Despite the success of Haemophilus influenzae type b (Hib) conjugate vaccines in reducing the burden of invasive Hib disease in developed countries, including a reduction in radiologically confirmed pneumonia by ∼20%, it is sparsely used in developing countries. 1–5 South Africa and The Gambia are the only African countries where Hib conjugate vaccine has been included in the national Expanded Program of Immunization (EPI). 5 This is mainly the consequence of the relatively high cost of the vaccine. Although Hib conjugate vaccines have been available since 1988, their effectiveness in countries where there is a high prevalence of pediatric HIV-1 infection remains unknown.
Reports regarding the immunogenicity of the Hib conjugate vaccines in HIV-1-infected children vary. Although the proportion of HIV-1-infected children producing anti-polyribosylribitol phosphate (PRP) serum antibody titers ≥0.15 μg/ml and ≥1.0 μg/ml was similar to that of HIV-1-uninfected children, most studies report HIV-1-infected children to have lower geometric mean Hib anti-PRP antibody titers than HIV-1-exposed uninfected children given the vaccine. 6–9 Further there appears to be little correlation between anti-PRP antibody response and clinical or immunologic staging and/or HIV-1 viral load in HIV-1-infected children. 7 The number of HIV-1-infected children achieving anti-PRP antibody titers ≥1.0 μg/ml, a correlate of long term immunity against invasive Hib disease, varies between 46 and 88% in these studies. 6–9 HIV-1-infected children are also less likely to have persistent antibody titers ≥1.0 μg/ml, 1 year after vaccination than HIV-1-uninfected children (57%vs. 89%, respectively). 6
A limitation of extrapolating these results to developing countries is that children from developed countries might have been receiving antiretroviral therapy and are scheduled for immunization with Hib conjugate vaccine at an older age than children from developing countries; both factors could possibly influence the immune response to vaccine. Antiretroviral therapy is unavailable in most developing countries where the burden of HIV-1 infection is greatest.
Because HIV-1-infected children are predisposed to developing bacterial infections and contribute significantly to the burden of invasive bacterial disease in areas with a high prevalence of pediatric HIV-1 infection, the effectiveness of bacterial vaccines in such areas would also depend on their success in preventing disease in these children. 10–13
Hib conjugate vaccine became available as part of the EPI in South Africa in July, 1999, but vaccination of infants in Soweto commenced in March, 1998, as part of a Phase III trial evaluating the efficacy of a nonavalent pneumococcal conjugate vaccine, which had earlier been shown to be immunogenic in this population. 14 This allowed us to evaluate the effectiveness of Hib conjugate vaccination in an area where the prevalence of HIV-1 in women attending antenatal clinics increased from 17.1% to 23.9% between 1997 and 1999. 15, 16
Study site and population
This study was conducted at Chris Hani-Baragwanath Hospital, a combined secondary and tertiary level hospital, which serves the population of Soweto, South Africa. Soweto is an urban black community with a population of 1.2 million, including 120 000 children <5 years of age. 17 It was estimated that 90% of hospitalizations involving children from the study area would occur at this hospital. The birth cohort at Chris Hani-Baragwanath Hospital and the associated midwife obstetric units in 1997 was 22 000 (Department of Obstetrics and Gynecology, Chris Hani-Baragwanath Hospital, unpublished data). The HIV-1 prevalence amongst women attending antenatal clinics in 1997 was 17.1% [95% confidence interval (95% CI) 15.1 to 19.2%] and this increased to 22.5% (95% CI 19.2 to 25.7) and 23.9% (95% CI 21.7 to 26.0) in 1998 and 1999, respectively. 15, 16 Antiretroviral therapy was not available in the public sector for either the prevention of mother-to-child vertical transmission of HIV-1 or for the management of HIV-1-infected children. In the absence of any active intervention, it was estimated that the mother-to-child vertical transmission rate of HIV-1 was 26% for Hib-vaccinated and -unvaccinated children. 18
Thus ∼4.45% of all the children born in Soweto in 1997 and ∼6.03% of children born between March, 1998, and June, 1999, were estimated to be HIV-1-infected. Infants are vaccinated at birth and 6, 10 and 14 weeks of age, and the immunization coverage for the three doses of diphtheria-tetanus-pertussis vaccine, trivalent oral polio vaccine and hepatitis B vaccine in this area was 93.8% in 1996. 19
Based on a neonatal mortality rate of 16.4 per 1000, it is estimated that 21 639 children, including an estimated 963 (4.45%) HIV-1-infected children born in 1997, would have received at least 1 dose of vaccine. 19, 20 Of these an estimated 19 394 HIV-1-uninfected children and 903 HIV-1-infected children would have received all 3 doses of vaccine. Children born in 1997 did not receive Hib conjugate vaccine.
