Yellow Fever
WHO/EPI/GEN/98.11
ORIGINAL: ENGLISH
DISTR.: GENERAL
Yellow fever
Written by:
Dr Jari Vainio and Dr Felicity Cutts,
London School of Hygiene and Tropical Medicine
DIVISION OF EMERGING AND OTHER COMMUNICABLE
DISEASES SURVEILLANCE AND CONTROL
GLOBAL PROGRAMME FOR VACCINES AND IMMUNIZATION
EXPANDED PROGRAMME ON IMMUNIZATION
World Health Organization
Geneva
1998
The Global Programme for Vaccines and Immunization
thanks the donors whose unspecified financial support in 1997
has made the production of this document possible.
Ordering code: WHO/EPI/GEN/98.11
Printed : September 1998
This documents is available on the Internet at:
http://www.who.ch/gpv-documents/
Copies may be requested from:
World Health Organization
Global Programme for Vaccines and Immunization
CH-1211 Geneva 27, Switzerland
• Fax: +22 791 4193/4192 • E-mail: gpv@who.ch •
© World Health Organization 1998
This document is not a formal publication of the World Health Organization (WHO), and all rights are
reserved by the Organization. The document may, however, be freely reviewed, abstracted, reproduced
and translated, in part or in whole, but not for sale nor for use in conjunction with commercial purposes.
The views expressed in documents by named authors are solely the responsibility of those authors.
2
Yellow fever
Contents
Glossary ............................................................................................................................. v
List of abbreviations ....................................................................................................... vi
Preface ............................................................................................................................. vii
Summary .......................................................................................................................... 1
I. Introduction ............................................................................................................... 4
II. Historical review ....................................................................................................... 8
II.1. Pre-vaccination epidemiology 1700-1930 .................................................... 8
II.2. Pre-vaccination epidemiology in Africa ..................................................... 11
II.3. Development of vaccines .............................................................................. 12
II.4. ......................................... Early post-vaccination epidemiology 1940 - 1980
15
III. Epidemiology .......................................................................................................... 19
III.1. Vector reservoir, vertebrate maintainer and amplifier ............................ 19
III.2. Yellow fever virus ........................................................................................ 19
III.3. Transmission cycles and factors that affect them ..................................... 20
III.4. Distribution, ecological zones and types of transmission in Africa ....... 21
III.5. Recent epidemiology in Africa ................................................................... 24
III.6. Risk factors .................................................................................................... 27
III.7. Epidemiology in the Americas .................................................................... 28
III.8. Yellow fever and Asia .................................................................................. 30
IV. Cost effectiveness of yellow fever vaccination ................................................. 33
V. Surveillance ............................................................................................................... 36
V.1. Definition of surveillance ............................................................................... 36
V.2. Yellow fever sentinel surveillance in Kenya98 ........................................... 40
VI. Comments and suggestions ................................................................................. 44
Summary of recommendations .................................................................................. 48
Appendix I: Examples of historical yellow fever epidemics .................................. 54
Appendix II: Yellow fever cases reported in Africa 1900-1996 ........................... 58
WHO/EPI/GEN/98.11
3
Appendix III: African vectors .................................................................................... 63
Appendix IV: African vertebrate hosts ................................................................... 65
Appendix V: South American vectors and vertebrate hosts ............................... 67
Bibliography ................................................................................................................... 69
Maps and figures
Chart 1: Resurgence of yellow fever, Africa and Latin America, 1980-1995 ....... 6
Figure 1: Cycles of yellow fever transmission ........................................................... 23
Map 1:
Countries at risk for yellow fever and having reported at least
one outbreak, 1985-1998 ............................................................................... 7
Map 2:
Limits on endemic and epidemic areas of yellow fever ............................ 10
Map 3:
Reported yellow fever immunization coverage in countries at
risk for outbreaks, 1993-1995 .................................................................... 47
Tables
Table 1: Yellow fever vaccinations with French neurotropic vaccine and
cases of yellow fever in Africa, 1934-1953 ............................................... 16
Table 2: Milestones in the history of yellow fever ................................................... 18
Table 3: Transmission cycles, vegetational zones and vectors ............................... 22
Table 4: Epidemics reported 1984-1996 ................................................................... 26
Table 5: Ecological factors affecting yellow fever transmission ............................ 27
Table 6: Age and sex distribution of yellow fever cases in South America .......... 28
Table 7: Methods of surveillance for yellow fever ................................................... 39
Table 8: Attributes of yellow fever surveillance system in Kenya ........................ 42
Table 9: Yellow fever outbreaks, immunization coverage & performance in
African countries at risk for yellow fever outbreaks ................................ 49
Table 10: Prioritising 34 African countries at risk for yellow fever for support;
highest priority ............................................................................................... 50
Table 11: Prioritising 34 African countries at risk for yellow fever for support;
high priority .................................................................................................... 51
Table 12: Prioritising 34 African countries at risk for yellow fever for support;
medium priority ............................................................................................. 52
Table 13: Prioritising 34 African countries at risk for yellow fever for support;
lowest priority ................................................................................................ 53
4
Yellow fever
Glossary
Endemic disease:
The constant presence of a disease or infectious agent within a
given geographic area or population group.
Enzootic:
The constant presence of a disease within a given animal
population
Epidemic:
An outbreak of disease in human population
Epizootic:
An outbreak of disease in animal population
Information bias: Systematic error due to differences in accuracy or completeness
of recall of prior events or experience
Reservoir:
Any person, animal, arthropod, plant , soil, or substance, or a
combination of these, in which an infectious agent lives and
multiplies and where it reproduces itself in such a manner that
it can be transmitted to a susceptible host.
Selection bias:
Error due to systematic differences in characteristics between
those who are selected for study and those who are not
Strain:
A clone of organisms that differs in one or more inheritable
characteristics from other organisms assigned to the same
species.
Topotype:
Genetically distinct geographical variant of yellow fever virus2
Trophic:
Pertaining to an insect’s preference for the species to feed on:-
- anthropophilic:
Preference for feeding on humans
- primatophilic:
Preference for feeding on primates
- simiophilic:
Preference for feeding on monkeys
- zoophilic:
Preference for feeding on animals even
when human hosts are available
Vector:
An insect that transports an infectious agent from an infected
individual to a susceptible individual
Zoonosis:
An infection or infectious disease transmissible under natural
conditions from vertebrate animals to man. May be enzootic or
WHO/EPI/GEN/98.11
epizootic
5
List of abbreviations
AFRO
WHO Regional Office for Africa
CFT
complement fixation text
CIGN
country in greatest need
CRF
case fatality rate
DHF
dengue haemorrhagic fever
DMO
district medical officer
DTP
diphtheria-tetanus-pertussis vaccine
ELISA
enzyme-linked immunosorbent assay
EMC
Division of Emerging and other Communicable Diseases
Surveillance and Control
EPI
WHO Expanded Programme on Immunization
GPV
WHO Global Programme for Vaccines and Immunization
HI
haemagglutination inhibition (test)
HIV
human immunodeficiency virus
HRC
human resources coordinate
KEMRI Kenya Medical Research Institute
Ksh
Kenyan shillings
MOH
ministry of health
PAHO
Pan American Health Organization
PCR
polymerase chain reaction
PMO
provincial medical officer
RT-PCR
reverse transcription / polymerase chain reaction
UNICEF
United Nations Children’s Fund
YF
yellow fever
6
Yellow fever
Preface
This is a draft background document for the Yellow Fever Technical Meeting, Geneva,
March 1998 organized jointly by the WHO Division of Emerging and other
Communicable Diseases Surveillance and Control and the Global Programme on
Vaccines and Immunization’s Expanded Programme on Immunization.
We thank Dr T.P. Monath for extensive and detailed comments on an earlier draft.
WHO/EPI/GEN/98.11
7
8
Yellow fever
Summary
Yellow fever is a viral haemorrhagic fever which strikes an estimated 200 000 persons
world-wide each year and causes an estimated 30 000 deaths.3 Yellow fever virus is
the prototype of the family Flaviviridae, which currently contains over 70 viruses,
of which most are arthropod-borne, including the dengue viruses.4, 5 There are three
different epidemiological patterns of yellow fever virus transmission: the sylvatic or
forest pattern; the Aedes aegypti-borne urban cycle;6 and an intermediate cycle
that bridges these two patterns. The different epidemiological patterns of transmission
lead to the same clinical disease.7
The main vector of yellow fever within village and urban settlements is female Aedes
(Stegomyia) aegypti (only females feed on blood to obtain protein for egg production).
The virus is transmitted when a mosquito bites an infected human and then, after an
extrinsic (in the mosquito) incubation period of 12-21 days, bites a susceptible human.
Ae. aegypti breeds readily in all types of domestic and peridomestic collections of
fresh water, including flower vases, water drums, tin cans, broken coconut shells,
old tyres and gutters.4, 5, 7-9 In the forest pattern of yellow fever monkeys are the
primary host, and man is an accidental host (in South America yellow fever is an
occupational disease of people cutting down the forest9). Humans become infected
with yellow fever virus when bitten by the primary mosquito vector, Ae. africanus,
Ae. bromeliae or one of several other mosquito species. Most of these mosquitos
breed and live in holes and cracks in the upper part of the trees in the forest.4, 7-9
Intermediate epidemics are a mixture of man-to-man and monkey-to-man
transmission, and are often characterised by focal outbreaks separated by areas
without human cases.10 In some surveys, it has been possible to estimate an annual
incidence of infection of susceptible humans of at least 1%, so that, by adulthood,
immunity rates of 50% or more are not unusual.11 An attack of yellow fever is followed
by a solid, long-lasting immunity against reinfection.12
The incubation period in humans is generally three to six days after the bite of an
infected mosquito. The patient is only infectious to mosquitos for the first three to
four days after onset of symptoms.7 The disease is characterised by a sudden onset of
fever, headache, backache, general muscle pain, nausea, and vomiting.13 Milder cases
of yellow fever may not present with jaundice.7 There is a characteristic bradycardia
in relation to the temperature (Faget’s sign).6, 14 About 15% of those infected develop
a serious illness with several phases: an acute phase of about three days with sudden
onset of fever, headache, myalgia, nausea, and vomiting; remission for up to 24 hours
(characteristic “saddle-back” fever);14 and a toxic phase with jaundice and vomiting
(black vomitus), in which haemorrhagic signs (bleeding of gums, nose and haematuria),
albuminuria, and oliguria (reduction of urine production) may occur. The patient
may suffer from hiccups, diarrhoea, progressive tachycardia, and shock. Examination
WHO/EPI/GEN/98.11
9
of the abdomen reveals intense epigastric tenderness.7, 15 At least half of the individuals
who reach the toxic phase do not survive.7, 9, 10 Death usually occurs between the
seventh and tenth day after onset.7, 16
The possibility of yellow fever should not be dismissed in the absence of jaundice or
of albuminuria. Malaria and yellow fever may coexist in a region,17 and malaria usually
shows clinical symptoms almost identical with those of the early stages of yellow
fever: sudden onset, headache, generalised aches, and vomiting.7 Even with the finding
of malaria parasites in a blood smear, the possibility of yellow fever is not ruled
out.18 In the beginning of an infection, there is little to distinguish the illness from a
number of other febrile conditions. Typhoid fever, rickettsial infections, influenza,
leptospirosis, viral hepatitis, infectious mononucleosis, and other arboviral fevers
like dengue, Lassa fever and chikungunya may all resemble anicteric yellow fever.6,
7, 17
Later in the course of disease the following conditions must be taken into
consideration: the hepatitides, Weil’s disease, carbon tetrachloride poisoning, dengue
haemorrhagic fever, tick-borne relapsing fever, malaria or blackwater fever in addition
to several virus diseases with haemorrhagic manifestations (Argentine haemorrhagic
fever, Bolivian haemorrhagic fever, Crimean-Congo haemorrhagic fever, dengue,
Ebola fever, Kyasanur forest disease, Lassa fever, Marburg disease and Rift Valley
fever).17
The definitive diagnosis of yellow fever is made by serology or virus isolation, which
requires special reagents and techniques as well as expertise in the interpretation of
the test results. But prior to this, it is important that staff of the health facility are
alert to the possibility of yellow fever and have the means to collect appropriate
clinical specimens from suspected patients.7 Liver samples may be obtained from
fatal cases by the use of a viscerotome. The histopathological diagnosis is based on
eosinophilic degeneration of the hepatocytes leading to the formation of Councilman
bodies.10 In the 1930s, a viscerotomy program was instituted in South America. All
individuals who died after a short-term febrile illness had a liver punch specimen
taken by health officials and sent to specially trained pathologists.19 Liver biopsies
are not done in living patients because of the risk of severe haemorrhage. A
viscerotomy service has not been instituted in Africa.17
Yellow fever is endemic in 34 countries of Africa with a combined population of 468
million. Yellow fever vaccine, one of the earliest viral vaccines to be developed, has
proved safe and efficacious.5 The vaccine is transported and stored frozen.7 The
development of new protective additives have increased the thermostability of the
vaccine. The shelf life at -20o or 4oC is now up to two years, and the estimated half
life at room temperature is 10 months.5 However, once a vial is opened, the vial must
be kept cold and used within one immunization session and it must be discarded
after then (in this case, one imunization session is considered to be six hours).133 One
dose of yellow fever vaccine provides protection for at least 10 years and possibly
life-long.2, 20 A single dose will confer immunity in 95% of persons vaccinated.7
Four strategies have the potential to bring yellow fever fully under control in Africa:
epidemic control, mass immunization, routine childhood immunization and
surveillance.7, 21 In Africa, epidemic control often suffers from delays of two months
or more between the onset of epidemics and their recognition, partly due to the
1 0
Yellow fever
occurrence of the first cases in remote areas with few medical services and the
unfamiliarity of medical personnel with the disease. Responses to a possible outbreak
include collection and testing of specimens, epidemic investigation, emergency
vaccination, entomological investigation and vector control.7, 21 Emergency
vaccination takes place as soon as an outbreak has been confirmed, in an attempt to
limit the spread of infection by immunizing all persons in the focus, regardless of
their former immune status. Good surveillance is essential in all at-risk countries for
the early detection of cases which will allow fast action to control an outbreak. It has
often proved difficult to identify early, isolated cases before they trigger an epidemic
because of the difficulties of distinguishing yellow fever from diseases with similar
symptoms (e.g. malaria).20 Other potential problems with emergency campaigns
include difficulty in obtaining the large supply of vaccine, syringes and needles, and
sudden deployment at short notice of large numbers of health workers. Another
disadvantage is that immunity does not appear until seven days after immunization.10
It should also be noted that several operational difficulties in responding quickly to
outbreaks have been extensively documented for measles and meningitis, and this
strategy needs discussion.
One study estimated that including yellow fever vaccine in the routine EPI is more
cost-effective than conducting emergency campaigns in response to yellow fever
epidemics. That study was based on data from Nigeria, which is one of the countries
worse-affected by yellow fever, and it is possible that the costs and benefits of the
different strategies would be different in countries which have less frequent yellow
fever epidemics. No published studies have compared costs and benefits of preventive
mass YF vaccine campaigns (e.g. periodic vaccination of a wide age range) with
those of routine YF vaccination of infants. Nonetheless, in countries which have
recently experienced YF epidemics, there is likely to be a high level of natural immunity
in older persons, and it makes sense to concentrate efforts on implementing routine
YF vaccination of infants.
A potential scheme for prioritising countries in need of support specifically for yellow
fever control is presented for discussion and debate. Although the absence of reported
YF activity in a country in the “YF” zone does not mean that there is no risk of a
resurgence, it makes sense to focus activities first on those countries that have evidence
of recent YF activity and which have either not yet introduced YF vaccine into their
EPI or which have low coverage of YF vaccine. However, the prioritization of
countries and of areas within countries will need thorough discussion. Although in
most countries, nationwide YF vaccination is needed, countries such as Kenya,
Angola, and perhaps Mali have focal areas of YF activity, and can consider whether
vaccination can be concentrated in the districts at greatest risk. All countries in the
YF zone need better information on trends in the epidemiology of YF, and efforts to
improve YF surveillance should be intensified. The YF laboratory network needs to
be strengthened, and peripheral health workers need training in the use of the YF
clinical case definition.
WHO/EPI/GEN/98.11
1 1
I. Introduction
In 1988, the EPI Global Advisory Group reviewed the situation on yellow fever and
noted a relatively high incidence in children. It recommended that countries at-risk
for yellow fever (Map 1) should incorporate yellow fever vaccine into the routine
activities of the national immunization programme, and this was endorsed by a joint
WHO and the United Nations Children’s Fund (UNICEF) Technical Group on
Immunization in Africa.7 Due to a small risk of adverse reactions, yellow fever vaccine
should not be administered to children less than six months of age, so it is usually
administered at the time of the measles vaccination at nine months of age. Older
children should also be vaccinated routinely in areas at high risk for yellow fever
epidemics.2, 20, 22
Since the late 1980s, there has been a dramatic resurgence of yellow fever (chart 1).
Vaccination activities in many of the countries at risk, which include the poorest in
the world, are generally weak. Only five of 34 African countries at risk reported
yellow fever vaccine coverage data in 1996. Outbreaks were reported in several
countries in West Africa in 1994-1995, and in 1995, Peru experienced the largest
yellow fever outbreak reported from any country in the Americas since 1950.
The WHO therefore commissioned a literature review of yellow fever to provide
background material for assessment of the current strategies, focusing on the
following:
•
the epidemiology of yellow fever, particularly in Africa;
•
a review of yellow fever surveillance systems and their effectiveness;
•
a review of studies examining the cost-effectiveness of preventive yellow
fever vaccination programmes versus emergency vaccination programmes.
A Medline literature search was conducted, and further articles were obtained from
the bibliographies of papers reviewed. Published information on the surveillance of
yellow fever was complemented by a review of the Kenyan sentinel surveillance
programme conducted by Jari Vainio in July-August 1997 as part of his MSc
programme at the London School of Hygiene and Tropical Medicine.
In chapter II, “History of yellow fever”, the history of yellow fever and YF vaccines
is reviewed, focussing on the major epidemics which extended even to Europe and
North America that remind us the capability of yellow fever to spread to these
continents.
1 2
Yellow fever
In chapter III, “Epidemiology of yellow fever”, recent YF epidemiology is reviewed
with emphasis on data from the African region. The situation in the Americas is
briefly summarized, and the possibility of spread of YF to Asia discussed.
In chapter IV, “Cost-effectiveness of yellow fever”, the small literature on this topic
is reviewed, and the factors that are likely to affect costs of yellow fever vaccination
are discussed.
In chapter V “Surveillance”, the advantages and disadvantages of different methods
of yellow fever surveillance are discussed. A more detailed description of the unique
Latin American surveillance based on viscerotomy is given, and a summary of key
findings in the review of the Kenyan sentinel surveillance programme is presented.
The published results of mosquito and monkey surveys, which often only serve to
provide lists of names of vectors and primates, are summarised in Appendices III,
IV and V.
Finally, in the “Conclusion and recommendations” chapter, potential activities are
prioritized according to practical criteria, with the aim of stimulating further critical
discussion and comments.
WHO/EPI/GEN/98.11
1 3
14
Chart 1: Resurgence of yellow fever, Africa & Latin America, 1980-95
Afri
Africa ca
Latin A
mer
Latin Americaica
6 000
5 000
cases 4 000
GPV/EPI
d
r
t
e
o 3 000
p
e
f
r 2 000
Number of reported cases . oo 1 000
N
0
1980
1985
1990
19
Y
ellow fever
Year
Year
Estimated 200 000 cases annually!
WHO/EPI/GEN/98.1
Map 1: Countries at risk for yellow fever and having reported at least one outbreak, 1985-1998
a t le a s t on e o u tb re a k , 1 9 85 1 99 8
1
C ou n tri e s a t ri s k
R e p or te d o u tbr e a k
T h e de s i gn a ti o n e m p loy e d an d th e pre s e n tatio n of m a te ria l o n th i s ma p d o no t i mp l y the e x pr es s io n o f a ny o p in io n wh a ts o e ve r o n th e p a rt o f the s e cre ta ria t o f th e W or ld He a lth O rg a n iz a tion c o nc e rn i ng th e le g a l sta tu s o f a n y c o u ntry ,
G PV /E P I
The designation employed and the presentation of material on this map do not imply the expression of any opinion whatsoever on the part
te rri tory , c ity o r a rea o r o f its a u tho riti e s, o r c on c e rn in g th e d e li mita tion o f its fro n tie rs o r bo u n d a rie s . Do tte d l i n e s o n ma p s re pr es e n t ap p ro x im a te bo rd e r l ine s for w h ic h th ere m a y n o t y et b e fu l a g re em e n t.
GPV/EPI
of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the
15
delimitation of its frontiers or boundaries. Dotted lines represent approximate border lines for which there may not be full agreement.
