DISEASE DESCRIPTION

In its severe form, yellow fever (YF) is characterized by three clinical periods: the period of infection (about 3 days), the period of remission, and the period of intoxication. The incubation period ranges from 3 to 6 days. Onset is sudden, with fever accompanied by headache, weakness, lumbosacral pain, anorexia, nausea, and vomiting. Physical findings include torpor, relative bradycardia, conjunctival congestion, dry skin, fetid breath, and abdominal or lumbar tenderness. Fever may remit briefly on the second or third day, then return with increasing intensity. The period of intoxication is characterized by jaundice, oliguria, albuminuria, and hemorrhages (e.g., hemoptysis, epistaxis, metrorrhagia).

Death occurs usually between the seventh and tenth day of illness. Early death between the second and sixth day may be observed (Serie et al., 1968). Convalescence is slow and may last up to 3 months.

A live attenuated vaccine (17D strain) produced in embryonated chicken eggs has been used widely to prevent YF, especially in endemic areas of Africa and Latin America. The experience acquired over the last 4 decades, during which millions of doses of 17D vaccine have been administered, has demonstrated the immunogenicity of the vaccine (with a seroconversion rate of more than 99 percent), its relative safety, and its efficacy in preventing in clinical illness (Pan American Health Organization, 1981).

The need for a new vaccine stems in part from evidence that infants under 6 months of age may develop neurological complications when immunized with the 17D vaccine (Pan American Health Organization, 1984). Also, the reappearance of the insect vector of YF virus in some urban centers in South America and the potential for YF transmission to susceptible areas elsewhere has led to concern that the demand for the vaccine could increase suddenly. Modern cell culture techniques could greatly improve the speed and economy of vaccine production.

PATHOGEN DESCRIPTION

Yellow fever virus is an arthropod-borne virus in the Flavivirus genus of the family Togaviridae. It shares group-specific antigens with other members of the genus (former group B viruses, e.g., in Africa: Zika, West Nile, Wesselsbron, dengue, Uganda S, Spondweni, Banzi, and other viruses) (Wildy, 1971).

Yellow fever and other flaviviruses contain a single-stranded, positive-polarity RNA genome. Viral particles are 43 nm in size; they contain a ribonucleoprotein core and a lipoprotein envelope. The virus is inactivated by deoxycholate, ether, proteases, and lipases. The envelope contains a single glycoprotein with type- and group-specific antigens.

The morphogenesis of YF virus is similar to that observed for other flaviviruses; that is, viral synthesis and maturation appear to be predominant in the rough endoplasmic reticulum. The formation of the surrounding envelope of the virion remains unclear. Mature virus particles accumulate within the cisternae of membranous organelles and are released extracellularly by exocytosis or by plasma membrane rupture.

The virus is pathogenic for adult mice by intracerebral inoculation, and for suckling mice by intracerebral, subcutaneous, and intraperitoneal inoculation. The rhesus monkey is highly susceptible to YF virus, and this animal may be used as a model to define the pathogenesis of the disease (Monath et al., 1981).

Yellow fever virus replicates in cell cultures of different origin, but the cell cultures are variously sensitive. Cell lines of mosquito, monkey kidney, and hamster kidney are useful for propagation and assay. The 17D attenuated strain can be grown in several cell substrates, such as primary or subcultured chick embryo fibroblast and monkey kidney cells; virus titers observed in these systems are comparable to those obtained in embryonated eggs (Pan American Health Organization, 1981).

Wild strains of YF virus vary in their pathogenicity for hosts, but the molecular basis for virulence is poorly understood. Host factors, including genetic and immunological parameters, probably affect susceptibility.

