RUBELLA EPIDEMIOLOGY IN AFRICA IN THE PREVACCINE ERA, 2002-2009

Tuesday, 8th of September 2015

RUBELLA EPIDEMIOLOGY IN AFRICA IN THE PREVACCINE ERA, 2002–2009

James L. Goodson1, Balcha Masresha2, Annick Dosseh3, Charles Byabamazima4, Deogratias Nshimirimana2, Stephen Cochi1 and Susan Reef1

+ Author Affiliations

1. ¹Global Immunization Division, Centers for Disease Control and Prevention, Atlanta, Georgia
2. ²Immunisation and Vaccines Development Programme, World Health Organization, Regional Office for Africa, Brazzaville, Democratic Republic of the Congo
3. ³West Africa Inter-Country Support Team, Ouagadougou, Burkina Faso
4. ⁴East and South Africa Inter-Country Support Team, Zimbabwe

1. Correspondence: James L. Goodson, MPH, Global Immunization Division, Centers for Disease Control and Prevention, 1600 Clifton Road NE, MS-E05, Atlanta, GA, 30333 (jgoodson@cdc.gov).

This article is best viewed, with figures, at http://jid.oxfordjournals.org/content/204/suppl_1/S215.long

Rubella virus infection is transmitted by respiratory droplets and causes a generally mild disease characterized by a rash and fever, primarily in children. However, infection in women during early pregnancy may cause fetal death or congenital rubella syndrome (CRS) in the infant [1]. CRS is a significant cause of deafness, blindness, congenital heart disease, and mental retardation [2]; although precise burden of disease is unknown, it is estimated that 110,000 CRS cases occur each year in developing countries [1, 3].

Rubella is vaccine-preventable; the primary objective of rubella-control programs is prevention of congenital rubella virus infection, which includes CRS. Rubella is among the small number of viral diseases considered to be potentially eradicable [4, 5]. Rubella can be eliminated in countries that have introduced routine rubella vaccination for children and achieved high coverage in the population. In September 2010, achievement of the goal of rubella and CRS elimination in the region of the Americas was announced by the Pan American Health Organization [6, 7, 8]. However, countries that introduce rubella vaccine and achieve suboptimal vaccination coverage may be at risk for a paradoxical increase in susceptibility among older age groups, potentially leading to acquisition of rubella virus infections among women of childbearing age and to an increase in CRS cases [9–11]. To decrease the risk of rubella virus infections among pregnant women and consequent CRS cases, the World Health Organization (WHO) recommends introduction of rubella vaccine should be considered only in countries that have achieved high (>80%) coverage with the first-dose measles-containing vaccine [1].

WHO recommends that countries without rubella vaccination programs should assess the burden of rubella and CRS [2]. Integrated case-based surveillance with laboratory testing to detect measles and rubella is recommended in countries with an established measles elimination or rubella control goal [12]. In Africa, several countries have conducted subnational rubella seroprevalence surveys; however, none has established routine surveillance for CRS.

Since 1999, as part of the WHO and United Nations Childrens Fund (UNICEF) measles mortality reduction strategy in Africa, case-based surveillance with laboratory testing for all suspected measles cases has been established. By December 2008, 40 of 46 countries in the African Region of WHO—all except Algeria, Comoros, Guinea Bissau, Mauritius, Sao Tome and Principe, and Seychelles—established measles and rubella case-based surveillance [13] following the WHO African Regional Office (AFRO) measles-surveillance guidelines [14]. Laboratory testing for rubella IgM antibody is recommended for specimens found to be negative or indeterminate for measles IgM antibody [14].

Given that little is known about rubella in Africa, we provide an overview of published seroprevalence surveys and analysis of surveillance and laboratory data as a baseline of prevaccine-era rubella epidemiology in Africa, in preparation for the introduction and widespread use of rubella vaccine throughout the region.