II. Historical review
II.1. Pre-vaccination epidemiology 1700-1930
The first account of a sickness that can definitely be recognised as yellow fever
occurred in Guadeloupe and in Yukatan in 1648.23, 24
Slave trade in the seventeenth century formed an intimate bond between West Africa
and Spanish-Portuguese America. “Yellow Jack” was one of the most dreaded of
the diseases of the Atlantic trade routes; the legend of the “Flying Dutchman”, a
vessel doomed to haunt the seas around the Cape of Good Hope because yellow
fever broke out and no port would give her harbourage and all the crew perished, as
described by Sir Walter Scott, was inspired by stories of this disease.23
Lind’s account (1792) of fever aboard the vessel off the coast of Senegal in 1768 is
usually accepted as the first in which we can definitely recognise yellow fever in
Africa. No clinical description of the fever was given, but the evidence for its being
yellow fever was its occurrence first in men who had been ashore, and its apparent
propagation aboard ship. The first clinical report on yellow fever was published by
Schotte in 1782 on the “Synochus Atrabiliosa” in Senegal in 1778: “...the vomiting
continued...It became green, brown, and at last black, and was coagulated in small
lumps...A continual diarrhoea, with gripings, now took place, by which a great
quantity of black and putrid faeces were evacuated... The skin became now full of
petechiae...” 24
For more than two hundred years the tropical and subtropical Americas were subject
to devastating epidemics, while serious outbreaks occurred as far north as Boston
and as far away from the endemic centres as Spain, France, England, and Italy.12
Epidemics swept repeatedly over the West Indies, Central America, and the southern
United States decimating populations and paralysing industry and trade.12 In the
course of history Philadelphia suffered from 20 epidemics, New York 15, Boston 8,
and Baltimore 7.12
In 1848 Josiah Clark Nott (1804-1973) was the first to suggest that yellow fever
was spread by mosquitos: “We can well understand how Insects wafted by the winds
(as happens with musquitoes, flying ants, many of the Aphides, etc.,) should haul up
on the first tree, house, or other object in their course, offering a resting place; but
no one can imagine how a gas or emanation, entangled or not with aqueous vapor,
while sweeping along on the wings of the wind, could be caught in this way...” 25
But it was Cuban physician C. J. Finlay (1833-1915) who published in 1881 the first
really serious theory of the mosquito transmission of yellow fever:
1 6
Yellow fever
“1. The existence of a yellow fever patient into whose capillaries the mosquito is
able to drive its sting and to impregnate it with the virulent particles, at an
appropriate stage of the disease.
“ 2. That the life of the mosquito be spared after its bite upon the patient until it
has a chance of biting the person in whom the disease is to be reproduced.
“3. The coincidence that some of the persons whom the same mosquito happens
to bite thereafter shall be susceptible of contracting the disease” 26.
Because of the considerable difficulties caused by yellow fever for the American
army in Cuba during the Spanish-American war, the American authorities appointed
a Yellow Fever Commission with Walter Reed (1851-1902), an army surgeon, as its
head. In September 1900 the work of the Reed Commission proved conclusively
that
a) the mosquito was the vector of yellow fever;
b) there was an interval of about twelve days between the time that the mosquito
took an infectious blood meal and the time it could convey the infection to
another human being;
c) yellow fever could be produced experimentally by the subcutaneous injection
of blood taken from the general circulation of a yellow fever patient during
the 1st and 2nd days of his illness; and
d) yellow fever was not conveyed by fomites.12, 27
Reed and his co-workers suggested that the spread of yellow fever could be most
efficiently controlled by antimosquito measures and the protection of the sick from
the bites of mosquitos.27 The Commission also demonstrated for the first time that a
filterable virus caused a specific human disease.12 The conclusions of the Reed
Commission were confirmed in practice by Gorgas, who eradicated yellow fever in
Havana and Panama in early 1900 by depriving the mosquito of its breeding places.25
Laboratory work on yellow fever was very much handicapped by the lack of an
experimental animal. In 1927 Dr A. F. Mahaffy and Bauer of the Commission’s
laboratory staff managed to transmit yellow fever to an animal other than man using
blood from a yellow fever patient (a 28 year-old West African man named Asibi)
into a rhesus monkey.12 Propagation of the now famous Asibi strain of yellow fever
virus also began with this experiment.28
The same workers confirmed that12, 28, 29:
a) the causative agent of yellow fever was a filterable virus;
b) the infection was easily transmitted from monkey to monkey, or from man to
monkey, by injection of citrated blood taken from early in the disease;
c) that it was transmitted from monkey to monkey by Aedes aegypti mosquitos;
d) that once infected, mosquitos remained infective for the entire period of their
lives, which in some instances exceeded three months; and
e) that the bite of a single infected mosquito was sufficient to produce a fatal
infection in a monkey.
WHO/EPI/GEN/98.11
1 7
Dr Max Theiler described in 1931 the use of mice in testing sera for protective
substances against yellow fever virus. This mouse protection test became one of the
principal tools in yellow fever research and epidemiological investigations30, 31.
Serological surveys helped to delineate the areas in Africa in which yellow fever had
occurred.
Map 2: Limits on endemic and epidemic areas of yellow fever
Western topotype
Western topotype
Eastern topotype
Eastern topotype
Limit of potential epidem
Limit of potential epidemics
ics
Limit of endemic
Limit of endemic areaarea
GPV/EPI
1 8
Yellow fever
II.2. Pre-vaccination epidemiology in Africa
From 1906-1922 cases of yellow fever were apparently rare in former French Africa.
From 1922 to 1927 very numerous small outbreaks without apparent inter-
connection were reported in West Africa. In all these small outbreaks the infected
area was extremely localised.32
From 1927 to 1931, disease incidence decreased markedly and seemed to disappear
from one colony after another. In 1931, however, yellow fever reappeared. The
almost simultaneous reappearance of cases of yellow fever, with no connection
between them, in a large number of places scattered over West Africa and in countries
where the disease had not been reported at all for several years, was explained by
the persistence of latent yellow fever foci in these countries. In the epidemic periods,
it was the Europeans in particular who were affected as they had not gained protection
through a previous attack.33
The number of serological studies increased considerably after Theiler’s discovery
enabled mice to be used, instead of monkey Macacus rhesus, for protection tests.34
The results of these tests were often positive in Sierra Leone and Southern Nigeria,
and fairly frequently positive in Northern Nigeria. The prospecting mission of
Stefanopoulo in 1931-32 in former French West Africa discovered a good number
of positive sera in the west and south of Senegal, and along the upper course of the
Senegal River, in the Macina area (former French Sudan) and the former Upper
Volta Territory.34 Tests also gave positive results in the parts of former Togoland
under French Mandate. On the other hand, the tests gave negative results in almost
all the places studied in Guinea and the Ivory Coast (except Grand Bassam).
W.A. Sawyer’s researches showed high percentages of sero-positivity in Anglo-
Egyptian Sudan, and in the west of Uganda. The exceptional positive results obtained
in Kenya, Tanganyika and Northern Rhodesia were not considered as sufficient proof
that yellow fever had existed in these countries, since serology was not 100% specific
for yellow fever34.
In places where the disease was endemic, the proportion of positive immunity tests
increased fairly regularly with age, whilst in places where the disease appeared
sporadically the immunity curve according to age was irregular. Negative results for
tests among children indicated the absence of yellow fever during recent years from
the area or place in question. In the same way, the ages of children giving a positive
reaction to the test fixed the epidemic years.35 Van Campenhout’s report for the
former Belgian Congo showed that in the Matadi area, where yellow fever was
prevalent in 1928, both children and adults gave positive reactions in equal
proportions. On the contrary, above Leopoldville, where yellow fever was not notified
during the previous years, the sera from children were negative, whereas, in the case
of adults, the older the subjects, the more often were the results positive.33
The disease appeared in June 1934 for the first time farther east, at Wau, in Bahr-el-
Ghazal Province in former Anglo-Egyptian Sudan. Sero-protection tests reported
by Sawyer in 1931 had revealed one third of individuals (7/27) immune to the
disease.36
WHO/EPI/GEN/98.11
1 9
The information on the geographical distribution of the disease was more useful
than the number of cases recorded, since the number of typical recognised cases was
infinitely small compared with the benign atypical cases and subclinical infection in
the affected areas. The figures published in 1930s related almost exclusively to
Europeans, and often only one or two cases were reported from each affected locality.
Thus reports only served to show the presence of yellow fever and not its true
burden.36
Certain epidemiological features were noted by 192832:
1. Yellow fever followed the trade routes such as rivers, roads and railways.
2. The disease was pre-eminently urban.
3. Nonetheless, outbreaks often occurred in isolated spots in the jungle.
4. Almost inevitably, outbreaks followed the arrival of large numbers of non-
immunes, or the incursion of susceptible troops in infected territory, or other
mass population movements.
5. Newcomers to endemic foci suffered disease almost exclusively, with high
attack rates in non-immunes, while the indigenous population enjoying a
relatively very high degree of immunity.
6. Attack rates were higher when infected localities were visited at night.
II.3. Development of vaccines
Two live attenuated YF vaccines were developed in the 1930s; the French neurotropic
vaccine from human virus passaged in mouse brain and the 17D vaccine from human
virus passaged in embryonated chicken eggs.
Milestones in the development and use of French neurotropic vaccine are summarized
in the box. Between 1939 and 1952 over 38 million doses were administered (mostly
by scarification along with smallpox vaccine) in Francophone countries of West Africa,
and incidence declined dramatically (Table 1). However, a high incidence of
encephalitic reactions in children led to its use in children under 10 years being
stopped in 1961, and manufacture of the vaccine was discontinued in 1980.
II.3.1. Milestones in the use of French neurotropic vaccine2, 9, 37, 38
1927: One of the first strains of yellow fever virus was isolated at the Institut
Pasteur at Dakar.
1928: The virulent organs from an infected monkey were transported to
Europe and America, where they were placed at the disposal of various
laboratories under the name of “French strain”.
1931: The first trials on humans by simultaneous injection of a suspension of
the French strain and a certain quantity of human immune serum (the serum
was added to limit the potential virulence of the vaccine strain).
2 0
Yellow fever
1932: A method involving the subcutaneous inoculation of the modified
French strain alone, without immune serum, was introduced.
1941: A departmental order made yellow fever vaccination by scarification
compulsory for the whole civilian and military population of French West
Africa. Yellow fever virtually disappeared from colonial French West and
Equatorial Africa by virtue of a programme of compulsory immunization
initiated in 1942 (Table 1). The same period was marked by major epidemics
in the British colonies of Gold Coast and Nigeria, which had not implemented
a policy of preventive immunization.
1951-1952: During the epidemics in Panama, Honduras, and Costa Rica,
and again in eastern Nigeria, when the French neurotropic vaccine was used,
cases of post vaccinal encephalitis were seen. Encephalitis was reported to
occur in Nigeria at a rate of 3-4/1000 vaccinations, mainly in children, with
a case-fatalty rate of 38%.
1961: The French neurotropic vaccine stopped being recommended for
children under 10 years, because of the noted association with a high
incidence of encephalitis reactions in children.
1980: The manufacture of the French neurotropic vaccine was discontinued.
Today, 17D is the only type of YF vaccine produced. Its development and use is
summarised in box II.3.2. The immunological basis for its use has been reported
elsewhere.
II.3.2. Milestones in the use of 17D vaccine2, 9, 12, 15, 39, 40, 41
1936: The Asibi strain of yellow fever was successfully established in a culture medium
containing embryonic mouse tissue and 10% normal monkey serum in Tyrode’s
solution. After cultivation through 18 subcultures in this medium, cultivation of the
virus was initiated in a medium containing minced whole chick embryo. After 58
subcultures in the latter medium, the tissue component of the medium was modified
by removing the brain and spinal cord from the chick embryo before mincing. The
virus was later maintained continuously in this medium for over 160 subcultures.
The resultant strain was designated as 17D.
WHO/EPI/GEN/98.11
2 1
1937: Theiler and Smith reported on the use of the 17D strain for human
immunization.
1938: After a year’s experience in the production and application of yellow
fever vaccine made from the cultured 17D virus strain, Smith, Penna, and
Paoliello reported that there was available a practicable safe method of large-
scale immunization against yellow fever. Vaccination of 59 000 persons in
Brazil showed a) that mild reactions occurred five to eight days after
vaccination in 10-15% of the persons vaccinated, with more intense reactions
in only 1-2%, and b) that the vaccine was harmless even for children and for
women at any stage of pregnancy. Laboratory studies indicated that about
95% of the vaccinated had acquired immunity as measured by specific
antibodies.
1940: The first cases of jaundice and encephalitis as side-effects of 17D
vaccinations in Brazil were recorded. In August of 1940, the practice of
adding 10% normal human serum (necessary for the filtration of the virus)
to the vaccine was given up. However, serum was used in preparing vaccine
in the US, resulting in a major outbreak of hepatitis in the military in 1942.
The practice had resulted in the transmission of the virus of infectious
hepatitis, which for many years contaminated yellow fever vaccine.
1945: A 17D virus seed lot system was established to resolve the problem of
over-attenuation or under-attenuation of 17D vaccine.
1950-1953: Violent outbreaks of yellow fever occurred in southern Brazil.
During this period of intensive campaigns against epidemics about 12 000
000 people were vaccinated with 17D vaccine.
1951-1952: The occurrence of postvaccinal encephalitis in 15 infants from
UK, US and France formed the basis for a recommendation that excluded
use of 17D vaccine in infants under six months of age.
1958: The 17D vaccine was shown to induce very long-lasting immunity,
providing the basis for new recommendations regarding reimmunization of
travellers at 10-year intervals.
1966: Initiation of 17D manufacture in Dakar, Senegal in response to
encephalitis during the FNV vaccination campaign in 1965.
1976: WHO: no vaccine shall be manufactured that is more than one passage
level from a seed lot that has passed all safety tests.
1988: The Joint WHO/UNICEF Technical Group on Immunization in
Africa recommended incorporation of yellow fever vaccine in routine child
immunization programmes of countries at risk for yellow fever.
1997: There are 34 African countries at risk for yellow fever: 17 of these
countries have a policy to include yellow fever vaccine in the EPI.
2 2
Yellow fever
II.4. Early post-vaccination epidemiology 1940 - 1980
The first experiments in large-scale vaccination against yellow fever were conducted
in French Africa south of the Sahara. During 1934 and 1935, 5699 persons were
given three successive subcutaneous inoculations with the French neurotropic
vaccine.37, 38
Before mass immunization campaigns were started in Africa, typical urban outbreaks
occurred in Lagos, Nigeria, in 1925-1926, in Accra, Ghana in 1926-1927 and again
in 1937, and in Banjul , the Gambia, in 1934-1935.10 In 1940, mass immunization
was initiated in French-speaking countries in West and Equatorial Africa where 25
million people were immunized about every four years (Table 1). As a consequence,
yellow fever disappeared gradually in these countries, while epidemic and endemic
activity continued in countries without immunization programmes.37, 38
WHO/EPI/GEN/98.11
2 3
Table 1: Yellow fever vaccinations with French neurotropic vaccine and cases
of yellow fever in Africa, 1934-1953
Cases of yellow fever notified*
Year
Number of yellow fever
French West
Other
Total
vaccinations by
Africa and
African
for
scarification in French
Togoland
territories
Africa
West Africa and Togoland
1934
23
41
64
1935
12
16
28
1936
24
19
43
1937
48
122
170
1938
27
49
76
1939
101 633
15
43
58
1940
372 632
4
4
8
1941
2 018 954
17
19
36
1942
4 932 068
10
6
16
1943
7 890 417
12
20
32
1944
11 577 269
2
11
13
1945
14 563 092
1
17
18
1946
17 179 812
1
51
52
1947
20 289 249
3
1
4
1948
24 293 762
2
4
6
1949
28 662 214
0
37
37
1950
32 530 124
0
17
17
1951
36 789 119
2
39
41
1952
42 095 954
1
53
54
1953
46 391 582
2
28
30
38* Cases occurring during the 1940 epidemic of yellow fever in the Anglo-Egyptian
Sudan are not included.
The largest yellow fever epidemic ever recorded was in Ethiopia in 1960-1962,
affecting 10% of the 1 000 000 residents of south-western Ethiopia, a population
without a background immunity. The epidemic caused about 30 000 deaths. There
had been some yellow fever activity west of Ethiopia in the end of 1950s: in former
Belgian Congo, Sudan and Uganda. A curiosity was the appearance of many fatal
cases with a fulminating, two to three-day course without hepatic or renal signs.
2 4
Yellow fever
Entomological investigation implicated Aedes africanus in monkey-to-monkey and
low-level monkey-to-human transmission, with in addition, intense interhuman spread
by Aedes simpsoni. In 1964, an isolated human case in Uganda was thoroughly
investigated; the evidence also implicated A. africanus in monkey-to-human
transmission, and was supported by observations that this species descends to feed
at ground level during the day.9
In 1961, immunization of children under 10 years old was suspended in Francophone
West Africa. In 1965, yellow fever appeared explosively in a dry savannah region of
Senegal, affecting children born since the last cycle of routine vaccinations. Although
only 243 cases were officially documented, the true incidence may have been 20 000
cases, with a case-fatality rate of 10%. Efforts to control the epidemic resulted in a
tragic, iatrogenic outbreak of postvaccinal encephalitis. A mass campaign with
neurotropic vaccine was undertaken; 248 cases of encephalitis were identified, with
a case-fatality rate of 22%. The manufacture of the French neurotropic vaccine was
suspended in 1982.9
Epidemic activity continued in the 70s, but at a lower level than in the preceding or
succeeding decades. A small outbreak in Okwoga District, Nigeria in 1970 provided
the first evidence that A. africanus, the classic enzootic vector, was responsible for
interhuman epidemic transmission.9
The decreasing number of cases resulted in a lack of interest in yellow fever and
surveillance and immunization were progressively neglected after the early 1960s.
In 1971, yellow fever appeared in Angola for the first time in 99 years. The official
incidence understates the true impact of the epidemic, and a serosurvey indicated
that at least 13% of the urban population had been infected.9
Between 1977 and 1979, Ghana experienced a series of epidemics. As in other
epidemics in anglophone countries that had not practised preventive immunization,
there was a high attack rate in adults.9
The subsequent epidemiology will be discussed by region in later sections. Major
milestones in the history of yellow fever are summarized in Table 2.
WHO/EPI/GEN/98.11
2 5
Table 2: Milestones in the history of yellow fever
Year
Milestones
Comments
- 1700 - epidemics with uncertain YF diagnosis in
- discovery of “The New World”
San Domingo, West Africa, Cuba, West Indies
and Barbados in 1600s.
- the first generally accepted description of YF
- slave trade from West Africa to
in Guadeloupe, the other French Antilles and
Spanish-Portuguese America
in Yucatan in 1648.
- YF reached New York in 1668, Boston in 1691
and Charleston in 1699 by ship.
1700s - first generally accepted epidemic of YF in
- in 1730 an epidemic in Cadiz,
Europe.
Spain with 2200 deaths in
September and October
- first descriptions of YF in Africa.
- Lind’s account on YF in 1768 was
based on epidemiological
evidence in Senegal, while
Schotte’s publication in 1782 is the
first clinical YF report from Africa
1800s - YF was believed to be caused by poisonous
- devastating epidemics in the
“miasmata” from swamps, “effluvia” from the
tropical and subtropical Americas
filthy docks of the port, “breath of other people”
decimated populations and
or divine displeasure.
paralysed industry and trade
- ships arrived at European ports with YF on
- YF epidemics in Europe: Brest in
board
1802, 1839, 1856, St. Nazaire in
- theories of the mosquito transmission of YF
1865, Swansea in 1843, 1851,
by Nott (1848) and Finlay (1881)
1864 and 1865, Southampton in
1852, 1866 and 1867
1900 US Reed Board proved that YF is transmitted by mosquito Aedes aegypti
1900-
- causative agent of YF was isolated.
- Aedes aegypti was found to
1930
transmit YF from monkey to
- experimental laboratory animal is found:
monkey in 1927, A. luteocephalus,
first monkey in 1927 and later mouse in 1930.
A. apicoannularis and
Eretmapodites chrysogaster in
- urban outbreaks in Africa.
1928, A. vittatus, A. africanus,
A.simpsoni-group, A. scapularis
- entomological and ecological knowledge
and A.albopictus in 1929, and
increased.
Taeniorhynchus africanus in 1930
1930-
- serological surveys in Africa helped to
- mass immunization for YF was
1960
delineate the boundaries of the area in which
continued
the disease had occurred (see Map 2).
- the first YF vaccine in 1931.
- 17D and French neurotropic vaccines were
developed concurrently in 1930s.
- mass immunization campaigns began in
Brazil in 1938 and French-speaking countries
in West Africa in 1940.
1960-
- the decreasing number of cases of YF
- severe epidemics in Ethiopia in
1985
resulted in lack of interest in YF, and
1960-1962 and West Africa in
surveillance and immunization were
1969-1970
progressively neglected.