HOST IMMUNE RESPONSE

Neutralizing antibodies usually are detectable on the sixth or seventh day after onset of primary infection and are responsible for immune elimination of the virus. It is not unusual to find both infectious virus and antibody in serum, but the role of immune complexes in the pathogenesis of the disease remains uncertain. Antibody responses may be accelerated and broadened in individuals with prior flaviviral immunity. Yellow fever virus infection also may alter the immune response itself. Depression of delayed hypersensitivity (tuberculin skin test reactivity) has been observed after administration of 17D vaccine (Monath, 1984). Whether the B-cell necrosis described in YF of rhesus monkeys is associated with suppression of antibody formation is not known.

DISTRIBUTION OF DISEASE

Geographic Distribution

Yellow fever is endemic in extensive parts of tropical South America and sub-Saharan Africa. The annual incidence of officially reported YF cases varies from 50 to 300 cases in South America and from 5 to 700 cases in Africa (Table D-19.1). Investigations during several outbreaks indicate that morbidity and mortality rates are significantly under-estimated (Monath, 1984).

TABLE D-19.1

Yellow Fever in Africa and the Americas During the Period 1965 to 1983.

Bolivia, Brazil, Colombia, Ecuador, and Peru account for the majority of cases in South America. Virus reservoirs are maintained in tropical forests, such as those of the Amazon region and the Orinoco and Magdalena valleys. In some years, the virus can extend to other areas, including Central America, northern Argentina and Paraguay, and Trinidad and Tobago, causing significant outbreaks. Currently YF in the Americas occurs exclusively in its jungle form.

In Africa, YF is endemic or epidemic in 29 countries of the tropical zone between the fifteenth parallel north and the tenth parallel south. It occurs as sporadic cases of jungle YF, or as outbreaks mainly in savannah areas. Several sizable epidemics have occurred during the past 5 years, such as in Ghana (1977 to 1979 and 1983), Gambia (1978 to 1979), Senegal (1981), Ivory Coast (1982), and Burkina Faso (formerly Upper Volta) in 1983.

Disease Burden Estimates

The disease burden estimates for YF appear in Tables D-19.2. through D-19.4.

TABLE D-19.2

Disease Burden: Yellow Fever (Africa).

TABLE D-19.4

Disease Burden: Yellow Fever—All.

Disease burden estimates were based on reported cases of YF to the World Health Organization, multiplied by a correction factor. The correction factor was determined by dividing the number of reported cases in certain areas into more realistic assessments based on epidemiological investigation in these areas. A median value was taken from the range of values derived from one South American and five African studies (Table D-19.5).

TABLE D-19.5

Comparison of Epidemiologically Determined Cases to Reported Cases of Yellow Fever.

From Table D-19.1 an average reported annual incidence was derived from the most recent 10 years for Africa and South America. These values, 181.3 and 136.3, were multiplied by the correction factor of 257.6 to get an annual incidence rate of 46,707 and 35,114 for Africa and South America, respectively.

Because of YF's different epidemiological patterns, the distribution of cases into categories of severity and age groups was calculated differently for the two continents. For Africa, the estimated age distribution of cases is 15 percent in the under 5 years age group, 45 percent in the 5 to 14 years age group, 40 percent in the 15 to 59 years age group, and virtually 0 in the 60 years and over age group. The estimated breakdown of cases in each age group is as follows: under 5 years of age, 50 percent in morbidity category A, 30 percent in category B, 20 percent in category C; 5–14 years of age, 40 percent in category A, 30 percent in category B, 30 percent in category C; 15 to 59 years of age, 30 percent in category A, 30 percent in category B, 40 percent in category C. Deaths (category H) are estimated to be half of the severe cases (category C) in each age group.