METHODS

Literature Review

To identify previously published rubella seroprevalence surveys in Africa, we conducted a Medline search using the National Library of Medicines PubMed online search engine. Keywords included rubella combined with Africa. The starting year was not defined, the end date was June 2010, and no language priority was chosen. A related article hyperlink was followed for each retrieved article. The reference list of the retrieved articles was used to identify related literature. We included a study in the summary table if the article reported seroprevalence results and contained a description of study design, study population, and age group(s) tested. If the original article did not report a confidence interval for prevalence, we calculated a Wald confidence interval for a binomial proportion.

Surveillance Data Analysis

We analyzed regional measles case-based surveillance data collected and reported by 40 African countries during 2002–2009. We included data from each country in the analysis, starting in the year following completion of the nationwide catch-up measles supplementary immunization activity and continuing through all subsequent years. The data were collected following the WHO guidelines [14] using a standard case definition for a suspected measles case, which is illness with a generalized maculopapular rash and fever, and at least 1 of the following: cough, coryza (runny nose), or conjunctivitis [14]. For each case, surveillance officers collected detailed information including case-patient data on age, sex, address, and number of doses of measles vaccine received. Blood specimens were collected from persons with suspected measles cases identified within 30 days of rash onset and tested for measles-specific immunoglobulin M (IgM) antibody using standard enzyme-linked immunosorbent assay [15]. If results were negative or indeterminate for measles IgM, sera were subsequently tested for presence of rubella-specific IgM antibody using standard enzyme-linked immunosorbent assay. Laboratory testing was performed at an accredited national measles laboratory with support from WHO AFRO [14].

We included any country with ≥30 laboratory-confirmed rubella cases reported during 2002–2009 in our analysis of age, sex, and location of cases by country. We divided the countries included in the analysis into 4 sub-Saharan African subregions based on geography: West (Benin, Burkina Faso, Côte dIvoire, Gambia, Ghana, Guinea, Liberia, Mali, Mauritania, Nigeria, Niger, Senegal, Sierra Leone, and Togo), Central (Angola, Cameroon, Central African Republic, Chad, Congo, Democratic Republic of the Congo, Equatorial Guinea, and Gabon), East (Burundi, Eritrea, Ethiopia, Kenya, Rwanda, Tanzania, and Uganda), and South (Botswana, Lesotho, Madagascar, Malawi, Mozambique, Namibia, South Africa, Swaziland, Zambia, and Zimbabwe). To graphically explore the temporal trends of confirmed rubella cases in each geographical subregion, we overlay a smoothed curve over a histogram of cases counts for each month from year 2002 to 2009. We used cumulative percent curves to graphically describe the age distribution of laboratory-confirmed rubella cases. We categorized age into 5 groups (<1 y, 1–4 y, 5–9 y, 10–14 y, and ≥15 y), and for analysis of the surveillance data, reproductive age was defined as 15–49 years. We log-transformed age data and used a 2-sample t test to compare the means of rural versus urban settings. The test accounted for the potential cluster effect of country with standard errors estimated via the Taylor-series method. We analyzed data using SAS version 9.2 and R version 2.11.0 (Free Software Foundation, GNU project).

Vaccination Coverage Review

National administrative vaccination coverage is reported annually by each country, using the WHO and UNICEF Joint Reporting Form [16]. WHO and UNICEF, in collaboration with each national Ministry of Health, generate vaccination coverage estimates based on the reported national administrative coverage, available surveys, and results from independent coverage verification [17].

RESULTS

Literature Review

The search of published literature found 22 reports of rubella seroprevalence surveys during 1963–2009 from 14 (30%) of the 46 countries in the African region. Rubella seropositivity ranged from 68%–98%, and varied by age group (Table 1). Of the 22 studies, 17 reported seropositivity results for group aged >12 years or described as “women of reproductive age”. Of these 17 studies, the rubella seropositivity ranged from 71%–99%, and varied by study group and age group (Table 1). In 3 studies of rubella seropositivity among women of reproductive age with a study population >1000, rubella seropositivity was found to be 94.1% (95% confidence interval [CI], 93.0%–95.2%) in women 14–18 years of age (N = 1696) in Ethiopia, 90.1% (95% CI, 89.2–91.1) in women 15–45 years of age (N = 3471) in Senegal, and 84% (95% CI, 83.0%–85.0%) in women 15–34 years of age (N = 4866) in Côte dIvoire.