1986-
- 1986-1991 was an extremely active period for
- 17 of 34 at risk countries in Africa
YF.
have a policy to include YF vaccine
- YF vaccination (17D) was recommended to be
in the EPI
incorporated in routine child immunization
programmes of countries at risk for YF in 1988
2 6
Yellow fever
III. Epidemiology
III.1. Vector reservoir, vertebrate maintainer and amplifier
Yellow fever occurs in tropical areas of South America and Africa. Aedes aegypti-
infested areas of Central America, the Caribbean, North America, and Europe were
subject to introduction and spread of the disease up to the early part of this century
and must still be considered receptive areas.4
The reservoir of yellow fever virus is the susceptible vector mosquito species that
remains infected throughout its life and can transmit the virus transovarially.10 Yellow
fever can persist as a zoonosis in tropical areas of Africa and America, with nonhuman
primates responsible for maintaining the infection.10 Man and monkey play the role
of amplifiers of the amount of virus available for the infection of mosquitos.10
III.2. Yellow fever virus
The causative agent of yellow fever is an arthropod-borne virus from Flavivirus
genus of the family Flaviviridae. The virus possesses a single-stranded, positive-
polarity RNA genome. Viral particles are 43 nm in size; they are made up of a
ribonucleoprotein core and a lipoprotein envelope.10
Considerable heterogeneity between isolates from Africa and South America has
been observed among yellow fever strains.43 However, there is very little empirical
evidence for differences in virulence between wild strains of YF virus.
The prevailing view is that there are only two genotypes of YF in Africa and one or
possibly two in South America, found by sequencing wild-type yellow fever virus
strains of different geographic origin. The data base includes the entire genome
sequences of the Asibi and French viscerotropic viruses (Ghana and Senegal, 1927)
and partial sequences of the E gene, the 5’ and 3’ termini, and of the NS4a-NS4b
region of multiple isolates from South America and Africa isolated over a 60 year
period. Yellow fever strains in Africa fall into only two genotypes, one represented
by West African viruses and the other by Central and East African strains.44 South
American viruses fall into one major phylogenetic group with respect to the E gene
sequence. In contrast to the situation in Africa, the two South American genotypes
do not segregate into discrete geographic distributions, but one genotype has not
been recovered since 1974, suggesting that this virus may have been lost.
WHO/EPI/GEN/98.11
2 7
III.3. Transmission cycles and factors that affect them
III.3.1. Vertical, passive transmission in mosquitos by passage of virus from a
vector to its progeny
In 1981, the vertical transmission of yellow fever virus in Haemagogus equinus was
demonstrated.45 This followed Cornet’s 1979 recovery of YFV from male Ae. furcifer-
taylori in Senegal and Aitken’s experimental work showing vertical transmission in
Ae. aegypti.46 An explanation was now at hand for yellow fever virus to survive in
nature without the need to postulate alternate vectors, prolonged survival, retarded
transmission by long-lived, drought-resistant, adult female mosquitos, persistent
infections of vertebrates, or reintroduction of virus from distant enzootic foci.9
Recently, natural YF virus vertical transmission has been demonstrated in Ae. aegypti,
its epidemic vector, in Senegal. It was thought that vertical transmission played a
major role in the spread of the epidemic.47
The role of vertical transmission in nature has been proved by the isolation of several
strains of virus from wild-caught males of vector species.48 Its efficiency could be
increased by the possibility of venereal infection of females by males.49 By vertical
transmission, the vector can keep the virus for very long periods and is thus the true
reservoir.50 The occurrence of vertical transmission has two important epidemiological
implicatons. The first is that the virus can be transmitted only a few days after the
emergence of Ae. aegypti females, theoretically at the first blood meal, without being
delayed until the viral extrinsic cycle is completed 8 to 12 days later. Transmission in
the human population will be more frequent than if there were only horizontal
transmission. The second implication is that the YF virus can persist in the area until
the next rainy season inside infected eggs laid in peridomestic breeding sites which
dry up, such as used tyres and old pots.47
III.3.2. Horizontal, active transmission between vertebrates by passage of virus
from one vertebrate host to another through a vector in which the
virus replicates
This may occur in one of two ways depending on ecological factors which affect the
degree of contact with susceptible hosts 49:
1) Maintenance cycles, with a relatively stable prevalence of infection: vector-
vertebrate contact is loose and yellow fever will appear in enzootic or
endemic form.
2) Amplification cycles, with an increase of the amount of circulating virus:
vector-vertebrate contact is close and yellow fever will appear in epizootic or
epidemic form.
A number of ecological factors may affect horizontal transmission. The degree of
contact between vectors and susceptible vertebrate hosts, and thus the mode of
transmission depends on the amount of virus, the abundance of vectors and
vertebrates. The infection of the vector depends on the specific intrinsic relations
between the virus and its invertebrate host (e.g. dissemination of the virus in the
invertebrate host: crossing the gut barrier, invading different tissues), but also the
extrinsic factors which are independent of the virus: the vector must become infected
2 8
Yellow fever
after a blood meal on an infected vertebrate host, the virus must replicate in the
tissues of the invertebrate host and after, the virus must be inoculated with saliva
into another vertebrate host. The invertebrate host must thus live long enough for
the virus development inside its body. The mosquito must have trophic preferences
for primates to act as a vector in nature.
III.4. Distribution, ecological zones and types of transmission in Africa
III.4.1. Vegetation
The distribution of yellow fever in Africa is best understood in terms of vegetational
zones which reflect rainfall patterns and determine the abundance and distribution
of mosquito vectors and vertebrate hosts (Table 4).
III.4.1.1. Equatorial rain forest (Enzootic; mainly sylvatic)
The great equatorial rain forest zone extending from Guinea in the west to Uganda
in the east and south to Equatorial Guinea and northern Angola is the zone of year-
round enzootic yellow fever transmission between monkeys and A. africanus. The
virus activity is generally at a low level and sporadic cases or focal outbreaks are the
rule, in a manner analogous to jungle yellow fever in South America. Transmission is
predominantly monkey-to-monkey, and human infection is sporadic (Figure 1).
III.4.1.2. Humid/semihumid savannah (emergency zone; cyclic epizootics and
epidemics; either monkey-to-monkey or monkey-to-man transmission;
major area of risk)
Extending outwards from the rain forest zone, with decreasing rainfall, are found in
sequence savannah-forest mosaic and moist (Guinea) savannah. During the rainy
seasons these regions are prone to repeated emergence of yellow fever activity, which
may occur at a high rate of transmission due to the presence of vector and host
populations. Sylvatic Aedes (e.g., A. furcifer, A. luteocephalus, A. vittatus) reach
very high densities during the rainy season, and are responsible for cyclic epizootics
in monkey populations and epidemics with interhuman transmission.51 This zone is
also known as the intermediate zone of transmission (fig 1). Vertical transmission in
these mosquitos assures virus survival and continuation of epizootic waves. It is in
this vegetational zone that most epidemics of yellow fever have occurred. There
may be focal outbreaks separated by areas without human cases.
III.4.1.3. Dry savannah (mainly man-man transmission; potential for
epidemics)
In the dry savannah zones the rainfall is very low and the rainy season abbreviated.
The sylvatic vector populations are too low or active for too short a period to sustain
an epizootic. The virus may nonetheless be introduced into a cycle of interhuman
transmission by Aedes aegypti, either if an epizootic extends from the humid
savannah, or if infected individuals move to villages with the domestic vector in the
dry savannah. If the virus is introduced into urban or very dry savannah regions
where the human population stores water and lives in association with domestic A.
aegypti, explosive outbreaks of A. aegypti-borne yellow fever (urban-type
transmission, fig 1) may ensue.9 Usually the outbreak spreads from village to village
following the lines of communication used by humans. When the epidemic has started,
the virus can be transported to distant places either by infected persons or by infected
mosquitos.
WHO/EPI/GEN/98.11
2 9
Table 3: Transmission cycles, vegetational zones and vectors9, 49
Rainforest
Moist savannahs
Dry savannahs / urban areas
Enzootic area
Zone of emergence
Epidemic zone
-cyclic epizootics & epidemics-
Endemic area
Potential epidemic area
Jungle yellow fever
Intermediate
Urban yellow fever
monkey-mosquito-monkey
monkey-sylvatic Aedes-human
human-mosquito-human
(human infection is sporadic
(both humans and wild
and often unrecognised)
vertebrates are involved in the
virus cycle)
Aedes africanus
Aedes furcifer
Aedes aegypti
Aedes luteocephalus
Aedes metallicus
Aedes neoafricanus
Aedes opok
Aedes simpsoni group*
Aedes taylori
Aedes vittatus
* Probably Aedes bromeliae.
III.4.2. The vectors in Africa
The main vectors of yellow fever in Africa are mosquitos of the genus Aedes,
subgenera Stegomyia and Diceromyia. Seven species are thought to play an important
role in nature: Aedes (Stegomyia) aegypti, A. (Stegomyia) africanus, A. (Stegomyia)
opok, A. (Stegomyia) luteocephalus, A. (Stegomyia) simpsoni group, A. (Diceromyia)
furcifer, and A. Diceromyia) taylori.49
The eggs of the vectors are resistant to desiccation; they remain quiescent during the
dry season and hatch only when rain fills the breeding places. In savannah areas
there are no adults during the dry season and transmission is discontinuous.
Aedes-vectors can be classified into three categories according to their contact with
humans:
a) domestic (i.e. around the household) - mainly A. aegypti
b) wild - all other species
c) semi-domestic - wild vectors which can acquire domestic habits - A. furcifer,
A. africanus, A. luteocephalus.
3 0
Yellow fever
Figure 1: Cycles of yellow fever transmission
Yfcycles
South America
Haemagogus
Aedes aegypti
Monkey
Monkey
Human
Human
Haemagogus
Aedes aegypti
Jungle yellow fever
Urban yellow fever
Africa
Aedes africanus
Sylvan Aedes spp*
Aedes aegypti
?
Monkey
Human
Human
Monkey
Human
Aedes africanus
Sylvan Aedes spp*
Aedes aegypti
Jungle yellow fever
Sylvatic yellow fever
Urban yellow fever
Rainforest zone
Moist savannas
Dry savannas and
Forest savanna
urban areas
ecotone
* West Africa: Ae. furcifer, taylori, Ae. luteocephalus, Ae. africanus, Ae. opok, Ae. vittatus, Ae. metallicus
East Arica: Ae. bromeliae, Ae. africanus, others
III.4.3. Animal vertebrate hosts in Africa
In 1928 Stokes described the susceptibility of an Asiatic monkey, Macacus rhesus,28
which became the first laboratory animal. In Africa, almost all zoological groups
have been studied, but only primates are implicated in the natural transmission cycles
of yellow fever virus, because other animals have low viraemia and/or lack of contact
with known vectors.
WHO/EPI/GEN/98.11
3 1
Monkeys remain the main vertebrate hosts involved in the circulation of yellow
fever virus in Africa; galagoes (bush babies) may also play an important role. The
viraemia developed by monkeys is always short, two to five days, with a maximum
of nine days.52 After infection they have life-long immunity, so they cannot be virus
reservoirs.49 Those monkeys which stay in the canopy (top of the forest trees) will
be the main vertebrate hosts in the wild cycle (e.g. Cercopithecus mitis), while those
which come to ground level (Cercocebus) or leave the forest to enter plantations
(Cercopithecus aethiops) will be the link between the wild cycle and humans. In
savannah areas, monkeys usually live at ground level, but sleep in trees where they
are exposed to mosquito bites. There, monkeys such as patas or baboons easily
disseminate virus because their territory is very large.49
III.5. Recent epidemiology in Africa
The period 1986-1991 was an extraordinarily active period for yellow fever in Africa.
The world wide total of 20 424 reported cases and 5447 deaths represented the
greatest yellow fever activity reported to WHO since reporting began in 1948
(Table 5 / Map 3).53, 54, 55
The largest number of cases was reported from Nigeria, where a resurgence of yellow
fever has been noted since 1984. In 1986 and 1987, the Nigerian Ministry of Health
and WHO supported two epidemiological studies to try to determine the extent of
YF outbreaks. One was in Oju, one of two major epicentres in the 1986 sylvatic YF
epidemic, the other being in Cross-River state. In 1986, surveys in treatment centres
and nine villages in the Oju area of Benue state established an overall attack rate of
4.9% and a mortality rate of 2.8%. The population at risk in Oju was 200 000, thus
the study suggested that 9800 cases and 5600 deaths occurred in Oju. Official 1986
figures indicated 559 reported cases and 200 reported deaths for all of Benue state,
thus under-reporting in the state was at least 17 and 28 times respectively.
In 1987, during surveys in 17 hospitals and three villages in Oyo state, 3.6% of the
60 000 village residents were interviewed. Results indicated an attack rate of 2.9%
and a mortality rate of 0.6%. The 1987 outbreak was in a densely populated area,
and was an urban type spread by Aedes aegypti. The population at risk was estimated
at four million, thus 116 000 cases and 24 000 deaths were estimated to have occurred
in Oyo state; 130 and 50 times respectively those reported.
From 1984 to 1993, Nigeria reported over 20 000 cases and 4000 deaths. Because of
under-reporting, YF is estimated to have affected at least one million people in Nigeria
over this resurgence. In 1994, another outbreak in Imo State, Nigeria, spread to
neighbouring districts of Cameroon.
The first epidemic of YF in Cameroon occurred in 1990, during the second half of
the rainy season. There were 180 known cases, of which 125 died. The affected area
was in the yellow fever belt, situated around latitude 11 degrees North and 14 degrees
East. This is a mountainous area of scattered villages. A serosurvey in 11 villages
found 20% YF-IgM-positive individuals among 107 tested; most were aged under-
10 years. IgM assays for other flaviviruses were negative, while there was substantial
cross-reactivity in IgG. It was estimated that less than 4% of cases had been reported
and that the real number of cases could have been between 5000 and 20 000 with
500-1000 deaths.56
3 2
Yellow fever
The years 1992 and 1993 were relatively mild in terms of the total numbers of cases
of yellow fever, but the first outbreak ever recorded in Kenya was documented. The
outbreak was the jungle type (monkey-man transmission), and affected predominantly
young men, but concerns that the disease could spread to cities, where Aedes aegypti
was present, led to mass vaccination of almost one million persons in affected districts.
This represented the first report of YF from East Africa for almost 50 years. The
incidence of yellow fever increased slightly in 1994 and 1995, compared with the
previous two years, yet remained lower than incidences reported before 1992.21
From November 1994 through January 1995, Gabon reported its first outbreak ever.
The outbreak started as jungle-type YF in a remote mining camp in the forest, but
spread rapidly to villages outside the forest where A. aegypti mosquitos were present,
indicating a shift to person-person transmission. This epidemic was identified as Y.F.
according to first serological and reverse transcription/polymerase chain reaction
(RT-PCR) results, but no virus was isolated. Subsequently sera were obtained from
37 people who lived in the same region and presented symptoms compatible with
YF. In ten, YF virus RNA was obtained by RT-PCR. Nucleotide sequence of two
regions of the RNA in three sera differed from that of the Asibi strain of YF, and the
presence of a new topotype was hypothesized.57
In late 1995, an outbreak was detected in Liberia. The first case was in a Nigerian
soldier of the West African Peace Keeping Force stationed in Buchanan. By the end
of 1995, 360 cases and 9 deaths were reported, and a yellow fever case had been
confirmed in Sierra Leone. Mass immunization campaigns delivered nearly one million
YF vaccine doses in response to this outbreak.
In Senegal, an outbreak of intermediate-type YF occurred in 1995, killing at least 46
humans among an estimated exposed population of 9000. It was quickly stopped,
following a prompt immunization campaign (quoted in Fontenille).
All the outbreaks occurred in the vicinity of the emergence zone in the moist
savannah, or in the dry savannah, along the genuine “yellow fever belt” stretching
from Senegal to Ethiopia and Kenya. The rural populations were the most affected,
and the vector in the initial phases was frequently an anthropophilic wild mosquito
(A. africanus, A. bromeliae, A. furcifer). The towns, however, were not spared
(e.g. Luanda, Angola 1971, several cities of south-west Nigeria 1987, Buchanan,
Liberia 1995).
In many countries, only part of the countries concerned is exposed to the risk of
severe outbreaks. However, experience in Ghana 1977-1979 and Nigeria 1986-1987
shows that yellow fever can be transported from an epidemic focus to a distant area
with different climate and environment, and there produce a secondary epidemic if
the conditions are favourable.
Despite the clear indication of the potential risk from YF, it is not widely recognised
as an endemic problem in Africa. For the most part, only case clusters and outbreaks
are recorded. This reflects the insensitivity of reporting, as endemic virus transmission
must certainly occur. Nonetheless, the actual disease burden from this endemic
infection is difficult to estimate - see section V.
WHO/EPI/GEN/98.11
3 3
34
Table 4. Epidemics reported 1984-1996 21 53 54 55 58
Country
Year
Cases
Age
Sex
Place
Time
Occupation
Other comments
Angola
1988
37
Luanda (urban)
- 85% of population of Luanda was vaccinated in response
Benin
1996
124
Atakora-Borgou
July-
Burkina faso 1984-1985
24
South-east
- cases notified on the basis of clinical data alone; they occurred at
the
same time as an outbreak of hepatitis.
Cameroon
1990
173
51% <5,
Northern
children
79% <10
Cameroon
1994
10
Adamaoua Prov.
Nov. - Dec.
- outbreak limited due to its occurrence within the dry period
Gabon
1994-1995
44
4% <15
Majority
north-eastern
Nov. - Jan.
forest workers
- heavy rains preceded
male
mine area
Ghana
1993-1994
118
Male 67%
Upper West: Jiripa Oct. - May
Dist.
Ghana
1996
27
Upper East
Guinea
1987
5
Siguiri (border)
- connected to Malian epidemic
Kenya
1992-1993
54
33% <19
Male 65%
Kerio Valley
Sept. - March young male
- first epidemic reported from Kenya
Liberia
1995
360
Buchanan
November
- vaccination coverage in Buchanan reached 80% in response
M a l i
1987
305
70% <15
Male 62%
near Bamako
Sept. - Nov.
children
- 1969 mass immunization protected adults
- urban risk: infected Aedes aegypti was found in Bamako
Mauritania
1987
21
south-west
Oct. -Dec.
- connected to Malian epidemic, but coinceded with an epidemic of
Rift Valley Fever
Nigeria
1986-1992
18 940
50% adults Male 58%
Oju, Oyo, Kano,
- the population as a whole was susceptible (because of the
abscence
in 1987,
in 1987,
Kaduna, Bauchi,
of regular vaccination, and long time from the last epidemic).
46% in
52% in
Ipetu-Iyesa
1991
1991
Nigeria
1993-1994
152
9% <5
Male 57%
Imo State
Oct. - Jan.
18% <15
Nigeria
1994
1227
Imo State
Sept. - Dec.
Niger
1990
- spread from epidemic in Nigeria. First outbreak ever in that country
Y
ellow fever
Senegal
1995
79
Male 53%
Kounglel Dist.
Oct. - Nov.
- 8000 people exposed. Prompt immunization campaign conducted
Sierra leone
1995
33
Eastern Prov.
Nov. - Dec.
- a single case of YF was serologically confirmed within weeks of the
YF outbreak in the bordering area of Buchanan, Liberia (suspected
cases 33).
III.6. Risk factors
The distribution of cases by age depends on the immune status of the population at
the time of the outbreak. When the entire population is devoid of natural or
vaccination-induced immunity, the distribution of cases parallels the demographic
distribution (e.g. Nigeria 1987). On the other hand, when the population has been
subjected to an epidemic and/or a mass vaccination campaign several years earlier,
the adults still have some protection and are relatively less affected by the epidemic.
For example, in 1969, the population in the northern area of Ghana was immunized
in a large campaign conducted in response to an outbreak of YF. When YF recurred
in the same area in 1977-1980, the epidemic involved mainly children under 15 years
of age, who were too young to have been immunized in 1969. Thus 67% of cases
and 82% of deaths in 1977-1980 occurred in this age group. Similar phenomena
occurred in several African countries including Burkina Faso 1983; Mali 1987;
Cameroon 1990, where over 70% of cases were in children under 15 years of age.
This led to the EPI recommendation that YF vaccine be included in the routine EPI
in countries at risk for yellow fever.
Climatic conditions affect both the abundance of the vector and the incubation period
(the time between the infecting blood meal and the time of the first transmission)
which is shorter with higher temperatures. Human activities can also influence
transmission by action on host abundance, either negative (hunting monkeys and
thus reducing the number of hosts; mosquito control), or positive (creating artificial
breeding places; overpopulation). Forestry practices such as felling trees may increase
transmission by bringing the treetop-dwelling mosquitos down closer to human
contact (Table 5).
Table 5: Ecological factors affecting yellow fever transmission
Virus
- amount of virus in the beginning of amplification cycle
- virulence
Vector
- abundance
- longevity
- trophic preferences
- number of blood meals/day
- length of incubation of YF virus in the vector
- vector competence
Vertebrate
- abundance
Host
- immunity rates
- susceptibility (duration, height of viraemia)
Climate
- temperature
- humidity
- duration of rainy season
Human behaviour
- hunting monkeys
- creating artificial breeding places for vectors (pots, tyres, etc.)