TABLE D-19.3Disease Burden: Yellow Fever (Americas)

View in own window

		Under 5 Years		5–14 Years		15–59 Years		60 Years and Over
Morbidity Category	Description	Number of Cases	Duration	Number of Cases	Duration	Number of Cases	Duration	Number of Cases	Duration
A	Moderate localized pain and/or mild systemic reaction, or impairment requiring minor change in normal activities, and associated with some restriction of work activity	211	7	702	10	5,969	10	140	10
B	Moderate pain and/or moderate impairment requiring moderate change in normal activities, e.g., housebound or in bed, and associated with temporary loss of ability to work	737	14	2,458	20	20,893	20	491	20
C	Severe pain, severe short-term impairment, or hospitalization	105	21	351	30	2,985	30	70	30
D	Mild chronic disability (not requiring hospitalization, institutionalization, or other major limitation of normal activity, and resulting in minor limitation of ability to work)		n.a.		n.a.		n.a.		n.a.
E	Moderate to severe chronic disability (requiring hospitalization, special care, or other major limitation of normal activity, and seriously restricting ability to work)		n.a.		n.a.		n.a.		n.a.
F	Total impairment		n.a.		n.a.		n.a.		n.a.
G	Reproductive impairment resulting in infertility		n.a.		n.a.		n.a.		n.a.
H	Death	53	n.a.	176	n.a.	1,492	n.a.	35	n.a.

For South America, the age distribution is estimated differently: under 5 years, 3 percent; 5 to 14 years, 10 percent; 15 to 59 years, 85 percent; and 60 years and over, 2 percent. For each age group, 20 percent of cases are assumed to fall in category A, 70 percent in B, and 10 percent in C. Half of category C cases die, and they comprise category H.

PROBABLE VACCINE TARGET POPULATION

Yellow fever vaccination is recommended for all persons who live in rural communities in endemic areas and whose occupations bring them into forests where the virus circulates. It is also recommended for persons who plan to visit an endemic zone.

The majority of cases recorded in South America involve males 15 to 45 years old, but children in the 1 to 4 years age group also can be affected (Pan American Health Organization, 1983).

In Africa, children and young adults up to 15 years of age are mainly affected by the severe form of the disease. In recent outbreaks recorded in Ivory Coast (1982) and Burkina Faso (1983), most cases occurred among children (World Health Organization, 1984).

There is evidence that a significant proportion of infants under 6 months of age may develop neurological complications when immunized with the 17D vaccine (Louis et al., 1981). Hence, the existing vaccine should be given only after 6 months of age, and some authorities withhold it until 1 year of age (Pan American Health Organization, 1984).

General adoption of the World Health Organization Expanded Program on Immunization (WHO-EPI) offers an opportunity to add an improved YF vaccine to the six vaccines already included in the program, providing a vaccine that is adequately safe in young children can be developed. In risk areas (which have to be defined in each country), the YF vaccination could be coupled with the measles vaccine; a single injection should provide lifelong immunity. The cold chain requirements would be similar to those for the measles vaccine. Herd immunity could be built up progressively in endemic areas, but national authorities would have to be aware that a risk group would remain among individuals older than the EPI-vaccinated group. New techniques for the preparation of YF vaccine should lower the cost of the vaccine to about the same as that of the measles vaccine.

SUITABILITY FOR VACCINE CONTROL

The epidemiological features of YF dictate that its prevention by active immunization is highly desirable and practicable.

Large-scale immunization against YF has been performed for more than 4 decades with considerable success. Certain problems (failure to immunize, post-vaccinal encephalitis, and hepatitis) observed in early years of the vaccination program were readily overcome by adopting appropriate measures. For the reasons noted above, an improved vaccine is needed.

If urban YF were to reappear in South America, the consequences could be catastrophic. However, an urban population can be immunized rapidly if adequate stocks of vaccine exist and vaccination teams are available.

PROSPECTS FOR DEVELOPMENT OF A YELLOW FEVER CELL CULTURE VACCINE

Most YF vaccine used at the present time in South America is produced by two national government controlled laboratories, one in Rio de Janeiro, Brazil, and the other in Bogota, Colombia. Although these two laboratories have provided an uninterrupted supply of vaccine, until recently their procedure for producing the vaccine from eggs remained basically the same as when the vaccine was first introduced 40 years ago. Since 1981, however, the Pan American Health Organization (PAHO) has been actively involved in improving the quality of YF vaccine produced in the two laboratories. As a result, significant progress has been achieved in modernizing the production facilities and techniques for vaccine manufacture in both countries.