Table 1.

Summary of Published Rubella Sero-Survey Study Results by Country, World Health Organization African Region, 1963-2009

Surveillance Data Analysis

The number of countries included in the analysis of the case-based surveillance data was 10 from 2002, 15 from 2003, 26 from 2004, 31 from 2005, 37 from 2006, 39 from 2007, 40 from 2008, and 40 from 2009. During 2002–2009, 180,284 suspected measles cases were reported; of these, 105,625 had a specimen sent for laboratory testing to detect rubella-specific IgM antibody (Table 2). Among those tested, 25,631 cases (24%) were confirmed as rubella, 70,218 (66%) had negative test results, 4677 (4%) had indeterminate test results, and 5099 (5%) were unknown with pending results (Table 2).

Table 2.

Laboratory Results for Rubella-Specific Immunoglobulin M Antibody Testing of Suspected Measles Cases by Year, World Health Organization African Region, 2002–2009

Of the 25,631 confirmed rubella cases-patients, 25,097 (98%) had information on age; 3% were aged <1 year, 28% were aged 1–4 years, 47% were aged 5–9 years, 16% were aged 10–14 years, and 5% were 15 years of age or older (Table 3). The mean age of persons with rubella virus infection was 7.3 years (interquartile range [IQR] = 4.2–9.0 y) (Figure 1). Overall, 12,271 (50%) of confirmed rubella cases were among females, and 1293 (5%) were among women of reproductive age (15–49 y). Information on household location (rural vs urban) was available for 16,782 (65%) case-patients; of these, 10,527 (63%) were from rural settings and 6255 (37%) from urban settings (Table 3). Information on age and household location was available for 16,627 (65%) confirmed cases; in urban settings, mean age was 6.8 years (IQR = 3.9–8.5 y) and in rural settings, mean age was 7.5 years (IQR = 4.2–9.8 y) (P = .004).

Table 3.

Laboratory-Confirmed Rubella Cases by Sex, Setting, Age Group, World Health Organization African Region, 2002-2009 (N=25,631)

Figure 1.

Frequency of laboratory confirmed rubella cases (as reported by countries using measles case-based surveillance) by age in years with a cumulative age distribution curve, 2002–2009, World Health Organization African Region, N = 25,097.

Figure 2.

Laboratory confirmed rubella cases (as reported by countries using measles case-based surveillance) by month of rash onset in 4 geographic subregions, 2002–2009, World Health Organization African Region.

Note. Confirmed rubella cases as reported by countries using measles case based surveillance to the World Health Organization African Regional Office.

Yearly seasonality of reported laboratory-confirmed rubella cases by month of rash onset varied among the 4 subregions (Figure 2). In the West subregion, during 2003–2009, marked seasonality of rubella occurred each year with sharp increases in reporting during January with peaks in March–April followed by sharp declines in May, leading to troughs during October–December each year. In the Central subregion, data were sparse; however, peaks generally occurred during February–March and troughs were observed during September–November. In the East subregion, reporting varied by year; however, biphasic reporting was generally observed with peaks in March–April and in September–October and annual troughs in December–January and May–June. In the South subregion, a distinct annual seasonality was observed with consistently few cases reported during January–June each year, followed by gradual increases in June–July and peaks in September–October.

Vaccination Coverage Review

Of the 46 countries in the WHO African Region, 17 (37%) had estimated first-dose measles-containing vaccine coverage >80% in 2009 (Table 3). Of these, 2 (12%), Mauritius and Seychelles, have introduced rubella vaccine into the national routine childhood immunization schedule [18].