- forestry practices
- population growth
- urbanization
- migration
- political unrest
WHO/EPI/GEN/98.11
3 5
III.7. Epidemiology in the Americas
Two types of epidemiological cycles have operated in South America: jungle and
urban. The last urban-type epidemic was in 1928-1929 in Rio de Janeiro (the last
case of urban yellow fever was notified in 1942). The two forms differ in that the
urban form is transmitted by A. aegypti while the jungle yellow fever is transmitted
by the bite of a Haemagogus or other forest-breeding mosquito that was previously
infected by feeding on an infected vertebrate host.49 In 1949, the ten countries most
afflicted by urban yellow fever (Brazil, Bolivia, British Guyana, Colombia, Ecuador,
French Guyana, Panama, Peru, Surinam and Venezuela) unleashed a vast campaign
against A. aegypti, destroying its urban breeding grounds, that by 1965 had eradicated
the mosquito and the disease from most urban areas in the continent. 59 But
A. aegypti has now reinfested most of Central and South America, and occupies
habitats adjacent to the areas where endemic yellow fever transmission occurs.53
The endemic zone corresponds to rain forests drained by the great river systems,
where the yellow fever virus is circulating “silently” among the monkeys, but the
emergence of human cases is rare.60 Epizootic waves exhaust susceptible nonhuman
primate populations, and five to 10 years may elapse before a sufficiently large non-
immune population of slow-breeding primates is reconstituted to permit recurrent
virus transmission.
Over the last 25 years, a mean of 115 cases have been reported annually from Latin
America. In 1990s, five American countries have reported yellow fever: Brazil,
Bolivia, Colombia, Ecuador and Peru. Bolivia and Peru has accounted for 82% of
the human cases (PAHO, 1998. Personal communication). In 1995, Peru reported a
jungle-type outbreak with 440 cases and a case fatality rate of 38%. This was the
largest outbreak in the region since the 1950s. Yellow fever affects mainly
unvaccinated people who enter the forest for hunting, fishing, or wood cutting and
become infected within the sylvatic cycle - about 80% of cases are reported in young
adult male forest workers. Thus yellow fever in South America can be considered an
occupational disease.15
Table 6: Age and sex distribution of yellow fever cases in South America
Country
Year
0-1 years old 1-15 years old More than 15
Male: female
ratio
Bolivia
1997
0
13
87
2.9 : 1
Brazil
1996-97
0
16
84
2.4 : 1
Peru
1997
0
10
90
3.2 : 1
Total
0
15
85
3 6
Yellow fever
The apparently low virus circulation for a long period in the Americas may have
been due to massive campaigns of vaccination and to vector control. The situation
can suddenly reverse as observed in the past with yellow fever.61 Aedes aegypti is
present in urban areas in the Americas (including southern parts of the USA).15
Aedes albopictus, which was discovered in the Americas in mid-1980s, and was
probably imported from Asia in used car tyres62, and which seems to have a great
adaptive capacity, is now in an intermediate position between the forest galleries and
the urban areas infested with Aedes aegypti,60 and thus increases the risk for
introduction of yellow fever to the urban environment.62
Concerns about the potential for YF epidemics to increase in Latin America were
highlighted in two recent reports from Brazil. Yellow fever virus transmission was
very active in the rainy season in Maranhao State in Brazil in 1993 and 1994.63 In
1993, of 932 people examined from Maranhao, 70 were positive for YF serologically,
histopathologically, and/or by virus isolation, and another four cases were diagnosed
clinically and epidemiologically. In Mirador (17 565 inhabitants), the incidence was
3.5 per 1000 people while in a rural yellow fever risk area (14 659 inhabitants), the
incidence was 4.2/1000; 45.2% of 62 infections were asymptomatic. In 1994, 49
serum samples were obtained and 16 infections were confirmed (two by virus isolation,
two by seroconversion, and 12 by serology). The investigation suggested that this
was the most extensive outbreak of yellow fever in the last 20 years in Brazil, and
was related to lack of vaccination.
Aedes aegypti, after having been eradicated in 1954, reappeared definitively in 1976-
1977. A second Flavivirus vector, Aedes albopictus has been present in Brazil for
about 10 years in some States, including Sao Paulo. Analysis of the distribution of
YF cases between 1972 and 1994 showed two epidemiologic regions.64 In the first
region, the endemic area, the YF virus circulates “silently” among monkeys, and the
emergence of human cases is rare. In the second region, the epidemic area, some
epizootics occur in a more or less cyclic way, and human cases can be numerous.
Nevertheless, these outbreaks are considered “sylvatic” epidemics, since as Ae.
aegypti is not involved. From the Amazonian region, the virus moves forward along
the forest galleries of the Amazone tributaries, from North to South. Dengue
epidemics appear in all States, and reflect the geographical distribution of Ae. aegypti.
Recently, Ae. aegypti was found in the southern part of the Para State, in the Carajas
region considered to be the source of the main YF epidemics. Furthermore, Ae.
albopictus is now increasing its distribution area, specially in the suburban zones.
This potential vector has an intermediate position between the forest galleries, where
the YF virus circulates, and the urban agglomerations infested with Ae. aegypti.
Thus, the importance of intensified surveillance of the epidemiological situation of
YF in Brazil was stressed.
WHO/EPI/GEN/98.11
3 7
III.8. Yellow fever and Asia
In 1934 Dudley was already concerned about the possibility of yellow fever to spread
from its endemic home in West Africa to the coast of East Africa and from there to
Asia.65 Since then there have been epidemics in Sudan in 1940,66 Ethiopia in 1960-
6267 and Kenya in 1992-93,68, 69 but yellow fever never spread to Asia. The following
reasons have been postulated but none provides a completely satisfactory explanation:
1) yellow fever was never introduced to Asia, 2) humans vary in susceptibility, 3)
there is cross-protection between flaviviruses, 4) the maintenance cycle is absent, or
5) there is variation in vector competence and/or behaviour (C. Leake, 1997, personal
communication).
III.8.1. Yellow fever was never introduced in Asia?
There have been multiple opportunities for the introduction and spread of yellow
fever to Asia. The opening of the Panama Canal in 1914 brought Asiatic ports into
more direct contact with the old endemic homes of the disease.32 The chances for
yellow fever to be introduced in Asia increased with the marked increase in air
travel since the 1960s. An average of about 200 000 passengers a year disembarked
at Calcutta airport in 1982-1988, and with about 25% of the passengers it was not
noted whether they possessed a valid yellow fever vaccination certificate or not.70
III.8.2. Humans vary in susceptibility?
There is no convincing evidence of innate differences in susceptibility to YF. The
apparently lower susceptibility to YF of the Indian population appears to reflect
cross-protection from other flaviviruses. Out of 876 specimens of human sera
collected from Australia, Ceylon, China, Java, India, the Malay States, The Philippine
Islands, and Syria in 1937, only two gave protection against yellow fever. Both of
them came from India, and as far as could be learned neither of the donors had ever
been exposed to yellow fever.71
In the 10th Regiment of Infantry during the Napoleonic wars, out of 408 officers
who contracted yellow fever in Gibraltar in 1815 after service in India, only four
died, in contrast to 21 deaths in 55 men who had not been to India (case-fatality
rates respectively 1% and 38%).72 Following slave emancipation in the British West
Indian colonies in 1838, hundreds of thousands of workers were brought from India
for the sugar plantations. These Indian immigrants were little affected by the outbreaks
of yellow fever which occurred among new arrivals from Europe.73
The issue of genetic or racial differences in susceptibility deserves further analysis.
Genetic determinants are known to affect the pathogenesis of flavivirus infections,
and resistance to yellow fever virus in mice is determined by an autosomal dominant
allele (Flv). Genetic background has been shown also to influence the immune
responses to flaviviruses in mice. The role of genetic factors in human responses to
yellow fever infection is uncertain. The older literature makes repeated reference to
racial differences in the lethality of yellow fever, rates being lower in blacks than
whites during outbreaks in West Africa, tropical Africa and the US.74, 75 It is uncertain
whether the apparent increased resistance of blacks reflects acquired immunity or is
due to genetic factors. In the case of dengue, whites had a higher incidence of dengue
haemorrhagic fever than blacks during the 1981 epidemic in Cuba, a finding that
3 8
Yellow fever
could not be explained on the basis of a racial difference in the background of
immunity. An association between HLA haplotype and disease severity also was
found in patients with DHF. The question of racial differences in susceptibility to
yellow fever will be resolved only by well-controlled epidemiological and serological
studies in the setting of an outbreak affecting both races. (Monath, T.P. Written
communication, 1998)
III.8.3. Flavivirus cross-protection?
Serologic cross-reactions between flaviviruses lead to difficulties in laboratory
diagnosis and cross-immunity to other flaviviruses has been observed to influence
susceptibility to other flavivirus infections. Some measure of protection may be
associated with antibodies to epitopes shared by distantly related flaviviruses.76
A small study of response to YF vaccination of Malay soldiers around 1960 found
that most soldiers had antibody prior to vaccination that cross-reacted with YF
antibody assays. Response rates after vaccination were equal in those with or without
prior antibody, but the antibody levels achieved were lower in the former.77
It has been thought possible that dengue immunity protects against clinical yellow
fever, or by reducing viraemia, decreases the possibility of secondary spread following
a chance introduction.4 However, the mouse-protection test gave no support to the
suggestion that an attack of dengue protects against yellow fever, since it was found
that the blood-serum from a recovered case of dengue contained no immune body to
yellow fever.78 However, cross-protection may be dependent on the specific virus
causing primary infection, the interval between primary and secondary infection,
and on quantitative and qualitative aspects of the heterologous immune response,
including the cellular immune response.
III.8.4. Maintenance cycle absent?
The jungle, intermediate and urban cycles of yellow fever need different vectors. A
missing link can cause the cycles to break and an epidemic to die out. The environment
is divisible into “niches” which are occupied by one or more species of organism. A
species is so closely adapted to its niche in the local environment that it is unlikely to
be replaced by an alien emigrant race.65 In the East African emergence zone the
strains of Aedes bromeliae were found to be virtually non-man biting. Thus the link
between the forest maintenance cycle and establishment of an urban Aedes aegypti-
man cycle was broken79. On the other hand, as early as 1929, Dinger and his team
reported the transmission of yellow fever in Java with Aedes (Stegomyia) albopictus,12
which has the potential of bridging the gap between jungle and urban yellow fever
cycles.80 A. albopictus is found from Madagascar eastward through Asia to Japan,
Korea, and northern China. 81 Furthermore, there is an abundant monkey population,
Macacus rhesus, which is extremely susceptible to the virus, in the plain of the Indus
and Ganges. M. sinicus is also susceptible, on the Deccan (south of the Godavari
River).12 Thus, there is no definitive reason for failure of maintenance cycles in Asia.
WHO/EPI/GEN/98.11
3 9
III.8.5. Variation in vector competence and behaviour?
Asian strains of A. aegypti may be less efficient vectors of yellow fever virus than
African or American populations.4 Hindle’s experiments in 1929 showed that the
one Indian strain of A. aegypti was a less effective vector than the African strains of
mosquitos for the one strain of the virus in question.82 But the studies by Aitken and
Tabachnick showed Asian populations of A.aegypti to be better vectors than West
African populations.83 It was also shown by Miller et al, that in the presence of high
population density an incompetent mosquito vector can initiate and maintain virus
transmission resulting in an epidemic.84 Vector incompetence thus becomes less tenable
as an explanation for the absence of yellow fever in Asia.
In summary, it is not known why yellow fever never spread to Asia, but there is no
evidence to show that this could not occur. All South- East Asia countries should
ensure that persons arriving from the Latin American and African countries at risk
for yellow fever have a valid yellow fever vaccination certificates.
4 0
Yellow fever
IV. Cost effectiveness of
yellow fever vaccination
Vaccinations provided through the EPI are believed to be one of the most cost-
effective child survival interventions at a cost between $10 and $15 per child.85 The
Joint WHO/UNICEF Technical Group on Immunization in Africa recommended
in 1988 incorporation of yellow fever vaccine in routine child immunization
programmes of countries at risk for yellow fever, and the World Bank’s 1993 World
Development Report also strongly endorsed adding yellow fever vaccine to the EPI
of the at-risk countries.15
YF vaccine previously suffered from poor thermo-stability. Improved stabilisers have
provided a product whose shelf life is up to two years at a temperature of -20°C or
+4°C. In 1995, of the eleven vaccine manufacturers approved by WHO, seven were
producing vaccines that met stability standards.86 An update on this situation will be
discussed at the workshop.
There is only one published cost-effectiveness analysis of preventive yellow fever
vaccination versus emergency mass vaccination campaigns, conducted by Monath
and Nasidi (1993). This analysis was done for Nigeria under conservative assumptions
of vaccine coverage and efficacy. The models used are explained in detail in the
publication “Should yellow fever vaccine be included in the Expanded Programme
of Immunization in Africa? A cost-effectiveness analysis for Nigeria”.87 Using
assumptions based on data from other African countries, the cost of adding yellow
fever vaccine to the existing EPI was estimated as US$ 0,65 per fully immunized
child, whereas the cost of emergency vaccination was estimated as US$ 7,84/person.
For an epidemic of moderate size the cost-effectiveness of emergency mass
immunization for control of hypothetical yellow fever epidemics was two-fold higher
than that of the EPI. However, the efficiency of the EPI was seven-fold greater in
terms of cases and deaths prevented.87
In the Gambia, yellow fever vaccine was added to the EPI in 1979 following a
successful mass campaign in which 97% of the population over six months of age
received a dose. There have been no subsequent reports of yellow fever from the
Gambia. Adding yellow fever vaccine did not significantly increase the per dose cost
of immunization delivered in the EPI. According to the estimated cost of EPI with
and without yellow fever vaccination, the average total cost per yellow fever vaccine
dose had a difference of only US$ 0,01 (adding the cost of YF vaccine to the total
cost of the EPI and averaging this over all vaccine doses given).88 However, the
average cost per CHILD obviously increased with the addition of another vaccine.
WHO/EPI/GEN/98.11
4 1
Concerns have been raised over the affordability of national immunization programmes
and the need for continued donor support, identification of other financing
mechanisms, or reconsideration of policies aimed toward accelerating and maintaining
immunization coverage has been highlighted.85 Integration of YF vaccine into the
EPI needs to be done in the context of strengthening the immunization programmes
in general in many of the affected countries, where current infrastructure for the
routine programme is weak.
The cost of preventive immunization campaigns (as opposed to reactive campaigns
in the face of an outbreak) is unknown. The increasing promotion of measles vaccine
campaigns raises the question as to the integration of YF vaccine in such campaigns.
It is of note that most campaigns conducted in the first decades of YF immunization
programmes used either scarification or jet injector equipment. Such methods are no
longer recommended; in particular WHO does not recommend the use of jet injectors
because it is difficult to guarantee that they are free from risk of transmission of
bloodborne infection. Mass campaigns using autodestruct needles and syringes are
likely to be more costly, and perhaps require more human resources. If YF vaccine is
included with measles campaigns, questions would also arise about repeat vaccination.
For measles control/elmination, follow-up campaigns would be needed every five
years or so (or even more frequently, depending on coverage in the routine
programme). YF vaccination need not be repeated, although data on any potential
adverse events are lacking.
Immunization campaigns in the lowest income countries tend to involve a substantial
importation of cold chain equipment, supplies, vehicles, and technical assistance from
developed countries in order to attain coverage goals within a short period of time.
These country programmes are vulnerable to breakdowns of equipment and vehicles
and limited access to financial resources needed to cover recurrent costs of operating
at this level of activity. Rapid increases in coverage rates are sometimes followed by
periods of low immunization activity in some countries, demonstrating the need for
sustainable immunization strategies in the future.85 Mass immunizations require a
well-co-ordinated and planned effort on the part of national authorities, intensive
social promotion, and strong management.89
The cost of adding YF to immunization programmes (routine vaccination of infants
and/or preventive campaigns in high-risk areas) relates mainly to the cost of the
vaccine (12 to 25 cents per dose) and to safe injection and disposal equipment (around
13 cents per injection). In 1995, it was estimated that to provide YF vaccine to the
(then) 31 countries at risk, with a combined population of around 18 million infants,
would cost some four to five million US dollars per year (S.Robertson, unpublished
trip report 1995). Because YF vaccine can be given simultaneously with the EPI
vaccines, its delivery should not involve an extra visit to a health facility, which
greatly reduces total costs. In practice, however, health workers in many countries
are reluctant to administer YF vaccine as a separate injection to children who are
already receiving two injections (eg those who come late for DTP and hence are due
for DTP and measles on the same day). Mothers are then requested to return on
another occasion for YF vaccine, increasing the programme costs as well as reducing
coverage because many mothers fail to return.
4 2
Yellow fever
As discussed in section 6 and Table 9, current coverage with YF vaccine in most of
the countries that use it is inadequate to prevent an epidemic. The 1995 outbreak in
Senegal occurred when reported coverage was 46% (quoted in 47); and in the same
country in 1967, an epidemic occurred despite prior immunity levels of 57% in
children under 10 years.102 The prevalence of immunity in humans required to prevent
an epidemic has been estimated between 60 and 90% depending on vector biting
rate and vector competence.87
A strategy which allows administration of YF vaccine using the same syringe as
measles vaccine would not only reduce costs (the cost of safe injection equipment
more or less doubles the cost of each dose of measles or YF vaccine); it would also be
likely to increase coverage. A study in Mali used a combined vaccine in which 17D
YF vaccine and Schwarz measles vaccine had been lyophilized together91, and
seroconversion to both components was high. A combined preparation of measles
and YF vaccine would facilitate the achievement of high coverage, but the fact that
the market would be restricted to countries at risk for YF may make its production
unattractive to manufacturers. Its distribution could also be complicated in countries
where only part of the country is at risk for YF, so that health facilities in some areas
would get single-antigen measles vaccine while others would need the combined
preparation.
Studies in Cote d’Ivoire and Cameroon showed that YF vaccine mixed with measles
vaccine immediately before injection produced similar seroconversion rates as when
the vaccines were given separately92,93. However, in these studies the investigators
administered the reconstituted mixture of vaccine within one hour, not after several
hours, as would occur in a routine immunization clinic. There is insufficient data
available on the stability of YF vaccine mixed with measles vaccine and maintained
for several hours94. YF vaccine once reconstituted retains potency for up to three
hours if kept in ice, but loses potency much more quickly if it is kept at room
temperature. Thus it would be essential to keep vaccine on ice. Strict precautions
would also be needed to prevent bacterial contamination. Also we lack data on possible
Adverse reactions caused by mixing vaccines from different manufacturers. The
EPI/WHO policy does not recommend mixing different vaccines in one syringe
before injection.134
Cost-effectiveness analyses should also consider factors affecting efficacy and safety
of the vaccine. For theoretical reasons, YF vaccine is not recommended for pregnant
women, though benefits outweigh the risks in an outbreak setting. A study from
Nigeria showed reduced seroconversion to YF vaccine in pregnant women.87 Yellow
fever vaccine is not recommended for symptomatic HIV infected persons or other
immuno-suppressed Individuals.69 A recent study showed poor response to YF
vaccine in HIV-infected children aged 7-14 months in Cote d’Ivoire. Only 17% of
18 HIV-infected children sero-responded adequately compared with 74% of 54 HIV-
uninfected children133. The implications of HIV infection for mass YF vaccine use in
areas of high HIV-prevalenceneeds to be assessed.
WHO/EPI/GEN/98.11
4 3
V. Surveillance
V.1. Definition of surveillance
Surveillance and rapid response to identified disease threats are at the core of
preventive medicine. A well-designed and well-implemented infectious disease
surveillance programme can provide a means to detect unusual clusters of disease,
document the geographic and demographic spread of an outbreak, estimate the
magnitude of the problem, describe the natural history of the disease, identify factors
responsible for emergence, facilitate laboratory and epidemiological research, and
assess the success of specific intervention efforts. The effectiveness of surveillance
depends on the speed of reporting and analysing the results.10
Monitoring of factors such as population growth and migration, vector abundance
(e.g. the effect of the spread of potential yellow fever vector Aedes albopictus in
Americas, including USA), development projects that disturb the environment (e.g.
forest cutting in Brazil), and natural environmental factors (temperature and rainfall:
e.g. global warming and the effects of “El Nino”) is an essential component of
surveillance. These factors can influence both the spread of yellow fever and the
effectiveness of efforts to control them.