Despite these improvements, efforts are under way to develop new techniques for the production of YF vaccine. These efforts have resulted, in part, from increased concern over the risk of YF extending to certain cities near jungle YF foci in South America. These cities have been heavily “reinfested” with Aedes aegypti (an urban vector of YF virus). Moreover, other areas in the world, such as North Africa, the Middle East, Southeast Asia, and the Far East, are known to be vulnerable to YF because of the presence of the same mosquito vector in high densities. The reappearance of YF in urban centers of South America, and its importation into susceptible areas elsewhere, would immediately produce a great demand for the vaccine. The demand would be very difficult to meet with currently available vaccine stocks. The reappearance of urban YF transmission at a time when the world's stock of YF vaccine is said to have dropped to a dangerously low level could be catastrophic, not only for the Americas, but also for the rest of the world.

Development of a YF vaccine grown in cell culture by modern techniques is a solution to the problem. Cell culture would greatly improve the speed and economy of vaccine production and provide for rapid expansion in the event of an emergency. In addition, safety would be improved because the amount of potentially allergenic proteins in the vaccine would be reduced. Because of its purity, a YF cell culture vaccine also would be more suitable for combined use with other live viral vaccines, such as measles.

Current experience indicates that efforts to develop a 17D cell culture vaccine will be successful. The 17D vaccine virus has been grown on various cell substrates up to titres comparable to those obtained in embryonated eggs (Pan American Health Organization, 1981). These substrates include primary chick and duck embryo fibroblasts, as well as a diploid cell strain. Many have been used for other vaccines and have met WHO safety requirements. It is expected that the efficacy of a 17D cell culture vaccine will be excellent (90 to 100 percent), with fewer side effects than the currently available embryonated egg vaccine.

It is anticipated that studies required for developing a YF cell culture vaccine will encompass four phases:

1.

the development, characterization, production, and certification of a primary seed virus for a cell culture 17D vaccine

2.

research, preferably conducted in an existing YF vaccine production facility, leading to the development of production protocols for a cell culture YF vaccine

3.

research to improve biological markers and reduce neurovirulence of the 17D vaccine and to define the dose response of the product on the basis of studies in man

4.

formulation and evaluation of more satisfactory thermal stabilization agents for the currently available YF vaccine.

If chick embryo cell cultures are used as the substrate, they should be derived from embryonated eggs from a monitored, specific pathogen-free flock of chickens. Diploid cell culture products must meet WHO requirements for freedom from cellular DNA and should be used only if they have been used in the past for a live attenuated virus vaccine. The original seed virus should be a 17D derivative that is free from leucosis virus and that meets WHO requirements for a YF fever vaccine seed.

The thermal instability and relatively short shelf life of the live-attenuated YF vaccine probably could be corrected through inexpensive and straightforward investigations. Such investigations could provide a cheaper, longer lasting, and more abundant YF vaccine.

At present, basic information about the 17D virus strain is lacking. Basic research studies on the biological and biochemical properties of this virus strain should be conducted simultaneously with the investigations described above to provide a better understanding of genetic variation from passage in cell cultures. Particular emphasis should be given to defining genetic markers useful for the characterization of cell culture derived vaccines. The availability of in vitro and in vivo markers for the characterization of the cell-culture adapted YF seed to be used for vaccine production is vitally important, and several marker systems should be developed for this purpose. The markers should be shown to be reliable and reproducible, and each passage of virus should be monitored for changes.

