the other was born within the facility. All of the
remaining 1431 samples from the other four breeding and export facilities and all holding facilities were negative for detectable IgG antibody (table 3).
Human studies. Two hundred fifty persons were included
in the study. One of 18 employees in Facility A was positive
for IgG antibody to EBO-R. No individuals from the other
facilities had detectable antibody. The positive individual previously had been identified as positive in 1992. A blood sample
that was obtained from this personss spouse was EBO-R antibody negative. Blood samples were also obtained from 3 family
members of another employee who was antibody positive in
1992 but who had no detectable EBO-R IgG antibody in 1996;
the samples were antibody negative.

JID 1999;179 (Suppl 1)

JID 1999;179 (Suppl 1)

EBO-R in Philippine Primate Facilities

/ 9d49$$se37

12-27-98 05:33:14

jinfa

Acknowledgments

We extend our sincere gratitude to the members of the Primate

Exporters and Breeders Association of the Philippines, the Protected Areas and Wildlife Bureau, the Department of Environment
and Natural Resources, and the Bureau of Animal Industry of the
Department of Agriculture, Republic of the Philippines, for their
assistance in the conduct of this investigation; to Mary Lane Martin, A. J. Williams, Maurice Curtis, and David Bressler for technical assistance in laboratory testing; to Laura Morgan for data management assistance; and to Kent Wagoner for the graphics.
References
1. Jahrling PB, Geisbert TW, Dalgard DW, et al. Preliminary report: isolation
of Ebola virus from monkeys imported to USA. Lancet 1990; 335:502
5.
2. World Health Organization. Viral haemorrhagic fever in imported monkeys. Wkly Epidemiol Rec 1992; 67; 24:183.
3. Centers for Disease Control. Update: filovirus infection in animal handlers.
MMWR 1990; 39; 221.
4. Centers for Disease Control. Update: filovirus infections among persons
with occupational exposure to nonhuman primates. MMWR 1990; 39:
266 7, 273.
5. Miranda ME, White ME, Dayrit MM, Hayes CG, Ksiazek TG, Burans JP.
Seroepidemiological study of filovirus related to Ebola in the Philippines. Lancet 1991; 337:425 6.
6. Centers for Disease Control. Update: Ebola-related filovirus infections
in nonhuman primates and interim guidelines for handling nonhuman
primates during transit and quarantine. MMWR 1990; 39:22 4, 29 30.
7. Department of Agriculture/Department of Natural Resources (Republic of
the Philippines). Monkey controls (circular). Manila, Philippines, 1994.
8. Rollin PE, Williams RJ, Bressler DS, et al. Ebola (subtype Reston) virus
among quarantined nonhuman primates recently imported from the Philippines to the United States. J Infect Dis 1999; 179(suppl 1):S108 14.
9. Sanchez A, Ksiazek TG, Rollin PE, et al. Detection and molecular characterization of Ebola viruses causing disease in human and nonhuman
primates. J Infect Dis 1999; 179(suppl 1):S164 9.
10. Ksiazek TG, Rollin PE, Jahrling PB, Johnson E, Dalgard DW, Peters CJ.
Enzyme immunosorbent assay for Ebola virus antigens in tissues of
infected primates. J Clin Microbiol 1992; 30; 4:947 50.
11. Ksiazek TG. Laboratory diagnosis of filovirus infections in nonhuman
primates. Lab Anim 1991; 20:34 46.
12. Ksiazek TG, West CP, Rollin PE, Jahrling PB, Peters CJ. ELISA for the
detection of antibodies to Ebola viruses. J Infect Dis 1999; 179(suppl
1):S192 8.
13. Hayes CG, Burans JP, Ksiazek TG, et al. Outbreak of fatal illness among
captive macaques in the Philippines caused by an Ebola-related filovirus.
Am J Trop Med Hyg 1992; 46; 6:664 71.
14. Jahrling PB, Geisbert TW, Jaax NK, Hanes MA, Ksiazek TG, Peters CJ.
Experimental infection of cynomolgus macaques with Ebola-Reston
filoviruses from the 1989 1990 US epizootic. Arch Virol 1996;
11(suppl):115 34.
15. Centers for Disease Control. Requirement for a special permit to import
cynomolgus, African green, or rhesus monkeys into the United States.
Federal Register 1990; 55; 77:15210 1.
16. Fisher-Hoch SP, Perez-Oronoz GI, Jackson EL, Hermann LM, Brown BG.
Filovirus clearance in non-human primates. Lancet 1992; 340:451 3.

UC: J Infect

Downloaded from http://jid.oxfordjournals.org/ by guest on April 3, 2014

determined. It is unclear if low-level transmission of the virus

had been maintained within the facility or if there was recent
reintroduction of the virus from the wild. In February 1996,
the deaths (retrospectively confirmed to be due to EBO-R) of
2 animals from this facility (Miranda ME and Ksiazek TG,
unpublished data) demonstrated the unrecognized circulation
of the virus in the facility a month before the exportation of
infected primates to the United States. It is still uncertain if
monkeys are the natural host of EBO-R, although mortality is
high among those infected, and laboratory studies have found
that experimentally infected survivors do not harbor the virus
for even moderate time periods [16]. Following the 1989 1990
episodes of EBO-R associated with Facility A, all animals in
the facility were reportedly destroyed [13]; however, interviews
with individuals associated with the facility strongly suggest
that it was not completely depopulated. This leaves open the
possibility that there was some means of virus persistence
within animals that continuously populated the facility, perhaps
by slow serial transmission of the virus in the remaining primates and among newly introduced susceptibles, who were
added as breeding stock, and their progeny or by intermittent
transmission through medical fomites, such as multi-use vials
of tuberculin or other biologicals regularly used within the
facility. Alternatively, the virus may have been reintroduced
repeatedly into the facility with newly captured wild-caught
monkeys from the capture sites in Mindanao or by other
sources, such as bats, rodents, insects, or other animals that
may come into regular contact with the monkeys in the facility.
The viral genetic data of Sanchez et al. [9] is not incompatible
with any of these hypotheses for the maintenance of EBO-R.
In an attempt to contain the spread of infection, closer disease
monitoring and selective euthanasia was attempted in Facility
A in the weeks following this investigation. This entailed
weekly monitoring and the euthanasia of all animals housed
in the same cage unit as EBO-R antigen positive animals by
officers of the Philippine Department of Environment and Natural Resources and the Department of Agriculture. After the
depopulation of about half of the animal population, it was
noted that there was no significant improvement in infectioncontrol practices and there was poor compliance with the monkey mortality surveillance. Additional antigen-positive animals
were continuously being detected in new units. This implies
that EBO-R virus transmission continued in the facility and
had spread beyond the cages where the virus was originally
detected. By February 1997, the complete depopulation of the
facility with the eventual permanent revocation of the facilitys
permit to operate was executed by Philippine government authorities.

S119