Official reports do not give an accurate picture of incidence or distribution of yellow
fever. Evidence for activity obtained by serological surveys is also largely incomplete
or out of date.4 Cases reported to WHO show only the presence of the virus, and
greatly underestimate the real number of cases.49 In seven epidemiological studies,
undertaken during yellow fever outbreaks over the last 25 years in Africa, morbidity
and mortality were constantly under-reported by 10 to 500 times.61 The majority of
the mild infections are undetected, which explains the high case fatality rates calculated
from these reports.49
Only limited attempts have been made to define the incidence of endemic yellow
fever infection. In Nigeria (1970-71), a laboratory diagnosis of yellow fever was
made in two (1%) of 205 patients hospitalized with jaundice in areas without epidemic
activity.95 Using data from serological surveys in Nigeria and an estimated 7:1 infection
: illness ratio, the annual incidence of overt infection was estimated to be between
1.1 and 2.4 per 1000 population, and yellow fever death between 0.2 and 0.5 per
1000.87 While indicating that endemic yellow fever may be a ‘silent’ cause of
significant morbidity, the incidence levels are 25 to 50-fold lower than those occurring
during epidemics, and are thus below the threshold of detection by existing passive
surveillance systems. The continued activity of YF in Kenya after the 1993 epidemic
detected by active surveillance further supports this view. It is likely that endemic
yellow fever activity is geographically focal and that it varies considerably from
4 4
Yellow fever
year to year, but that it causes thousands of unrecorded deaths annually in West
Africa. This provides a strong rationale for preventive immunization, but lacks
supporting data from surveillance or serological surveys. {Monath, T.P. Written
communication, 1998}
Surveillance can take many forms, each having advantages and disadvantages (Table
6). Although most infections with YF do not cause jaundice, this sign is the easiest
on which to base case-reporting. Most countries have a system for reporting cases
of hepatitis, or jaundice, which can be adapted to provide information about the
occurrence of suspected yellow fever. An unusually high case-fatality rate among
cases of jaundice might indicate the possibility of an outbreak of yellow fever.
For effective YF surveillance, the following measures are essential: identification of
suspected patients; prompt investigation of each suspected case with collection of
appropriate clinical specimens; transport of specimens from the field to the laboratory
in cold boxes; reliable completion of laboratory tests; forwarding specimens, as
appropriate, to higher-level laboratories should additional tests be indicated, and
rapid feedback to the district and national levels so that disease control measures,
including mass immunization campaigns, can be instituted.15 Some of the practical
difficulties faced in implementing these measures are described in the review of the
Kenyan sentinel surveillance system, below.
Active surveillance takes the form of regular contacts by surveillance officers with
health-facility staff, and/or periodic house-to-house surveys, which can provide more
complete information on YF incidence and risk factors for YF. Active surveillance
may be particularly indicated in the ecological zone of emergence (moist savannah).
Annual serologic surveys of young children can also provide important information
on the circulation of YF virus (since most infections are subclinical). Surveys
conducted during the dry season can detect any increase in seropositivity since the
previous rainy season, to help predict a danger of further increase in the following
year. Surveys of YF antibodies in monkeys can also provide evidence of YF
circulation.10 (Appendices III, IV and V). Because of an association between increased
YF isolates from mosquito vectors and outbreaks of YF in West Africa, vectors have
been monitored by Institut Pasteur in Senegal and Ivory Coast.
To improve surveillance, WHO AFRO has developed a training workshop and field
guide for district level staff on EPI target disease surveillance, which focuses on
poliomyelitis, measles, neonatal tetanus, and yellow fever. A series of laboratory
workshops on YF diagnosis have been held in the African region in the 1990s, for
Anglophone and Francophone countries (Dr S. Robertson, unpublished trip report
Sept 1995). Actions to strengthen the laboratory network for YF have followed on
from the success of the WHO polio laboratory network programme. WHO is
currently revising its reference materials for surveillance, and has developed suggested
case definitions, types of surveillance, minimum data elements and data analysis, and
case investigation forms (Appendix).
WHO/EPI/GEN/98.11
4 5
Since one of the main aims of YF surveillance is the early detection of outbreaks,
appropriate training and provision of resources for outbreak preparedness and
response is an essential complement to surveillance. Outbreak response consists of
mass immunization locally, with possible extension to surrounding areas. Data on
age-specific attack rates help in deciding the age group to include in mass vaccination.
WHO maintains an emergency stock of vaccine which can be made available rapidly
when there is an outbreak. A vector control specialist should also be consulted to
determine if specific mosquito control measures are indicated.
In addition to disease surveillance, monitoring immunization coverage also needs
improving. Data on YF incidence and vaccine coverage have been included in the
WHO EPI computerized monitoring system since 1991. Comparison of YF with
measles vaccine coverage can show missed immunization opportunities that may be
caused by vaccine supply problems, and/or concern about administering multiple
injections on one day.
In South America, routine yellow fever surveillance has been augmented by
histopathological review of liver specimens collected post mortem from patients who
died following an acute febrile illness.15 The “viscerotome” was invented by Rickard
in 193035 to permit the rapid and convenient removal of liver specimens by laymen.96
From 1500 liver specimens secured in 1931 and 13 733 obtained in 1932, this
programme was expanded by 1940 to more than 30 000 viscerotomies per year in
Brazil alone. By 1949 more than 400 000 liver samples had been collected and
examined. This viscerotomy service was soon extended to Bolivia and Colombia.
Pathological examinations confirmed the sero-protection tests, but while positive
sero-protection tests revealed past or potential foci,35 the advantage of liver
examination is that it reveals active foci of YF. Surveillance of YF continues to be
based on viscerotomy in South America. But even with post mortem surveillance, it
is estimated that only a small proportion of yellow fever cases are detected.10 The
countries of Americas notified 2238 cases to the Pan American Health Organization
(PAHO) between 1965 and 1984. This figure which is based for the most part on
viscerotomy results, gives only an incomplete idea of the real incidence of the disease.97
WHO has not emphasised histopathological surveillance in Africa because patients
are more likely to die at home, and families are reluctant to provide consent for an
autopsy.12
In the African Region, until recently, the approach to YF diagnosis was by serology
(CFT, HI and NT (in mice or cell cultures). In addition, work was done on virus
isolation, again in mice and cell cultures. These tests were tedious and fraught with
problems of cross reactions with numerous flaviviruses in Africa. Currently,
technicians are being taught to use the IgM test, which is faster, quite sensitive and
specific, even in the presence of other flaviviruses. It was used test in 1995 during
the Liberian outbreak. Although several laboratories have the capacity to conduct
the IgM ELISA test, they have not been fully utilized. As the Kenya experience has
shown, laboratory-driven surveillance is limited in scope, and not as cost-effective
as a well organised disease/epidemiological surveillance including laboratory
diagnosis. The potential of laboratory diagnosis is only realized when it is integrated
with a reliable and dependable clinical-epidemiological surveillance system (Tomori
O, personal communication 1998).
4 6
Yellow fever
Table 7: Methods of surveillance for yellow fever
Method
Advantages
Disadvantages
Examples
Case reporting
- indicates the presence of
- under-reported; passive
- 34 countries
(routine, passive
virus and yellow fever
systems underestimate the
presently at risk: see
surveillance)
activity in the area
true incidence of disease
annex II.
- if high reporting
10- to 250-fold and the
completeness can be
incidence of infection
achieved, potentially most
by 20- to 5000-fold
sensitive method of clinical
- delayed an average of two
surveillance
months in Africa
- difficult clinical diagnosis
Monitoring
- gives an idea of immunization - good and reliable reporting
WHO provides a yearly
vaccination
programme performance
system essential
EPI global summary.
coverage
- low coverage alerts to the
- only 31% of countries
risk of a potential epidemic
reported YF immunization
- monitor missed opportunities
coverage in 1995
Sentinel
- can get better quality data
- costly
Kenya, 1993 -
surveillance
- if YF is focal in a country,
- laboratories essential
(Kerio Valley)
can concentrate resources
- is not easy to integrate to
on affected areas
existing governmental Public
Health activities
- location of YF activity may
be unpredictable
Hospital record
- gives an idea of hospital
- not all patients are
Nigeria 1994
review
incidence and geographical
hospitalised
distribution of yellow fever
- selection bias
House-to-house
- gives an estimation of
- “verbal autopsy”: difficulties Nigeria (Obibi
survey
incidence, age and sex
of distinguishing YF from
village), 1994
proportions
disease with similar
symptoms
- information bias
Serological
- reveal past or potential foci
- cross-reactions with other
Africa 1931-1935
survey
- gives an idea about the
flaviviruses confuse the
(helped to delineate
population’s immunity status
results
the boundaries of
and yellow fever circulation
endemic yellow
in an area.
fever)
- use of IgM tsst may reveal
recent YF activity
Viscerotomy
- pathological examinations
- not very sensitive
- South America
confirm the sero-protection
- mostly fatal cases are
from 1930 to today
tests
reported:
- the countries of
- detects unconfirmed and
- not successfully applied in
Americas notified
atypical yellow fever cases
Africa, where death
2238 cases
- reveal active foci
registrations and burial
between 1965 and
practices are poorly
1984.
controlled.
Entomological
- confirms virus transmission
- need high level of effort to
- Ivory Coast, 1982
survey
in mosquitos
collect adequate number of
- Senegal, 1993
- limited to research efforts in
adult mosquitos
routinely monitored in
enzootic foci
- may not help PH officials to
the Kedougou area,
- YF virus isolations from
improve control activities,
Senegal since 1976
vectors correlate with
especially in rural and jungle - Burkina Faso
outbreaks of YF in humans;
areas
1983-86
can warn of increased risk
Monkey survey
- determines the presence of
- not practical on a large scale - Colombia 1936,
yellow fever and provides a
- ecological grounds: monkeys
1942
quantitative index of the rate
need to be shot and killed
Sentinel monkeys
of virus transmission
have been used to
- epizootic can precede an
detect sylvatic
epidemic
yellow fever activity
in Uganda
- Trinidad 1988-1989
- Senegal 1993
WHO/EPI/GEN/98.11
4 7
V.2. Yellow fever sentinel surveillance in Kenya98
Although periodic outbreaks of yellow fever have been reported in East Africa since
1940, the first yellow fever outbreak reported in Kenya began in September 1992
and continued through March 1993. Sentinel surveillance was established after this
outbreak, initially involving 13 health facilities in the Rift Valley, and expanding to
its current number of 43. An epidemiological support unit was established at the
Kenya Medical Research Institute (KEMRI) laboratory to provide the link between
possible disease occurrences in the field and the diagnostic facilities at the distant
laboratory.99
Surveillance teams usually consist of locally recruited staff trained on the spot and
supervised by one or more epidemiologists. Their task is to collect the specimens
that are to be sent to laboratories to confirm a suspected diagnosis. Adequate transport
facilities and a means of rapid communication are essential for the investigation and
control of outbreaks. In July-August 1997 a review of the system was conducted;98
salient findings are summarised here.
V.2.1. Objective of sentinel surveillance in Kenya
To improve the ability to detect, diagnose, and prevent yellow fever outbreaks in
Kenya (and the East Africa Region) through training and education programmes
which increase awareness of the disease, improve detection by pathological
examination, promote rapid response and planning and stimulate the initiation of
isolation precautions for suspect cases, through
•
improving knowledge and increasing awareness of yellow fever within the
Kenyan medical community and public health facility practitioners.
•
implementing a series of sentinel stations equipped to collect samples to send
to appropriate laboratories for analysis by pathological examination and
immunohistochemistry techniques,
•
improving diagnostic capabilities at the KEMRI for viral haemorrhagic fevers
and other viral diseases.
V.2.2. Yellow fever case definition for surveillance:7
Patients presenting with at least two of the following five symptoms or signs should
be reported as suspected yellow fever and a case investigation should be carried out:
1. Fever:
>38oC or 100.4oF
2. Jaundice:
yellow eyes, elevated serum bilirubin, bilirubin in urine
3. Haemorrhages:
vomiting blood or coffee-grounds like matter, persistent
nose bleeds, bleeding gums, or melaena
4. Encephalopathy:
confusion, disorientation, extreme drowsiness,
convulsions
5. Renal problem:
decreased urine output, proteinuria, blood in urine
4 8
Yellow fever
V.2.3 Summary of Kenyan guidelines
According to the Ministry of Health of Kenya published booklet “Field Guide for
Yellow Fever Surveillance” the health facility chosen as a sentinel surveillance centre
collects a blood sample from patients who meet the surveillance case definition. The
serum is separated, stored frozen, and transported to KEMRI (according to the
guidelines, it is to be kept frozen even during transport) - either through courier
services or personal delivery. All positive cases are later followed and if the patient
is still alive, a convalescent sample is taken. “ A Yellow Fever Case Investigation
Form” is completed and sent together with the serum sample.7
Any case of suspected yellow fever should be reported to the District Medical Officer
(DMO) within 24 hours. The DMO’s office should assist the health facility to
transport the blood specimen to KEMRI, follow-up the case, search for other cases
in the affected area, and report the case to the Provincial Medical Officer (PMO).7
The Provincial Medical Office provides support and supervision to the district as
necessary, including transporting specimens and helping with investigations of
outbreaks. The Central Ministry of Health provides periodic analyses and feedback
regarding cases nation-wide, follows up on the outcome of cases, and provides
laboratory results to the PMO and DMO. The Virus Research Centre in KEMRI
processes specimens for virus isolation and detection of antibodies, and provides
results to central MoH and feedback to the health facility that submitted the specimen.7
All serum samples are to be re-tested and confirmed at the Division of Vector-Borne
Infectious Diseases, Fort Collins, USA.
V.3.4. Findings from the review
The blood is collected by the sentinel health facilities from some patients (an unknown
proportion of those meeting the clinical case definition), and the serum samples are
kept frozen in the laboratories. Because of difficulties in establishing a reverse cold
chain the surveillance team from KEMRI, equipped with dry ice or liquid nitrogen,
collects the samples (the delay between collection and transport can be months).
The surveillance system is mainly the responsibility of KEMRI, which follows up
the positive samples, searches for other cases in affected areas and reports to health
facilities and Central Ministry of Health. The surveillance team, which normally
consists of an epidemiologist and a driver, but sometimes also a laboratory
technologist, also refills the “yellow fever kits” (equipment for specimen collection)
during its irregular visits.
Between 1993-1995 only an average of five samples monthly were obtained and in
1996 only two samples were collected for the whole year. The cost was about
Ksh. 20 000 (roughly about Ksh. 4000 , about US$ 80, per sample collected), which
is very costly.
WHO/EPI/GEN/98.11
4 9
Table 8: Attributes of yellow fever surveillance system in Kenya
Attribute
Comment
Usefulness
Capacity to contribute to the prevention and control of YF
-
Simplicity
Refers to both structure and ease of operation; considers
+
flow of information and lines of response
Flexibility
Ability to adapt to changing information needs or operating
+
conditions with little additional cost, time, staff
Acceptability
Reflects willingness of individuals and institutions to participate
+
Positive value predictive
Proportion of persons identified as cases who actually do
-
have the condition under surveillance
Representativeness
Ability to accurately describe the occurrence of YF over time
-
and its distribution in the population, and thus of the ability to
generalise findings from surveillance data
Timeliness
Reflects the speed or delay between steps in a surveillance
-
system
(+= Good; - = Needs improvement)
Results of interviews with health professionals in the sentinel sites in August 1997
showed that:
1) The yellow fever surveillance is perceived to be important because of the high
case-fatality of the disease. It is potentially useful but because of the variable
participation and delays in the system the surveillance cannot lead to quick
implementation of prevention and control measures.
2) The system is otherwise simple and flexible, but needs cold-chain
transportation for the samples and depends on laboratory confirmation. The
fact that the system is not integrated with other MoH services makes it harder
to keep staff motivated. However, for some techniques such as antigen
detection, ELISA, PCR and serologic samples, cold chain requirements are
much reduced.
3) The staff are interested in the subject, find yellow fever surveillance
important, but need more training in the recognition of suspected cases of YF
and in the use of the field guide.
4) The Predictive Value Positive of the case definition is low, because of its
broadness and the similarity of symptoms with many other common infectious
diseases in the area.
5) The representativeness is poor, because of the few samples collected, but also
because only a minority of health facilities report regularly.
6) Presently the system is too slow to provide information useful for control
activities.
5 0
Yellow fever
Based on the review of sentinel surveillance in Kenya, a series of recommendations
were made for improvement.
An important aspect of the clinical diagnosis is case-definition. For effective
surveillance the broadest possible case-definition should be used. The case definition
for yellow fever surveillance in Kenya is too complex and needs to be changed to a
broader form - for yellow fever it could consist of the three major symptoms: fever,
jaundice and haemorrhages.
The guidebook and the case investigation forms should be simple enough to be
understood easily by different groups of medical staff with different educational
backgrounds.
Continuous training is needed to improve on-the-job performance. Every District
Health Office has a “Human Resources Co-ordinator” (HRC) whose responsibility
is to co-ordinate all in-service training and who could act as a facilitator for
workshops. The surveillance team should spend more time with the field staff with
supervision and training on the spot. If communication between the field and KEMRI
could be improved (through telephone, letters), a special staff meeting could be easily
organised with the surveillance team and the health facility staff. The health facilities
need prompt feedback of results to motivate the work.
Laboratory workers in health facilities were better trained than the other staff, but
they don’t have the mandate to prescribe laboratory investigations - a job clearly for
the clinicians. The workload of most Medical Officers in Kenyan hospitals mainly
consists of administration and surgery, and the work in medical wards and out-patient
clinics is done either by Clinical Officers or nurses. The clinicians (including nurses)
need more training on yellow fever and surveillance, because they are in a key position
to prescribe an investigation. Also Public Health Officers and Technicians should be
involved, because they have been trained to survey diseases in their areas.
The Public Health Department could be the missing link between the specialised
yellow fever surveillance programme and the government for co-operation and
integration, because the Public Health Officers/Technicians are trained for
surveillance and are used to follow-up not only cases but also outbreaks of other
infectious diseases. They have motorcycles, which can carry an additional person
(e.g. a laboratory technician if samples are needed). The PMO of Rift Valley Province
Dr. K. Chebet in an interview on the 24 th of July, 1997 requested KEMRI for
logistical support to his staff to participate in surveillance and regular - maybe
quarterly - meetings in order to strengthen the link between his office and KEMRI.
The surveillance should be seen as a MoH initiative for the sake of sustainability.
The present yellow fever surveillance system can be improved by reducing the number
of sentinel sites (it is better to concentrate on few, but well-chosen and well-supervised
sentinel sites, than many badly-supervised ones), training more staff locally (especially
the staff of public health department), and organising regular and reliable
communication between the sentinel sites and KEMRI. The yellow fever surveillance
should be integrated gradually with the MoH activities. Any other arboviral research
could be carried out in a few well-chosen sentinels for scientific interest and
“virological trends” only.
WHO/EPI/GEN/98.11
5 1
VI. Comments and
suggestions
Historically yellow fever has caused devastating epidemics in Africa, South, Central
and North America and Europe. Yellow fever has not spread to Asia for unknown
reasons. Recently there have been some reported cases of imported yellow fever
both in Europe and in the USA. It is of great importance to inform all the international
vaccination centres about the dangers of yellow fever.
Yellow fever resurged in the 1990s: 1989-1991 was an especially active period in
Africa, where 34 countries are at risk for yellow fever: 17 have a policy to include
yellow fever vaccine in the EPI, but 14 have poor immunization programme
performance and 14 out of 34 belong to the countries in greatest economical need.
It is very difficult to try to prioritise the 34 African at-risk countries in order of
highest to lowest risk, because the resurgence of yellow fever is unpredictable, but
certain criteria could be used: the reported epidemics during the past years categorised
under the different topotypes; reported cases over the past 15 years; immunization
coverage and performance including the reported measles immunization coverage.
1) Recent epidemics: According to Map 3, Nigeria, which has reported far more
cases than any other country during the past 15 years, seems to have been a
centre for epidemics, and needs a lot of support for controlling yellow fever,
as does Mali (which may have been the centre of epidemics in 1987) and
Liberia. Other countries in West Africa (including Burkina Faso, Ivory Coast,
Ghana, Togo, Benin, Nigeria, Niger and Cameroon) also need extensive
vaccination coverage. Some countries seem to need only partial yellow fever
coverage in the country (especially in East and Central Africa).
2) The reported cases during the past 15 years are from Angola, Benin,
Burkina Faso, Cameroon, Gabon, Ghana. Guinea, Kenya, Liberia, Mali,
Mauritania, Nigeria, Senegal, Sierra Leone and Togo. Out of them Angola,
Burkina Faso, Gabon, Ghana, Mauritania, Nigeria and Senegal already have a
policy to include yellow fever vaccination to the EPI. Of these countries the
immunization programme performance is poor (<50%) in Angola, Burkina
Faso, Cameroon, Kenya, Mali, Mauritania and Nigeria: Burkina Faso, Mali
and Mauritania are also included in the countries of the greatest economical
need. Only part of some countries is exposed to the risk of severe outbreaks,
e.g. in Angola, Mali and Kenya only some districts need to be included in
yellow fever vaccination programme. By now because of the extensive
epidemics from 1984-1994, much of the population of Nigeria has been
vaccinated through mass vaccination programmes or has gained natural
immunity - it is important now to emphasise the immunization of the new
susceptible population - infants.
5 2
Yellow fever
Reporting, in general, is poor and slow, and needs to be standardised. E.g. most
reports on yellow fever in Weekly Epidemiological Record lack the information about
sex-ratios and occupational data, and categories of ages can vary enormously making
it difficult to make cross-comparisons. The main use of (under-)reported numbers of
cases and deaths is only to indicate yellow fever activity in an area, and the Case
Fatality Rates, varying from 20 to 80% depending whether the samples are collected
from alive or dead persons (e.g. the high CFRs in South America because of the
practise of viscerotomy), are not very useful.