Currently, YF 17D vaccines are not recommended for use in infants under 6 months of age, and some countries do not require the International Certificate of Vaccination under 1 year of age. The incidence of encephalitis following YF 17D vaccination of infants is not known with any certainty and should be studied further. In one analysis, the incidence was estimated to be at least 0.3 percent in infants under 6 months of age (Louis et al., 1981). Because immunization of infants younger than 6 months of age is desirable in many circumstances, reduction of the existing level of neurovirulence of 17D vaccine should be a goal of YF research.

Such a research project could produce a vaccine ready for initial trials in humans in about 2 years, depending to some extent on the amount of vaccine and seed virus testing in monkeys. In any event, vaccine testing in humans is not expected to be a special problem.

REFERENCES

• Louis, J.J., P.Chopard, and F.Larbre. 1981. Un cas d'encephalite vaccination anti-amarile par la souche 17D. Pediatrie 36(7):547–550. [PubMed: 6119675]
• Monath, T.P. 1984. Yellow fever. Pp. 636–651 in Tropical and Geographic Medicine, K.S.Warren, editor; and A.F.Mahmoud, editor. , eds. New York: McGraw-Hill.
• Monath, T.P., K.R.Brinker, F.W.Chandler, G.E.Kemp, and C.B. Cropp. 1981. Pathophysiologic correlations in a rhesus monkey model of yellow fever. Am. J. Trop. Med. Hyg. 30:431–443. [PubMed: 7235133]
• Pan American Health Organization. 1981. PAHO meeting on modernization of yellow fever vaccine production techniques. Washington, D.C., January 12–14, 1981. Internal document.
• Pan American Health Organization. 1983. Yellow fever in the Americas, 1981–1982. Epidemiol. Bull. PAHO 4(1):1–5.
• Pan American Health Organization. 1984. PAHO meeting to develop guidelines and protocols for the production of a yellow fever vaccine in cell cultures. Washington, D.C., February 21–23, 1984. Internal document.
• Pinheiro, F.P. 1981. Yellow fever. Pp. 1155–1160 in International Textbook of Medicine, Medical Microbiology and Infectious Diseases, A.I.Braude, editor. , ed. Philadelphia: W.B.Saunders.
• Pinheiro, F.P. 1982. Unpublished data, Pan American Health Organization, Washington, D.C.
• Pinheiro, F.P., A.P.Travassos da Rosa, M.A.Moraes, J.C.Almeida Neto, S.Camargo, and J.P.Filgueirs. 1978. An epidemic of yellow fever in central Brazil, 1972–1973. I. Epidemiological studies. Am. J. Trop. Med. Hyg. 27(1):125–132. [PubMed: 626268]
• Poland, J.D., C.H.Calisher, T.P.Monath, W.G.Downs, and K.Murphy. 1981. Persistence of neutralizing antibody 30–35 years after immunization with 17D yellow fever vaccine. Bull. WHO 59(6):895–900. [PMC free article: PMC2396120] [PubMed: 6978196]
• Serie, C., A.Lindrec, A.Poirier, L.Andral, and P.Neri. 1968. Etudes sur la fievre jaune en Ethoipie. I. Introduction, Symptomatologie clinique amarile. Bull. WHO 38:835. [PMC free article: PMC2554517] [PubMed: 5303659]
• Wildy, P. 1971. Classification and nomenclature of viruses. Monographs in Virology, Volume 5. Basel, N.Y.; Karger.
• World Health Organization Expert Committee on Yellow Fever. 1971. Third Report. TRS No. 479. Geneva: World Health Organization.
• World Health Organization. 1984. Yellow fever in 1983. WHO Wkly. Epidemiol. Rec. 43:329–335.

Footnotes

*

Vaccine preventable illness is defined as that portion of the disease burden that could be prevented by immunization of the entire target population (at the anticipated age of administration) with a hypothetical vaccine that is 100 percent effective (see Chapter 7).

The committee gratefully acknowledges the efforts of F.P.Pinheiro, who prepared major portions of this appendix, and the advice and assistance of A.Shelokov and T.Monath. The committee assumes full responsibility for all judgments and assumptions.