DISCUSSION

Rubella virus is circulating widely in Africa and primarily infects young children. The surveillance data suggest that by 15 years of age most children have developed immunity from natural infection. However, 5% of reported rubella cases occurred in women of reproductive age, suggesting that rubella infection during pregnancy, which can potentially lead to CRS, occurs throughout the region and remains largely undetected. Underreporting of milder rubella cases among adults is a potential reporting bias because the measles-rubella surveillance system is primarily focused on detection and reporting of suspected measles cases that generally occur among children. Point prevalence estimates from serological surveys throughout Africa estimated susceptibility to rubella virus among adults ranged from 1%–29% among the various study populations, whereas the 3 largest studies in Africa of women of reproductive age found that 6%–16% were susceptible to rubella virus infection.

This is the first report to review rubella epidemiology in the African region. Our findings are similar to those observed during the prevaccine era in other regions [3, 19, 20]. In Europe and the Americas, the age distribution of cases was similar to that in Africa, and rubella was primarily a childhood disease that occurred mainly among 5- to 9-year-olds [21]. In some countries in Africa, the proportion of cases was highest among children <5 years of age, suggesting the possibility of infection at a younger age; however, the numbers of cases were small and limitations of the surveillance data should be considered. In the United States, prior to the use of rubella vaccine, 80% of reported rubella cases occurred by 14 years of age and 92% by 20 years of age; however, seroprevalence studies estimated susceptibility to rubella among persons 17–22 years of age was 15%–20% [22]. Susceptibility to rubella among persons in rural settings was lower than in urban settings during the prevaccine era in Europe, the Americas, and Asia; and is consistent with our finding of an older mean age of reported cases in rural settings than in urban settings in Africa [21, 23]. The significantly younger mean age of reported cases in urban settings compared with that of rural settings in Africa may be due to rubella infection occurring at younger age in areas with high population density and contact rates. In the United States, annual seasonality of rubella was observed, with an increase in cases occurring in the early winter, peaking in March and decreasing to a low point in late summer and autumn; in general, annual seasonal peaks in rubella cases occur during springtime in temperate climates [21, 24]. We found similar seasonality in the South subregion of Africa, which includes countries located in the southern temperate zone.

CRS data from Africa are limited to descriptive reports of individual cases from Kenya, Uganda, and Senegal and case-series studies of clinically diagnosed CRS from Tanzania, Nigeria, Zimbabwe, and Ghana [3, 25, 26]. Of these small studies, 3 reported ≥18 clinically diagnosed CRS cases. In Ghana, 18 infants with clinically diagnosed CRS born within a 5-month period were identified following a large epidemic that included 40,276 suspected measles cases reported in 1995; however, no cases of rubella were reported [25, 27]. Assuming that the outbreak was caused by rubella instead of measles, the estimated congenital rubella syndrome rate was .8 cases per 1000 live births [25]. A case-series study from Harare, Zimbabwe, reported 18 clinically-diagnosed CRS cases following simultaneous epidemics of rubella and measles associated with an influx of refugees in the 1970s [28]. In Nigeria, a retrospective medical record review from 2000–2007 in a tertiary hospital identified 19 clinically diagnosed CRS cases [26].

These findings should be considered with knowledge of several limitations. Rubella cases were detected through a surveillance system designed to detect measles, and the clinical presentation of rubella may not meet the suspected measles case definition; 20%–50% of rubella infections do not include a rash [8]. Therefore, the descriptive analysis represents a small fraction of all rubella cases that occurred during 2002–2009, which likely biased the epidemiological findings toward groups wherein suspected measles cases occurred, such as in younger age groups because measles in Africa remains primarily a disease of persons aged <15 years.