Map 4 shows yellow fever vaccine coverage in countries at risk from outbreaks.
Table 8 compares different African countries with vaccination performance,
immunization coverage, reported cases and epidemics, and economical status. In
Tables 11, 12, 13 and 14, the countries have been categorised into four recommended
priority groups from highest to lowest priority. This does NOT mean that the
“lowest” priority countries should be neglected, but that the emphasis of work,
especially for yellow fever, should initially be in the highest priority countries.
Concurrently, all these countries need readiness for mass vaccination campaigns.
The importance of surveillance should not be forgotten, especially taking the
experiences of the surveillance in Kenya into consideration. E.g. Gambia has had a
successful immunization programme with combined strategy (mass immunization
covered 95% of the population in 1978-1979 followed by preventive vaccination in
the EPI which has reached 87% of infants). The last Gambian epidemic was reported
in 1979, but a sensitive surveillance system is still needed to improve the detection,
diagnosis and prevention of yellow fever outbreaks and monitor yellow fever coverage
including details of vaccinations by age group given in response to outbreaks.
Immunizations given through the EPI are one of the most cost-effective child survival
interventions. Only one study has been done comparing the cost-effectiveness of
preventive yellow fever vaccination and emergency mass vaccination campaigns -
the efficiency of the EPI was found greater in terms of cases and deaths prevented.
The mass immunization campaigns are also effective, and reasonably cost-effective,
and they secure the additional (donated) funding needed. There are no studies about
the cost-effectiveness of combined preventive vaccination through EPI and
surveillance versus mass vaccination campaigns; such studies should be encouraged.
However, in countries that have recently experienced widespread epidemics, most
“preventive” campaigns are probably not indicated.
The cost of using YF vaccine could be greatly reduced if the vaccine can be
administered in the same syringe as measles vaccine. Laboratory tests to determine
the stability of YF and measles vaccines mixed after dilution and kept for up to four
hours (as would occur in most routine immunization sessions) would be easy to
perform and should be done using measles vaccines from the range of manufacturers,
as the effect of mixing could vary depending on vaccine source. Should laboratory
experiments be encouraging, a field study of seroresponse to such vaccine mixed
prior to vaccination could be done relatively quickly and inexpensively. Concurrently,
the potential interest of vaccine manufacturers in producing a combined vaccine
should be explored, but the practical implications of distributing this vaccine to only
certain countries, or regions within a country, should be assessed.
WHO/EPI/GEN/98.11
5 3
All Asian countries should make an effort to check that all persons arriving from
yellow fever endemic countries have a valid yellow fever vaccination certificate.
In South America yellow fever is a disease of forest workers, but there is a danger
that Aedes aegypti -mosquito, which has reinfested all over South and Central
America, could transmit yellow fever in an urban cycle. The urban epidemics were
prevented in the early 1900s just by depriving the mosquitos of their breeding places
- this could be also enough today to prevent urban yellow fever in South American
towns, combined with yellow fever vaccinations as part of “occupational” health
care for forest workers.
5 4
Yellow fever
Map 3: Reported yellow fever immunization coverage in countries
at risk for outbreaks, 1993-1995
Not considered at risk
61 to 100%
41 to 60%
21 to 40%
0 to 20%
No report received since 1993
GPV/EPI
The designation employed and the presentation of material on this map do not imply the expression of
any opinion whatsoever on the part of the World Health Organization concerning the legal status of any
country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or
boundaries. Dotted lines represent approximate border lines for which there may not be full agreement.
Data as of 22 July 1997.
WHO/EPI/GEN/98.11
5 5
Summary of
recommendations
The combined strategy of surveillance, outbreak response and prevention is still
needed to combat yellow fever.
The group “highest priority” needs yellow fever vaccine to be included in the EPI
urgently (either the whole country or only parts of it), but also efforts to improve the
immunization performance in general, followed by improved surveillance systems.
Practical problems will be caused by political unstability and weak infrastructure in
some of these countries.
The group “high priority” needs yellow fever vaccine to be included in the EPI
(most already have), but need support to improve the immunization performance in
general, together with an improved surveillance system.
Most of the countries in group “medium priority” have already included yellow
fever vaccine in the EPI and relatively low YF activity. However, many of these
countries are politically unstable and/or in great economic need, and their
immunization programmes and surveillance systems for all the EPI diseases need to
be strengthened. Reasons for differences between coverage of YF and measles vaccine
should be investigated locally. Some of these countries have had little YF activity for
decades, but surveillance needs to be continued and strengthened, as resurgence has
been documented after long intervals in other countries. The priority to give to
introducing YF vaccine into the countries in this group which have no YF vaccination
(Congo, Eq. Guinea, Ethiopia, Sierra Leone, Sudan, Uganda and Zaire) needs
discussion.
The “lowest priority” countries (excluding Gambia) have not reported yellow fever
cases. Surveillance and possible outbreak response is enough at this stage. Gambia
has already a good existing immunization programme, which naturally needs support
to continue.
Large countries like Mali, Mauritania, Niger, Chad and Sudan could make a decision
to immunize below, perhaps, 15° N based on ecological considerations, and similarly
Angola could prioritize the region above, perhaps, 12°. However, it needs to be
reiterated that such priority schemes are problematic where Ae. aegypti exists outside
the ‘zone of emergence’, creating a ‘receptive’ area within national boundaries.
Moreover, the problem of migration and movement of nonimmune persons into the
endemic region both in normal commerce and during political unrest is substantial.
These are precisely the issues now in South America, and there is a risk of recreating
them in Africa. Guidelines on targetting vaccination need to be developed.
5 6
Yellow fever
WHO/EPI/GEN/98.1
Table 9: Yellow fever outbreaks, immunization coverage & performance in African countries
at risk for yellow fever outbreaks.
Country
Total reported cases last time cases
Reported at least one
immunization programme
YF vaccine
YF immunization
measles
1982-1996
reported
outbreak 1982-1996
performance poor
included in the
coverage (year)
immunization
1
(<50% immunization coverage)*
EPI (even partially)
coverage (1995)
Angola
37
1988
+
+
+
34 (1994)
32
Benin
124
1996
+
81
Burkina Faso
280
1985
+
+
+
55 (1995)
55
Burundi
44
Cameroon
184
1994
+
+
51
Cape Verde Is.
66
CAR
+
+
52 (1995)
70
Chad
+
+
28 (1994)
Congo
1961
39
Eq. Guinea
1970
Eritrea
+
Ethiopia
1966
+
38
Gabon
44
1995
+
+
23 (1991)
Gambia
1979
+
68 (1994)
Ghana
523
1996
+
+
24 (1995)
54
Guinea
5
1987
+
Guinea Bissau
Ivory Coast
25
1982
+
+
43 (1995)
57
Kenya
64
1995
+
+
35
Liberia
360
1997
+
Mali
305
1987
+
+
3 (1994)
49
Mauritania
21
1987
+
+
+
32 (1990)
Niger
1939
+
+
27 (1995)
38
Nigeria
20 337
1994
+
+
+
1 (1993)
Rwanda
Sao Tome
2 (1994)
Senegal
79
1995
+
+
46 (1994)
80
Sierra Leone
33
1995
+
Somalia
Sudan
1942
77
Tanzania
75
Togo
7
1987
+
14 (1993)
65
57
Uganda
1971
Zaire
1972
+
8 (1992)
41
* Designated as in greatest need of improved programme performance and enhanced financial support (EPI Information System, 1997)
Table 10: Prioritising 34 African countries at risk for
yellow fever for support; highest priority
Highest priority
Country
Arguments
Comments
Recommendations
Nigeria
- huge epidemics during
- has already included
- yellow fever vaccination
recent years: highest
YF in the EPI, but
together with measles
number of reported
coverage only 1%
at the age > 6 months
cases in Africa
- the whole country
- centre of other
should be included
epidemics in topotype II
- improve surveillance
area
Cameroon
- neighbouring Nigeria
- 214 000 doses of YF
- as Nigeria
- two epidemics in 1990s
vaccine given in 1990
Kenya
- topotype III in Kerio
- mass vaccinations in
- yellow fever vaccination
Valley
1992-1993 covered
needs to be included
- has established
one million people
into the EPI in Kerio
sentinel surveillance for
Valley, also all
yellow fever
immigrants should be
- endemic area close to
vaccinated.
big highly populated
- existing surveillance
urban centres
system needs to be
improved
Liberia
- CIGN***
- mass vaccinations
- yellow fever vaccination
- most recent epidemic
covered one million
together with measles
reported in Africa (1997)
people in 1995
at the age > 6 months
- improve surveillance
Mali
- CIGN***
- mass vaccination
- EPI activities need
- centres the topotype I
campaign in 1969,
support
area
and three million were
- YF vaccination should
- reported one epidemic
vaccinated in 1987
be included in EPI in
in 1987; 70% < 15 years
high risk areas
old
- improve surveillance
**
No reports to WHO for the last three to five years.
***
Country in greatest need: EPI has identified 21 countries in greatest need of improved
vaccination programme performance and enhanced financial support.100
5 8
Yellow fever
Table 11: Prioritising 34 African countries at risk for
yellow fever for support; high priority
High priority countries
Country
Arguments
Comments
Recommendations
Angola
- reported an epidemic in
- has already included
- only part of the country
1988
YF in the EPI
needs YF vaccinations,
but the overall EPI
programme needs to
be improved
- support surveillance
Burkina
- CIGN
- has already included
- support EPI activities
Faso
- reported an epidemic in
YF in the EPI
- support surveillance
1985
Gabon
- reported an epidemic in
- has already included
- YF immunization
1995
YF in the EPI
coverage poor,
surveillance system
needed
Mauritania
- CIGN
- has already included
- only part of the country
- reported one epidemic
YF in the EPI
needs YF vaccinations
in 1987 - but connected
with Malian epidemic
Senegal
- reported an epidemic
- has already included
- support EPI and
in 1995
YF in the EPI
surveillance to sustain
- high risk: historically
- measles immunization
good control
many epidemics
coverage 80%, but YF
- monitor YF vaccination
immunization Coverage
coverage and missed
<50%
opportunities
Togo
- reported an epidemic in
- Difficult to assess
- Introduce yellow fever
1987
degree of risk
vaccination together
with measles at the
age > 6 months
WHO/EPI/GEN/98.11
5 9
Table 12: Prioritising 34 African countries at risk for
yellow fever for support; medium priority
Medium priority countries
Country
Arguments
Comments
Recommendations
Benin
- reported an epidemic in
- measles immunization
- training needed to add
1996
coverage good
YF vaccinations to EPI
- historically many
epidemics
CAR
- CIGN***
- has already included
- support EPI activities
- no reported epidemics
YF in the EPI
- support surveillance
for tens of years
Chad
- CIGN***
- has already included
- support EPI activities
- no reported epidemics
YF in the EPI
- support surveillance
for tens of years
Congo
- last epidemic in 1961
- support EPI activities
- support surveillance
Eq. Guinea
- CIGN***
- support surveillance
- last cases in 1970
Ethiopia
- CIGN***
- support EPI activities
- last epidemic in 1966
- support surveillance
Ghana
- reported the second
- has already included
- support surveillance
highest number of
YF in the EPI
cases during the past
- YF coverage > 50%
15 years
Guinea
- last cases in 1987
- support EPI activities
- support surveillance
Ivory Coast
- last epidemic in 1982
- has already included
- support surveillance
YF in the EPI
and EPI activities
Niger
- CIGN***
- has already included
- support surveillance
- last epidemic in 1939
YF in the EPI
and EPI activities
Sierra
- CIGN***
- include yellow fever in
Leone
- reported an epidemic in
the EPI
1995
Sudan
- last epidemic in 1942
- Part of country at risk
- include yellow fever in
the EPI, but only
partially
Uganda
- last reported cases
- support surveillance
from 1971
Former
- CIGN***
- support surveillance
Zaire
- last epidemic in 1972
and EPI activities
**
No reports to WHO for the last three to five years.
***
Country in greatest need: EPI has identified 21 countries in greatest need of improved
vaccination programme performance and enhanced financial support.100
6 0
Yellow fever
Table 13: Prioritising 34 African countries at risk for yellow fever for support;
lowest priority
Lowest priority countries
Country
Arguments
Comments
Recommendations
Burundi
- CIGN***
- surveillance and
- no reported epidemics
outbreak response
Cape
- no reported epidemics
- surveillance and
Verde Is.
outbreak response
Eritrea
- CIGN***
- surveillance and
- no reported epidemics
outbreak response
Gambia
- good immunization
- has already included
- support surveillance
performance: no
YF in the EPI
reported epidemics
since 1979
Guinea
- no reported cases since
- surveillance and
Bissau
1951
outbreak response
Rwanda
- no reported epidemics
- surveillance and
outbreak response
Sao Tome
- no reported epidemics
- surveillance and
& Principe
outbreak response
Somalia
- CIGN***
- surveillance and
- no reported epidemics
outbreak response
Tanzania
- no reported epidemics
- surveillance and
(even historical)
outbreak response
WHO/EPI/GEN/98.11
6 1
Appendix I:
Examples of historical yellow fever
epidemics12 23 24 25 27 101 102-106
South and Central America
Year
Place
Cases Deaths
Comments
164- West Indies
Epidemics: 1649, 1652, 1656, 1664, 1671, 1686, 1690,
1691, 1694, 1695, 1703, 1705, 1715, 1723, 1664, 1729,
1664 St.Lucia
94%
1731, 1734, 1735, 1740, 1741, 1743, 1750, 1751, 1754,
died of
1756, 1761, 1762, 1765, 1767, 1769, 1500, 1770, 1779,
1500
1780, 1781, 1791, 1793, 1795, 1796, 1800, 1801, 1802,
1803, 1804, 1807, 1813, 1816, 1780, 3500, 1817, 1818,
1780 Jamaica
3500
1819, 1820, 1821, 1822, 1823, 1824, 1825, 1826, 1827,
1828, 1829, 1830, 1837, 1838, 1839, 1841, 1842, 1843,
1847, 1850, 1852, 1853, 1854, 1855, 1856, 1858, 1860,
1861, 1862, 1864, 1865, 1866, 1867, 1868, 1869, 1877,
1881, 1887, 1889, 1891, 1894, 1895, 1901, 1907, 1908
1699 Mexico
Vera Cruz attacked: thought to be the first epidemic here
1725 Mexico
At Vera Cruz: reported in Clavigero’s Histoire de la Mexique
1760 Dutch Guiana:
Said by Fermin in his “Traite des Maladies les plus frequentes
Surinam
a Surinam” to be the first epidemic of yellow fever there
1762 French Guiana
At Cayenne where the outbreak continued for three years,
another epidemic in 1791
1793 Venezuela
Caracas attacked in October
1793 British Guiana
Outbreak in Demerara
1795 West Indies
31 000
Among European troops
180- Central- and
Venezuela in 1802, 1869, British Guiana in 1800, 1820,
South America
1821, 1825, 1837, 1838, 1840, 1841, 1881, Colombia in
1830, 1861, 1883, 1886, 1887, 1888, 1889, Dutch Guiana in
1800, 1835, 1837, 1841, 1854, Peru in 1842, 1852, 1854, 1869,
Mexico in 1846, 1863, 1865, 1868, 1875, 1878, Rio De Janeiro
in 1850, 1894, French Guiana in 1802, 1850, 1855-1858, 1877,
Honduras in 1860, San Salvador in 1868, Nicaragua in 1868,
1804 San Juan
Mortality was “inordinate”
Puerto Rico
1898 Rio de Janeiro
Mortality 95,5% ???
1900 Colombia
Aedes aegypti was very scarce and the cases may have
been of the rural type.
190- Dutch Guiana
Both epidemics in 1902 and 1908 coincided with the advent of
a large number of non-immunes
19— Colombia
Outbreaks in 1907, 1910, 1912, 1915, 1920, 1929,
19— Venezuela
Outbreaks in 1908, 1912, 1914, 1917, 1918, 1928-1929.
6 2
Yellow fever
North America
Year
Place
Cases Deaths
Comments
1668 North America
Particularly destructive in the cities of New York and
Philadelphia. Intercommunication between settlements in North
America and corresponding settlements in the West Indies was
frequent and the old wooden ships would harbour mosquitos
169-
North America
1690 in Charleston, 1691 in Boston, 1693 in Philadelphia,
Charleston and Boston, 1694 in Boston, New York and
Philadelphia, 1699 in Charleston and Philadelphia.
169-
Philadelphia
Epidemics: 1699, 1741, 1762, 1780, 1794, 1795, 1796, 1797,
1798, 1799, 1800, 1803, 1804, 1805. The worst one was in
1793.
17—
North America
1702 in New York, 1703 in Charleston, 1728 in Charleston,
1732 in Charleston, 1734 in New York, Boston, Charleston,
Philadelphia and Albany, 1737 in Virginia, 1739 in Charleston,
1741 in Virginia, 1743 in New York and Virginia, 1745 in New
York and Charleston, 1747 in New Haven, 1748 in Charleston,
1751 in New York and Philadelphia, 1778 in Philadelphia, 1783
in Baltimore, 1791 in New York and Philadelphia, 1792 in
Charleston, 1793 in Philadelphia.
1793 Philadelphia
4000
After the arrival of refugees from San Domingo in August,
“bilinous fever” started in September and lasted for seven
weeks.
180-
North America
Norfolk in 1801, New York in 1801, 1819, 1821, 1822, 1870,
Massachusetts in 1801, Philadelphia in 1802, 1803, 1805, 1819,
1820, 1821, 1867, Boston in 1803, 1821, Charleston in 1807,
1817, 1819, 1821, 1824, 1839, 1843, 1854, 1856, New Orleans
in 1811, 1817, 1819, 1820, 1821, 1822, 1824, 1827, 1828,
1829,
1837, 1841, 1847, 1854, 1856, 1867, 1873, 1878, Florida in
1811, New Jersey in 1811, Baltimore in 1817, 1819, 1821,
Alabama in 1821, 1854, 1873, banks of Mississippi in 1821,
1843, 1855, 1873 Key West in 1823, 1829, 1841, 1867,
Natchez in 1825, 1829, 1837, 1847, Washington in 1825, Mobile
in 1825, 1827, 1829, 1837, 1839, 1843, 1847, 1854, 1867,
Memphis in 1828, 1873, 1879, Galveston in 1839, 1843, 1867,
1839 Galveston,
About
Epidemic described in detail by Dr Ashbel Smith
Texas
100
1846 The Plains of
“The Mormons, during their march from Nanvoo to Utah,
US
suffered from remittent and yellow fevers. Their track across
the desert was marked by the graves of those who perished.”
(Walford).
185-
Charleston,
Several major epidemics: the most notable in 1852, 1854, 1856
South Carolina
and 1858. 682 deaths were reported in 1854
1853 New Orleans
4858
1870 New York
The last recorded outbreak of yellow fever in New York
1876 Charleston
The final yellow fever epidemic in Charleston
1877 Port Royal,
The last yellow fever epidemic in South Carolina
South Carolina
1878 Mississippi
100 000 20 000
An economic loss of more than 100 million dollars
Valley
1905 New Orleans
5000
1000
WHO/EPI/GEN/98.11
6 3
Europe
Year
Place
Cases Deaths
Comments
1649 Gibraltar, Spain
Brought by ships en route to the West Indies from Africa, or
returning
1700 Cadiz
According to Hirsch the first record of yellow fever in Cadiz
1723 Lisbon, Portugal,
Report on local epidemic by Pedro Francisco da Costa
London ???
Alvarenga - yellow fever appeared first time in Europe? The
disease is said to have appeared in London, being transmitted
from Lisbon.
1724 Spain and
The disease was attributed to the “eating of fruit and drinking
Portugal
snow-water.”
172-
Spain
Gibraltar in 1727, Cadiz in 1730, 1736, 1764,Malaga and
Cartagena in 1741
1730 Europe
During September and October 22 000 deaths took place after
the arrival of the flotilla of Pintado from Carthagena, where
many of his men had died of ‘el vomito prieto’. Bascome writing
of this outbreak adds: “It was probably this pestilence which
during the seven years 1729-35 raged in Vienna, Pignerol
Fossano, Nizza, Rivoli, Asti, Larti, Acqui, Basle, Silesia,
Thrasburg (Lower Rhine), Trino, Fresneuse (Lower Seine),
Vimeux (Seine et Oise), Orleans (Loiret), Plouviers (Loiret),
Meaux, Villeneuve, St.George (Seine et Maine), Bohemia,
Denmark, Sweden and Russia.”
1740 Cartagena
Due to neglecting the precaution of anchoring some distance
from the shore resulted an epidemic in the British Navy
1741 Malaga, Spain
8431
During a military operation
180-
Bres, Brittany
Small outbreaks in 1802 and 1856
1802 Cadiz, Spain
Spread by 1804 to Cordoba, Grenada, Valencia, Catalonia and
Gibraltar was seriously attacked
1804 Livorno, Italy
2000
650
181-
Spain
Outbreaks recorded in Barcelona, Cadiz, Cartagena, Malaga in
1810, in Gibraltar in 1813,
1821 Barcelona, Spain
5000-
French and English observers started to recognise that yellow
20 000
fever might move further and northward within Europe
1823 Portugal
Outbreak at Lisbon
1826 Dublin
The most remote reported epidemic of yellow fever in Europe,
according to Fannim, 1848.