The widespread circulation of rubella virus in Africa was confirmed through the measles case-based surveillance system and documented by several seroprevalence studies. Prevention of rubella through vaccination is largely unavailable in Africa despite the >80% routine measles vaccination coverage achieved by more than one-third of countries in the region [18]. As of October 2010, only 2 countries in the region have introduced rubella vaccine into their national immunization program.

A main concern when considering introducing rubella vaccination is that low vaccination coverage (typically described as <80%) may decrease virus circulation sufficiently to shift both the average age of exposure to rubella and rubella susceptibility from children to older age groups including women of childbearing age, and therefore may increase the prevalence of CRS. Such a shift in susceptibility to older age groups occurred and was documented in Greece and South Africa [10, 29]. As childhood rubella immunization programs mature, even with high coverage the percentage of remaining rubella cases shifts toward older age groups. However, the rubella susceptibility rate in older age groups, the absolute rubella risk, and the potential risk of CRS do not increase and may actually decrease. Several mathematical models have predicted that a certain minimum vaccination-coverage level may be important in determining the optimal strategy for the introduction of rubella vaccine; however, depending on the parameters used in these models, the minimum vaccination-coverage level can vary from 70%–85%.

In conclusion, we have documented that rubella virus circulates widely in Africa. Evidence of the CRS burden is limited to date; in 1996, it was estimated that 22,500 infants with CRS are born annually in the WHO African Region [30]. To fully understand the burden of rubella and CRS and to determine the most appropriate strategies for rubella control and CRS prevention, a comprehensive approach is needed. Following WHO recommendations, rubella surveillance should be integrated with measles case-based surveillance, CRS surveillance should be established, the routine immunization program should be strengthened, and the introduction of rubella vaccination should be carefully considered to ensure rubella immunity among women of reproductive age.

Funding

This work was supported by the World Health Organization (B. M., A. D., C. B., D. N.); and the Centers for Disease Control and Prevention (J. L. G., S. R., S. C.).

Acknowledgments

The authors would like to recognize the great efforts of all the surveillance officers and laboratory staff throughout the region.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Footnotes

• Potential conflicts of interest: none reported.

Supplement sponsorship: This article is part of a supplement entitled ”Global Progress Toward Measles Eradication and Prevention of Rubella and Congenital Rubella Syndrome,” which was sponsored by the Centers for Disease Control and Prevention.

• © The Author 2011. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com

References

1. 1.↵

World Health Organization. Rubella vaccinces: WHO position paper. Wkly Epidemiol Rec 2000;75:161-72.

1. 2.↵

World Health Organization. Preventing congenital rubella syndrome. Wkly Epidemiol Rec 2000;75:290-5.

1. 3.↵
2. 4.↵
3. 5.↵
4. 6.↵

Centers for Disease Control and Prevention. Progress toward elimination of rubella and congenital rubella syndrome—The Americas, 2003–2008. MMWR Morb Mortal Wkly Rep 2008;57:1176-9.

1. 7.↵
2. 8.↵

Centers for Disease Control and Prevention. Progress toward control of rubella and prevention of congenital rubella syndrome—Worldwide, 2009. MMWR Morb Mortal Wkly Rep 2010;59:1307-10.

1. 9.↵
2. 10.↵
3. 11.↵

World Health Organization. Accelerated rubella and congenital rubella syndrome programme, Costa Rica. Wkly Epidemiol Rec 2001;76:265-70.

1. 12.↵
2. 13.↵

Centers for Disease Control and Prevention. Progress toward measles control—African region, 2001–2008. MMWR Morb Mortal Wkly Rep 2009;58:1036-41.

1. 14.↵

World Health Organization Regional Office for Africa. Guidelines for measles surveillance. Brazzaville, Republic of Congo: World Health Organization regional Office for Africa; 2006.

1. 15.↵

Expanded Programme on Immunization (EPI) Team. Manual for the laboratory diagnosis of measles and rubella virus infection. Geneva, Switzerland: World Health Organization Dept of Immunization, Vaccines and Biologicals;; 2007. .

1. 16.↵