1828 Gibraltar
53 83
1183
“Atmospheric causation”
1852 Southampton
Small outbreak
1857 Oporto, Lisbon
“Awesome in both scale and mortality” -
Portugal
the last major European epidemic
1861 Saint-Nazaire,
“Anne Marie”, a small wooden sailing ship from Havanna had
France
40
26
already at sea a case-fatality of 22% because of yellow fever.
1865 Swansea,
27
17
During a spell of extraordinary hot weather a small number of
Wales
infected mosquitos from a cargo of copper ore from Cuba
established an epidemic in town
1870 Barcelona
1878 Madrid
6 4
Yellow fever
Africa
Year
Place
Cases Deaths
Comments
1751 West Africa:
Lind: “In several towns, among the negro population the
Guinea Coast,
mortality was so great that there were no sufficient left to bury
Senegal
the dead...”
181-
West Africa
Senegal in 1814, 1816, 1828, 1830, 1837, 1840-41, 1844, 1852,
1858, 1863, 1866, 1872, Sierra Leone in 1816, 1823, 1859,
1862, 1865, 1866, the Congo Coast in 1816, 1862, 1865,
Fernando Po in 1839, 1862, the Gold Coast in 1852, 1862,
Senegambia in 1858, 1859, 1866, Gambia in 1860, Angola in
1860, the Benin Coast in 1862, Canary Islands in 1810, 1862,
1888, Lagos in 1864,
1900 West Africa
Started in Senegal and spread along the railway from Kayes to
Dioubeba in the French Sudan, 1901. Dakar was almost
decimated.
1923- West Africa
Yellow fever became active in the Gold Coast, Nigeria,
27
Senegal, French Sudan, the Ivory Coast, Dahomey, Togoland,
the Upper Volta, and the Belgian Congo.
WHO/EPI/GEN/98.11
6 5
Appendix II:
Yellow fever cases reported in Africa,
1900-1996
6 6
Yellow fever
WHO/EPI/GEN/98.1
1900
-00
-01
-02
-03
-04
-05
-06
-07
-08
-09
-10
-11
-12
-13
-14
-15
-16
-17
-18
-19
-20
-21
-22
-23
-24
-25
-26
-27
-28
1
Angola
X
Belgian Congo
9
45
Dahomey
11
12
2
16
6
14
2
4
6
French Sudan
20
36
5
f e w
6
5
French Togo
8
Gambia
1
5
Gold Coast
8
7
27
107
4
Guinea
1
Ivory Coast
15
X
X
7
3
3
3
Liberia
5
5
Niger
1
Nigeria
1
21
11
3
Port. Guinea
X
Senegal
5
1
2
15
27
116
Upper Volta
2
X = cases reported
67
68
-29
-30
-31
-32
-33
-34
-35
-36
-37
-38
-39
-40
-41
-42
-43
-44
-45
-46
-47
-48
-49
-50
-51
-52
-53
-54
-55
-56
-57
Anglo-Eg. Sudan
1
15,
1
641
Belgian Congo
*
*
14
2
3
7
3
2
3
2
1
5
6
1
3
4
3
British Cameroon
8
British Togo
1
Dahomey
1
2
4
7
1
1
2
1
French Cameroon
1
1
2
French Eq. Africa
2
7
1
3
5
2
2
1
1
French Guinea
9
4
1
1
3
7
French Sudan
8
17
1
7
3
9
1
8
1
3
1
French Togo
3
4
3
1
1
1
Gabon
3
8
Gambia
3
3
Gold Coast
2
20
4
6
5
5
6
77
15
2
4
1
2
1
5
2
27
13
25
6
1
7
(Ghana)
Guinea
1
Ivory Coast
16
3
28
101
1
10
24
11
2
4
8
1
2
2
1
1
Kenya
1
1
Liberia
22
Mauritania
1
Niger
2
5
8
1
1
1
3
Nigeria
1
4
1
1
7
1
5
26
11
8
3
2
1
1
46
3
1**
13** 42**
18
1
2
Northern Rhodesia
3
Port. Guinea
46
7
8
2
2
Senegal
3
15
22
4
10
24
2
1
1
3
2
Sierra Leone
3
1
3
3
1
1
3
2
Uganda
1
1
Y
ellow fever
Upper Volta
10
1
*
Febrile jaundice, which was not confirmed serologically as yellow fever
**
A series of several outbreaks among indigenous population of Nigeria with at least 12 000 cases not reported officially
WHO/EPI/GEN/98.1
-58
-59
-60
-61
-62
-63
-64
-65
-66
-67
-68
-69
-70
-71
-72
-73
-74
-75
-76
-77
-78
-79
-80
-81
-82
-83
-84
-85
Angola
65
1
Cameroon
1
2
1
1
2
1
7
1
Congo
60
11
7
4
Eq.Guinea
4
Ethiopia
***
10
350
Gambia
30x
x
Ghana
2
3
5
12
3
5
5
1
2
2
110
213 494
8
4
6
372
Ivory Coast
25
M a l i
21
Nigeria
208 x
4 x
2 x
25
11
1
12
6
Senegal
243#
1
3
Sierra Leone
130
Swaziland
1
Togo
1
2
1
Uganda
1
1
106
14
Upper Volta
87
356 x 17
7
(Burkina Faso)
x = Estimated cases in Gambia 8,400 and in Nigeria about 100 000 in 1969 and 1000 in 1973 and 3000 in Burkina Faso in 1983.
# = Estimated cases 20 000.
***
100 000 cases in Ethiopia.
69
70
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Angola
37
Benin
124
Cameroon
173
10
Gabon
28
16
Ghana
39
79
27
Guinea
5
Kenya
27
27
7
3
Liberia
360
Mali
305 x
Mauritania
21
Nigeria
1289 x
2676 x
4920 x
3270
4075
2561
149
152
1227
Senegal
79
Sierra Leone
33
Togo
6
x =
Estimated morbidity 1525 in Mali in 1987, and in Nigeria 9800 in 1986, 120 000 in 1987 and > 1000 in 1988.
Y
ellow fever
Appendix III:
African vectors49, 51, 79, 83, 107-115
Mosquito
Breeding / living
Biting habit
Comments
Aedes
(Stegomyia)
aegypti
a) domestic
- breeds in artificial
- antropophilic
- its distribution correlates
form
containers
- bites inside & outside during
with human behaviour
(aegypti)
(containers for water
day time, especially late
- major & often the only
storage, old cans, tins,
afternoon
vector involved in man-
used tyres, etc.)
- all the year around
to-man transmission
- including dry season
b) wild form
- breeds in natural water
- zoophilic
- little role in yellow fever
(formosus)
collections (tree-holes,
- found during the rainy
transmission (life span
rock-holes, fruit-shells,
season & the beginning
is short, it has little
crab-holes, etc.)
of the dry season
contact with monkeys).
Aedes
- found in forested areas
- primatophilic
- main vector in the rain
(Stegomyia)
from the rain forest to
- after dusk in the canopy,
forest and as an
africanus
the dry savannahs
but may bite at any time
important vector in
during the day when a
forest galleries
convenient host is
introduced in its activity
area (intrusion effect)
Aedes
- known from savannah
- less primatophilic than
- virus often isolated from
(Stegomyia)
areas
A. africanus
A. opok in Central African
opok
Republic and in Ivory
Coast
- considered as an
important vector
Aedes
- known only from forest
- it is never very abundant
(Stegomyia)
galleries of eastern
but can act as an
neoafricanus
Senegal
effective local vector
(infectious rate is high)
Aedes
- in savannah areas,
- primatophilic
- incriminated in 1969 during
(Stegomyia)
extending to the Sahelian
- bites monkeys after dusk
the Jos Plateau epidemic
luteo-
zone, common in forested
in the canopy of forest/
in Nigeria
cephalus
areas
mangrove galleries
- main vector in savannah
- breeds in tree-holes
areas in West Africa
Aedes
- at least three species
(Stegomyia)
- isolated in Uganda in 1942
simpsoni
and many times during the
group
Ethiopian epidemic
1960-62.
- in West Africa human-
biting species of the
Simpsoni group were
recorded only in Nigeria
WHO/EPI/GEN/98.11
7 1
Mosquito
Breeding / living
Biting habit
Comments
a) A. simpsoni
- found in South Africa only
b) A. lilii
- non-primatophilic
- found in East Africa
- is not considered as a
yellow fever vector
c) A. brome-
- vegetal breeding places
- day-biter
- most probably the vector
liae
(leaf axils of banana
incriminated by Mahaffy
trees), tree-holes
or Haddow
under
the
name A. simpsoni
Aedes
- savannah dwellers
- take two blood meals
- species of this group
(Diceromyia)
during a single
first incriminated during
furcifer-
gonotrophic cycle
the Nuba Mountains
taylori group
epidemic in Sudan in 1940
- 58 yellow fever virus
strains were isolated from
male and female wild-
caught mosquitos during
an epizootic in Senegal in
1977
a) A. taylori
- in the canopy of forest
- more simiophilic than
- main vector from
galleries
anthropophilic
monkey to monkey in the
canopy of forest galleries
b) A. furcifer
- in open savannahs and
- anthropophilic
- monkey-to-monkey
in villages
- bites indoors and outdoors
transmission, but also
human infections
- main vector in intermediate
epidemics: Gambia in 1978
Burkina Faso in 1969 and
1983, Mali in 1987
Other
- were found infected in
mosquitos
nature, but none is
efficient vector
a) A. metallicus
- have no strict trophic
- never play an important
b) A. vittatus
preferences, or do not bite
role in dissemination or
c) A. dentatus
primates
amplification of the virus
d) A. stokesi
e) Eretmapo-
dites
f) Mansonia
- never play an important
g) Culex
role in dissemination or
amplification of the virus,
because the virus has too
long incubation period
Ticks
Amblyomma
- in nature of Central
- yellow fever virus was
veriegatum
African Republic
twice isolated from ticks
(once from males and
once from eggs)
- more important role in
maintenance of the virus
since the virus can be
vertically transmitted
7 2
Yellow fever
Appendix IV:
African vertebrate
hosts35, 115, 73, 107, 116-124
Name
Environment
Comments
Simioidea
- all African monkeys susceptible
(monkeys)
to yellow fever virus, developing a
viraemia followed by neutralising
antibodies, but have only a mild,
unnoticeable disease
a) Colobus
- East and central Africa
- viraemia is long (five to nine days)
abyssinicus
- antibodies appear after the eighth
day
- one strain of virus isolated during
the Ethiopia epidemic in 1960-62
b ) Colobus
- West Africa
- found serologically positive in
polykomos
Ghana in 1965, Sierra Leone in
c) Colobus
1937 and Senegal in 1936
badius
d) Cercopithecus
- East Africa: C. mitis: transmit the
- best studied genus
virus in the forest, C. aethiops and
C. nictitans feed in plantations and
area source of infection of
anthropophilic mosquitos
- West Africa: C. diana and C. mona
dwell in forest and C. aethiops in
savannah areas
e) Cercocebus
- forest monkeys which often come
- in Uganda, C.albigena had high
(mangabeys)
to the ground level to feed
antibody prevalence rates, but its
viraemia is short-lived and of low
level
f) Erythrocebus
- common in savannahs
- travels long distances and often
patas (red
feed in plantations
monkey or
- often serologically positive
patas)
g) Papio papio
- savannah dwellers
- seropositivity rate in East and West
h) Papio anubis
- live in large groups attractive for
Africa high
(baboons)
primatophilic mosquitos
i) Pan troglodytes - East, West and central Africa
- not very common, so it has not an
(chimpanzee)
important role
WHO/EPI/GEN/98.11
7 3
Name
Environment
Comments
Lemurioidea
(pottos and
galagoes)
a) pottos
- nocturnal, tree-dwellers of the rain
- too rare species to play an
forest
important epidemiological role
b ) galagoes
- live in forests and savannahs,
- in nature, serological surveys did
(Galago
active at night
not show any evidence of their
senegalensis,
participation in the circulation of
Galago
virus in West Africa, but in East
crassicaudatus)
Africa galagoes were found
positive in Kenya and in Uganda
Other
- Sudanese (Atelerix pruneri) and
- several found susceptible, but their
vertebrates
European hedgehogs
role can be limited because of their
(A. erinaceus)
lack of contact with known vectors
- bat (Epomophorus sp.)
- cross-reactions possible; many
- laboratory mice
positive responses are probably
- Guinea pigs (after intracerebral
caused by antibodies induced by
inoculation)
other flaviviruses
- Steatomys opimus
- carnivora usually resistant, except
Genetta tigrina and Nandinia
binotata
- Artiodactyla (camels, sheep, goats,
cattle, pigs etc.) found serologically
positive, but often in countries
where yellow fever virus has
never occurred
7 4
Yellow fever
Appendix V:
South American vectors and vertebrate
hosts17,35, 49, 125, 126, 73, 107, 127-131
Mosquito
Breeding / living
Biting habit
Comments
Aedes
- peridomestic
- imported from West Africa
(Stegomyia)
aegypti
Haemagogus
- rests in tree-holes &
- simiophilic, but bites
bamboo stems
occasionally humans
a) H.
- abundant in the humid
- the principal vector in South
(Haemagogus)
tropical forest canopy
America
janthinomys
of Colombia and the
- if the forest is cut, this vector
Atlantic coast of
will survive at ground level of
Panama
plantations
b) Hg.
- shares a similar
- principal vector in Guatemala
(Haemagogus)
ecology with H.
equinus
janthinomys
c) Hg.
- principal veector in Guatemala
(Haemagogus)
in 1956
mesodentatus
d) Hg.
(Haemagogus)
lucifer
e) Hg.
(Haemagogus)
iridicolor
f) Hg.
- major sylvatic vector found
(Haemagogus)
infected in nature:
capricornii
taxonomically confused with
Hg. janthinomys???
g) Hg.
- at ground level bites
(Haemagogus)
man both indoors and
albomaculatus
outdoors
h) Hg.
- lives in forests
- anthropophilic
(Conopostegus)
leucocelaenus
Sabethes
- rests in tree-holes &
- simiophilic, but bites
- highly tolerant of arid
(Sabethoides)
bamboo stems
occasionally humans
conditions: it may play an
chloropterus
- active all year long
important role in the
maintenance of the virus in
areas with long dry seasons
Aedes
- imported from Asia
albopictus
- potential role???
WHO/EPI/GEN/98.11
7 5
South American vertebrate hosts
Name
Environment
Comments
Alouatta
- live in groups in the crowns of trees
- very susceptible to yellow fever
(howler monkeys)
Ateles
- live in tree tops
- very susceptible
(spider monkeys)
Callithrix
- very susceptible
(marmosets)
Cebus
- relatively resistant
(capuchin monkeys)
Saimiri
- live in groups
- play a role in “wandering” epizootic”
(squirrel monkeys)
concept
7 6
Yellow fever
Bibliography
1.
Last JM. A Dictionary of Epidemiology. New York: Oxford Press, 1988.
2.
Robertson S. Yellow Fever; The Immunological Basis for Immunization 8.
WHO/EPI/GEN/93.18. Geneva, Switzerland, WHO, 1993.
3.
Division of Epidemiological Surveilance and Health Situation and Trend
Assessment. Global Health Situation and Projections: Estimates. Geneva,
Switzerland, WHO, 1992.
4.
Monath T. Yellow Fever. In: Monath T, editor. The Arboviruses;
Epidemiology and Ecology. Boca Raton, Florida: CRC Press, 1988;
139-231.
5.
Meegan JM. Yellow fever vaccine. WHO/EPI/GEN/91.06. Geneva,
Switzerland, WHO, 1991.
6.
Simpson DIH. Arbovirus infections. In: Cook GC, editor. Manson’s
Tropical Diseases. Bath, UK: Saunders, 1996; 637-42.
7.
Ministry of Health, Government of Kenya. Field Guide for Yellow Fever
Surveillance. Nairobi, Kenya, 1996.
8.
Reiter P, Cordellier R, Ouma JO, McLean RG, Cropp CB, Savage HM,
Sanders EJ, Marfin AA, Tukei PM, Agata NN, Gubler DJ. First
Recorded Outbreak of Yellow Fever in Kenya, 1992-1993. II.
Entomological Investigations. Not published.
9.
Monath TP. Yellow Fever: Victor, Victoria? Conqueror, Conquest?
Epidemics and Research in the Last Forty Years and Prospects for the
Future. Am. J. Trop. Med. Hyg: 1991; 45(1):1-43.
10.
WHO. Prevention and Control of Yellow Fever in Africa. Geneva,
Switzerland, WHO, 1986.
11.
Monath TP, Lee VH, Wilson DC, Fagbami A, Tomori O. Arbovirus
Studies in Nupeko Forest, a Possible Natural Focus of Yellow Fever Virus
in Nigeria. Trans. R. Trop. Med. Hyg. 1974; 68: 30-38.
WHO/EPI/GEN/98.11
7 7
12.
Strode GK, Bugher JC, Austin-Kerr J, Smith HH, Smithburn KC, Taylor
RM, Theiler M, Warren AJ, Whitman L, editors. Yellow Fever. New
York, McGraw-Hill Book Company, Inc. 1951.
13.
Mims CA, Playfair JH, Roitt IM, Wakelin D, Williams R, Anderson RM.
Medical Microbiology. London, UK: Mosby, 1995: 30.2.
14.
Peters W, Gilles HM. Color Atlas of Tropical Medicine and Parasitology.
London, UK: Mosby-Wolfe, 1995; 4-9.
15.
Robertson SE, Hull BP, Tomori O, Bele O, LeDuc J, Esteves K. Yellow
Fever. A Decade of Re-emergence. JAMA, 1996; 276(14): 1157-62.
16.
NasidiA, Monath TP, DeCock K, Tomori O, Cordellier R, Otaleye OD,
Harry TO, Adeniyi A, Sorungbe AO, Ajose-Coker AO, van Der Laan
G, Oyediran AB. Urban Yellow Fever Epidemic in Western Nigeria.
Trans. R. Soc. Trop. Med. Hyg. 1989;83: 401-06.
17.
Downs WG, Shope RE. Yellow Fever. In: Gear JHS, editor. CRC
Handbook of Viral and Rickettsial Hemorrhagic Fevers. Florida, USA:
CRC Press, 1984; 73-79.
18.
Downs WG. Malaria, the Great Umbrella. Bull. New York Ac. Med.
1975; 51: 984.
19.
Rickard ER. Organization of Viscerotome Service of Brazilian
Cooperative Yellow Fever Service. Am. J. Trop.Med. 1937; 17: 163.
20.
WHO. Inclusion of Yellow Fever Vaccine in the EPI, Gambia. Wkly
Epidemiol. Rec. 1996; 71: 181-85.
21.
WHO. Yellow Fever. Wkly Epidemiol. Rec. 1996; 71(42): 313-18.
22.
WHO. Field Guide for District Level Staff on Priority Communicable
Disease Surveillance. Brazzaville, Congo: WHO Regional Office for
Africa, 1994.
23.
Hobson W. World Health and History. Bristol: Wright, 1963.
24.
Carter HR. Yellow Fever: an Epidemiological and Historical Study of its
Place of Origin. Baltimore: The Williams & Wilkins Company, 1931.
25.
Smith A. Yellow Fever in Galveston, Republic of Texas, 1839. Austin,
Texas: University of Texas Press, 1951.
26.
Finlay CJ. Trabajos Selectos. Habana, Cuba, 1912.
27.
Kelly HA. Walter Reed and Yellow Fever. New York, McClune, Phillips
and Co, 1907.
7 8
Yellow fever
28.
Stokes A, Bauer JH, Hudson NP. Transmission of Yellow Fever to
Macacus Rhesus: Preliminary Note. JAMA, 1928; 90: 253-54.
29.
Stokes A, Bauer JH, Hudson NP. Experimental Transmission of Yellow
Fever to Laboratory Animals. JAMA, 1928;8: 103-64.
30.
Theiler M. The Development of Vaccines Against Yellow Fever - Les Prix
Nobel de 1951. Collected papers by members of the staff of the divisionof
medicine and public health of the Rockerfeller Foundation. New York:
Division of Medicine and Public Health of the Rockerfeller Foundation,
1952.
31.
Sawyer WA, Lloyd W. Use of Mice in Tests of Immunity Against Yellow
Fever. J. Exp. Med. 1931; 54: 533-55.
32.
League of Nations. Yellow Fever. League of Nations Monthly Epidemic
Report, 1928; 7(111): 57-69.
33.
League of Nations. Yellow Fever since the Beginning of 1931. League of
Nations Monthly Epidemic Report, 1932;11(160): 79-82.
34.
League of Nations. Yellow Fever in 1932-33. League of Nations Monthly
Epidemic Report, 1933; 12(169): 226-9.
35.
Biraud Y. Present-day Problems of Yellow Fever Epidemiology. League
of Nations Monthly Epidemic Report, 1935; 14(179): 103-173.
36.
League of Nations. Yellow Fever in 1933-4. League of Nations Monthly
Epidemic Report, 1934; 13(175): 207-9.
37.
Durieux C. Preparation of Yellow Fever Vaccine at Institut Pasteur,
Dakar. In: Smithburn KC, Duriex C, Koerber R, Penna HA, Dick GWA,
Courtois G, de Sousa Manso C, Stuart G, Bonnel PH. YF Vaccination.
Monograph Series No 30. Geneva: WHO, 1956: 31-32.
38.
Durieux C. Mass Yellow Fever Vaccination in French Africa South of
Sahara. In: Smithburn KC, Duriex C, Koerber R, Penna HA, Dick
GWA, Courtois G, de Sousa Manso C, Stuart G, Bonnel PH. YF
Vaccination. Monograph Series No 30. Geneva: WHO, 1956: 115-121.
39.
de Sousa Manso C. Mass Vaccination against Yellow Fever in Brazil. In:
Smithburn KC, Duriex C, Koerber R, Penna HA, Dick GWA, Courtois
G, de Sousa Manso C, Stuart G, Bonnel PH. YF Vaccination. Monograph
Series No 30. Geneva: WHO, 1956: 123-139.
40.
Penna MD. Production of 17D Yellow Fever Vaccine. In: Smithburn KC,
Duriex C, Koerber R, Penna HA, Dick GWA, Courtois G, de Sousa
Manso C, Stuart G, Bonnel PH. YF Vaccination. Monograph Series
No 30. Geneva: WHO, 1956: 67-68.
WHO/EPI/GEN/98.11
7 9
41.
Groot H, Ribeiro RB. Neutralizing and haemagglutination-inhibiting
antibodies to YF 17 years after vaccination with 17D vaccine. Bull.World
Health Organ. 1962; 27: 699-707.
42.
WHO. Yellow Fever in 1979. Wkly Epidemiol. Rec. 1980;
55(45): 345-51.
43.
Deubel V, Pailliez JP, Cornet M, Schlesinger JJ, Diop M, Diop A,
Digoutte JP, Girard M. Homogeneity among Senegalese Strains of YF
virus. Am.J.Trop.Med.Hyg. 1985; 34(5): 976-83.
44.
Chang GJ, Cropp CB, Kinney RM, Trent DW, Gubler DJ. Nucleotide
sequence variation of the envelope protein gene identifies two distinct
genotypes of yellow fever virus. J. Virol. 1995; 69(9): 5773-80.
45.
Dutary BE, Petersen JL, Peralta PH, Tesh RB. Transovarial Transmission
of Gamboa Virus in a Tropical Mosquito, Aedeomyia squamipennis. Am.
J. Torp. Med. Hyg. 1989; 40(1): 108-13.
46.
Beaty BJ, Tesh RB, Aitken THG. Transovarial Transmission of YF Virus
in Stegomyia Mosquitoes. Am J Trop Med Hyg. 1980; 29(1): 125-32.
47.
Fontenille D, Diallo M, Mondo M, Ndiaye M, Thonnon J. First Evidence
of Natural Vertical Transmission of Yellow Fever Virus in Aedes Aegypti,
Its Epidemic Vector. Trans. R. Soc. Trop. Med. Hyg. 1997; 91: 533-35.
48.
Cornet M, Robin Y, Heme G, Valade M. Isolement au Senegal Oriental
d’une Souche de Virus Amaril a Partir d’un Lot d’Aedes du Sous-genre
Diceromyia. C.R.Acad.Sci. Hebd. Seances Acad. Sci. D. 1978;
287(16): 1449-51.
49.
Digouette J-P, Cornet M, Deubel V, Downs WG. Yellow Fever. In:
Porterfield JS, editor. Exotic Viral Infections. London: Chapman & Hall,
1995; 67-102.
50.
Germain M, Saluzzo JF, Cornet JP, Herve JP, Suleau P, Camicas JL,
Robin Y, Salaun JJ, Heme G. Isolation of the Yellow Fever Virus from an
Egg-cluster of the Larvae of the Tick Amblyomma Variegatum.
C.R.Seanus.Acad.Sci.D. 1979: 289 (8): 635-7.
51.
Germain M, Francy DB, Monath TP, Ferrera L, Bryan J, Salaun JJ, et al.
YF in the Gambia, 1979-1979: Entomological Aspects and
Epidemiological Correlations. Am.J. Trop.Med.Hyg. 1980 29(5): 929-40.
52.
Smithburn KC. Rift Valley Fever : Neurotropic Adaptation of Virus and
Experimental Use of this Modified Virus as Vaccine. Brit. J. Exp. Path.
1949; 30 : 1-16.
53.
WHO. Yellow Fever in 1989 and 1990. Wkly Epidemiol. Rec. 1992;
67(33): 245-51.
8 0
Yellow fever
54.
WHO. Yellow Fever in 1987. Wkly Epidemiol. Rec. 1989; 64(6): 37-43.
55.
WHO. Yellow Fever in 1992 and 1993. Wkly Epidemiol. Rec. 1995;
70(10): 65-71.
56.
Vicens R, Robert V, Pignon D, Zeller H, Ghipponi PM, Digoutte JP.
Yellow Fever Epidemic in the Extreme North of Cameroon in 1990: First
Yellow Fever Virus Isolation in Cameroon. Bull. WHO1993; 71: 173-76.
57.
Pisano MR, Durand JP, Tolou H. Partial Genomic Sequence
Determination of Yellow Fever Virus Strain Associated with a Recent
Epidemic in Gabon. Acta Virol. 1996; 40: 103-05.
58.
WHO. Yellow Feverin 1985. Wkly Epidemiol. Rec. 1986; 61(49): 377-80.
59.
Maurice J. Yellow Fever Makes Comeback. Suomen Laakarilehti 1993;
48: 3057-61.
60.
Mondet B, da Rosa AP, Vasconcelos PF. The Risk of Urban Yellow Fever
Outbreaks in Brazil by Dengue Vectors, Aedes ASegypti and Aedes
Albopictus. Bulletin de la Societe de Pathologie Exotique 1996;
89(2): 107-13.
61.
Chippaux A, Deubel V, Moreau JP, Reynes JM. Current Situation of
Yellow Fever in Latin America. Bulletin de la Societe de Pathologie
Exotique 1993; 86(5): 460-64.
62.
PAHO. PAHO Advisory: Aedes Albopictus in the Caribbean. Epid. Bull
PAHO 1993; 14(3): 15-16.
63.
Vasconcelos PF, Rodrigues SG, Degallier N, Moraes MA, Travassos-Da-
Rosa JF, Travassos-Da-Rosa ES, et al. An Epidemic of Sylvatic Yellow
Fever in the Southeast Region of Maranhao State, Brazil, 1993-1994:
Epidemiological and Entomological Findings. Am. J. Trop. Med. Hyg.
1997; 57: 132-37.
64.
Traore-Lamizana M, Fontenille D, Zeller HG, Mondo M, Diallo M,
Adam F, et al. Surveillance for Yellow Fever Virus in eastern Senegal
during 1993. J. Med. Entomol. 1996; 33(5): 760-65.
65.
Dudley SF. Can Yellow Fever Spread into Asia? An Essey on the Ecology
of Mosquito-borne Disease. J. Trop Med. Hyg. 1934; 37(18): 273-78.
66.
Mahaffy AF, Hughes TP, Smithburn KC, Kirk R. Isolation of Yellow
Fever Virus in Anglo-Egyptian Sudan. Annals of Tropical Medicine and
Parasitology, 1941; 35: 141-48.
67.
WHO. Annual Epidemiological and Vital Statistics. Geneva, WHO, 1964.
WHO/EPI/GEN/98.11
8 1
68.
WHO. Yellow Fever. Wkly Epidemiol. Rec. 1995; 70(24): 175-6.
69.
WHO. Yellow Fever. Wkly Epidemiol. Rec. 1996; 71(13): 103.
70.
Sinha MK, Majumder NM. Will the Present Health Check-up System
Invite Yellow Fever in India? Indian Journal of Public Health, 1990;
34(2): 119-21.
71.
Sawyer WA, Bauer JH, Whitman L. Distribution of Yellow Fever
Immunity in North America, Central America, West Indies, Europe, Asia
and Australia with Special Reference to Specificity of Protection Test.
Am. J. Trop. Med 1937; 17: 137-61.
72.
Pym W. Observations upon the Bulam Fever which has of Late Years
Prevailed in the West Indies, on the Coast of America, at Gibraltar and
Other Parts of Spain. London; J. Callow, 1815.
73.
Wallbridge M, Downs WG. The Communicability of Yellow Fever. Br.
Guiana Med. Ann. 1891; 3: 7-18.
74.
Elton NW. Sylvan Yellow Fever in Central America. Publ Health Rep.
1952; 67(5): 426-32.
75.
Matas R. Nursing in Yellow Fever and the Duties of Trained Nurses in
Epidemics. Trained Nurse Hosp Rev 1905; Oct. -Dec. 1905: 3-24.
76.
Tsai T, Mitchell C. St. Louis Encephalitis. In: Monath TP, editor. The
Arboviruses: epidemiology and ecology. Boca Raton, Florida: CRC Press,
1988.
77.
Gordon-Smith CE, Turner LH, Armitage P. Yellow Fever Vaccination in
Malaya by Subcutaneous Injection and Multiple Puncture. Bull. WHO,
1962; 27: 717-27.
78.
Theiler M, Anderson CR. The Relative Resistance of Dengue-immune
Monkeys to YF Virus. Am J Trop Med Hyg;1975 24: 115.
79.
Huang Y-M. Aedes (Stegomyia) Bromeliae (Diptera: Culicidae), the
Yellow Fever Virus Vector in East Africa. J. Med. Entomol. 1986; 23(2):
196-200.
80.
Mitchell CJ, Miller BR, Gubler DJ. Vector Competence of Aedes
Albopictus from Houston, Texas, for Dengue Serotypes 1 to 4, Yellow
Fever and Ross River Viruses. JAMA, 1987;3(3): 460-65.
81.
Moore CG, Francy DB, Eliason DA, Monath TP. Aedes Albopictus in the
United States: Rapid Spread of a Potential Disease Vector. Journal of the
American Mosquito Control Association, 1988; 4 (3): 356-61.
8 2
Yellow fever
82.
Hindle E. An Experimental Study of Yellow Fever. Trans. R. Soc. Trop.
Med. Hyg. 1929; 22: 405.
83.
Tabachnick WJ, Wallis GP, Aitken THG, Miller BR, Amato GD, Lorenz
L, Powell JR, Beaty BJ. Oral Infection of Aedes Aegypti with YF Virus;
Geographic Variation and Genetic Considerations. Am J Trop Med
Hyg.1985 34(6): 1219-24.
84.
Miller BR, Mitchell CJ, Ballinger ME. Replication, Tissue Trophisms, and
Transmission of Yellow Fever Virus in Aedes Albopictus. Trans. R. Soc.
Trop. Med. Hyg. 1989; 83: 252-55.
85.
Brenzel L, Claquin P. Immunization Programs and Their Cost. Social
Science and Medicine 1994; 39(4): 527-36.
86.
Monath TP. Stability of Yellow FeverVaccine. In: Brown F, editor. New
Approaches to Stabilisation of Vaccines Potency: WHO Headquarters,
Geneva, May, 29-31, 1995. Basel, Karger, 1996: 219-25.
87.
Monath TP, Nasidi A. Should YF vaccine be included in the expanded
program of immunization in Africa? A cost-effectiveness analysis for
Nigeria. Am.J.Trop.Med.Hyg. 1993;48(2): 274-299.
88.
Robertson RL, Foster SO, Hull HF, Williams PJ. Cost-effectiveness of
immunization in the Gambia. Am.J.Trop.Med.Hyg. 1985;88(6):343-51.
89.
Dietz V, Cutts F. The Use of Mass Campaigns in the Expanded
Programme on Immunization: Advantages and Disadvantages.
Internal Journal of Health Services 1997; 27(4): 767-90.
90.
Chambon L, Wone I, Bres P, Cornet M, Cire L, Michel A, et al. Une
Epidemie de fievre jaune au Senegal en 1965. Bull. WHO, 1967;
36: 113-50.
91.
Soula G, Sylla A, Pichard E, Kodio B, Bentejac MC, Teulieres L, Saliou
P.. Etude d’un nouveau vaccine combine contre le fievre jaune et la
rougeole chez les enfants ages 6 a 24 mois au Mali. Bull. Soc. Path. Exper.
1991; 84: 885-97.
92.
Lhuillier M, Mazzariol MJ, Zadi S, Le-Cam N, Bentejac MC,
Adamowiez L, MarieFN, Fritzell B. Study of Combined Vaccination
against Yellow Fever and Measles in Infants from Six to Nine Months. J.
Biol. Stand. 1989; 17: 9-15.
93.
Mouchon D, Pignon D, Vicens R, Tu-Ha-Thanh, Tekaia F, Teulieres L,
Garrigue G.. Etude de la Vaccination Combine Rougeole-Fievre Jaune
chez l’Infant Africain Age de 6 Mois a 10 Mois. Bull Soc. Path. Exper.
1990: 83: 537-51.
94.
Galazka A. Stability of Vaccines. WHO/EPI/GEN 89.8:1990.
WHO/EPI/GEN/98.11
8 3
95.
Monath TP. Surveillance of Yellow Fever in Nigeria, 1970-71.
Nigerian Med. J. 1972; 2(4):179.
96.
Rickard ER. Organization of Viscerotome Service of Brazilian
Cooperative Yellow Fever Service. Am.J.Trop.Med. 1937;17: 163-83.
97.
PAHO. Yellow Fever in the Americas. PAHO Bull. 1985; 19(2): 209-12.
98.
Vainio JJ. Evaluation of Yellow Fever Surveillance in Kenya
(MSc Control of Infectious Diseases). University of London, 1997.
99.
Sanders EJ, Borus P, Ademba G, Kuria G, Tukei PM, LeDuc JW. Sentinel
Surveillance for Yellow Fever in Kenya, 1993 to 1995.
Emerging Infectious Diseases, 1996; 2(3): 236-238.
100.
EPI. EPI Information System: Global Summary. Geneva: Expanded
Programme on Immunization, Global Programme for Vaccines and
Immunization, WHO, 1996.
101.
Busvine JR. Disease Transmission by Insects: Its Discovery and 90 Years
of Effort to Prevent It. New York: Springer-Verlag, 1993.
102.
Leger M. Epidemiology of Yellow Fever in the French West African
Colonies. League of Nations Monthly Epidemic Report, 1925;
4(83): 472-80.
103.
Scott HH. A History of Tropical Medicine. London; Edward Arnold &
Co, 1942.
104.
Moe CE. Yellow Fever and Mosquito Control. Public Admin. Series:
Bibliography # P 479, 1979.
105.
Coleman W. Yellow Fever in the North - the Methods of Early
Epidemiology. Madison, Wisconsin: The University of Wisconsin Press,
1987.
106.
Newsom EY. Unto the Least of These : the Howard Association and
Yellow Fever. Southern Med. J. 1992; 85(6): 632-37.
107.
Monath TP. Chapter 71: Yellow Fever. IN: Warren KS, Mahmoud AAF,
editors. Tropical and Geographic Medicine. 2nd ed. New York:
McGraw-Hill, Inc. 1990: 661-74.
108.
Mahaffy AF, Smithborn KC, Jacobs HR, Gillett JD. Yellow Fever in
Western Uganda. Trans.R.Trop.Med.Hyg. 1942; 36: 9.
109.
Serie C, Lindrec A, Poirier A, Andral L, Neri P. Etudes sur la Fievre
Jaune en Ethiopie.I. Introduction - symptomatologie clinique amarile.
Bull. WHO 1968; 38(6): 835-41.
8 4
Yellow fever
110.
Haddow AJ. Yellow Fever in Central Uganda, 1964, part I.
Trans.R.Soc.Trop.Med.Hyg. 1965;59: 436.
111.
Haddow AJ. Mosquitoes of Bwamba County, Uganda III. Bull. Entomol.
Research 1945; 36: 297-304.
112.
Deubel V, Digoutte JP. Morphogenesis of Yellow Fever Virus in Aedes
Aegypti Cultured Cells. I. Isolation of Different Cellular Clones and the
Study of Their Susceptibility to Infection with the Virus. Am. J. Trop.
Med Hyg. 1981; 30(5): 1060-70.
113.
Deubel V, Digoutte JP, Mattei X, Pandare D. Morphogenesis of Yellow
Fever Virus in Aedes Aegypti Cultured Cells. II. An Ultrastructural
Study. Am. J. Trop. Med. Hyg. 1981; 30(5): 1071-77.
114.
Gilpin ME, McClelland GA. Systems Analysis of the Yellow Fever
Mosquito Aedes aegypti. Fortsch. Zool. 1979; 25(2-3): 355-88.
115.
Lumsden WHR. Probable Insect Vectors of YF Virus from Monkey to
Man, in Bwamba County, Uganda. Bull Entomol Res. 1951; 42: 317.
116.
Dick GWA. Further Studies on the Susceptibility of African Wild
Mammals to YF. Trans R Soc Trop Med Hyg. 1952; 46: 47-58.
117.
Durieux C, Boiron H, Koerber R. Sur l’Existence d’un Reservoir de Virus
Amaril Animal en Afrique. Bull. Soc. Pathol. Exot. 1947; 40: 111-18.
118.
Findlay GM. The Infectivity of Neurotropic Yellow Fever Virus for
Animals. J. Path. Bact. 1934; 38: 1-6.
119.
Findlay GM, Mahaffy AF. Susceptibility of Nigerian Hedgehogs to YF.
Trans R Soc Trop Med Hyg. 1936; 29: 417-18.
120.
Findlay GM, Stephanopoulo GJ, Davey TH, Mahaffy AF. Yellow Fever
Immune Bodies in Blood of African Animals. Preliminary Observations.
Trans. R. Soc. Trop. Med. Hyg. 1936; 29: 419-24.
121.
Haddow AJ, Dick GWA, Lumsden WHR, Smithburn KD. Monkeys in
Relation to the Epidemiology of YF in Uganda. Trans R Soc Trop Med
Hyg. 1951; 45: 189-224.
122.
Haddow AJ, Smithburn KC, Mahaffy AF, Bugher JC. Monkeys in
Relation to Yellow Fever in Bwamba County, Uganda. Trans. R. Soc.
Trop. Med. Hyg. 1947; 40: 677-700.
123.
Monath TP, Kemp GE. Importance of Non-human Primates in YF
Epidemiology in Nigeria. Tropical and geographical medicine, 1973;
25: 28.
WHO/EPI/GEN/98.11
8 5
124.
Smithburn KC. The Susceptibility of African Wild Animals to Yellow
Fever. III. Pottos and Galagos. Am. J. Trop. Med. 1949; 29: 411-23.
125.
Kumm HW, Cerqueira NL. The Role of Aedes Leucocelaneus in the
Epidemiology of Jungle Yellow Fever in Brazil. Bull. Entomol. Research,
1951; 42(1): 195-99.
126.
Schliessmann DJ. Progress Report of the Aedes Aegypti Eradication
Program in the United States for 1965. Mosq News 1965; 26: 484-89.
127.
Gillette HPS. YF in Trinidad. A Brief Review. Mosq News, 1956;
16: 121-25.
128.
Boshell MJ, Osorno Mesa E. Observations on Epidemiology of Jungle YF
in Santander and Boyaca, Colombia, September, 1941 to April, 1942.
Am J Hyg. 1944; 40: 170-81.
129.
Yellow Fever in the Americas, 1981 and 1982. Morbidity and Mortality
Weekly Report 1983; 32: 202.
130.
Laemmert HW, de Castro Ferreira L, Taylor RM. An Epidemiological
Study of Jungle Yellow Fever in an Endemic Area in Brazil; Part II -
Investigation of Vertebrare Hosts and Arthropod Vectors. Am. J. Trop.
Med. Supp. 1946; 26: 23-69.
131.
Bugher JC, Boshell MJ, Roca Garcia M, Osorno ME. Epidemiology of
Jungle YF in Eastern Colombia. Am J Hyg. 1944; 39: 16-51.
132.
Miller BR, Ballinger ME. Aedes Albopictus Mosquitoes Introduced into
Brazil: Vector Competence for Yellow Fever and Dengue Viruses. Trans.
R. Soc. Trop. Med. Hyg. 1988; 82: 476-77.
133.
WHO. Galazka A, Milstien J, Zaffran M. Thermostability of vaccines.
EPI. Advance draft. WHO/EPI/GEN/97. Rev.1, 1997.
134.
WHO. Immunization policy. WHO/EPI/GEN/95.03. 1995.
8 6
Yellow fever
WHO/EPI/GEN/98.11
8 7
Document Outline
- Front cover
- Contents
- Glossary
- Abbreviations
- Preface
- Summary
- I. Introduction
- II. Historical review
- III. Epidemiology
- IV. Cost effectiveness of YF vacc.
- V. Surveillance
- VI. Comments & suggestions
- Summary of recommendations
- Appendix I
- Appendix II
- Appendix III
- Appendix IV
- Appendix V
- Bibliography