In March, 1998, a Phase III trial evaluating the efficacy of a nonavalent pneumococcal conjugate vaccine was started in the study area. The vaccinated cohort included 19 267 children who were recruited between March, 1998, and June, 1999, into the trial. These children received Hib conjugate vaccine (PRP-CRM197-diphtheria-tetanus toxoids-whole cell pertussis; TETRAMUNE; Wyeth Lederle Vaccines and Pediatrics) and either placebo or a nonavalent pneumococcal conjugate vaccine (Wyeth Lederle Vaccines and Pediatrics). The immunization coverage rate for all 3 doses of Hib vaccine in this study was 98% (KP Klugman, unpublished data). The prevalence of HIV-1 in this cohort was estimated to be 6.03% (1162 children would have been HIV-1-infected).
Daily laboratory surveillance of culture-confirmed invasive Hib disease was undertaken from January, 1997. There was active case detection during prospective studies evaluating invasive bacterial disease in children between March, 1997, and September, 2000. 11, 12
Vaccine effectiveness was assessed by comparing the burden of invasive Hib disease in the Hib-unvaccinated cohort compared with the vaccinated cohort of children. Surveillance for invasive Hib disease in the latter cohort was identical with that of the unvaccinated cohort. All cases of invasive Hib disease occurring in either of these cohorts until September, 2000, were included in this report.
Ethical permission for the studies evaluating invasive bacterial disease and the Phase III pneumococcal conjugate vaccine efficacy trial were independently obtained from the Committee for Research on Human Subjects (Medical) at the University of the Witwatersrand and informed consent was obtained from the parent/s of the child.
Sample size calculations and data were analyzed using the Epi-Info Version 6.04c statistical package. 21 The number of HIV-1-uninfected children present in the two cohorts would have been adequate to identify vaccine effectiveness of >95% with 80% power and 95% confidence. The number of HIV-1-infected children in each cohort studied would have been adequate to detect vaccine effectiveness of 95% with 80% power and 90% confidence, based on 1.0% of the Hib-unvaccinated children developing invasive Hib disease before 1 year of age.
The burden of disease was calculated based on the estimated incidence rates of invasive Hib disease in the unvaccinated cohort. The at risk group cohort for the HIV-1-infected and -uninfected children was calculated based on the mean HIV-1 prevalence rates amongst women attending antenatal clinics at the given period and assuming a vertical transmission rate of 26% for both cohorts of children. 18
The estimated burden of disease and relative risks (95% CI) of culture-confirmed severe invasive Hib disease for the children born in 1997 was adjusted for estimated infant mortality rates of 32.5 per 1000 and 350 per 1000 in HIV-1-uninfected and -infected children, respectively. 20, 22
Continuous variables were analyzed with the Student t test for normally distributed data and equal group variances and the Kruskal-Wallis H test for nonparametric data. Categoric variables were analyzed with the Yates corrected test or Fisher’s exact test when the expected cell value was <5. An alpha value of ≤0.05 was considered significant. The effectiveness of Hib conjugate vaccine in HIV-1-infected and -uninfected children was analyzed on an intent-to-treat basis. All children who received at least one dose of vaccine, i.e. 6 weeks or older in the unvaccinated cohort, or had received at least one dose in the vaccinated cohort and included only children <1 year of age. The same analysis was performed for children who were considered fully vaccinated. In the unvaccinated cohort this included all cases detected in children between 4.1 and 12.0 months of age; in the vaccinated cohort it included all cases occurring at least 14 days after having received the third dose of Hib conjugate vaccine. The estimated vaccine effectiveness was calculated by using the vaccine efficacy calculation function for cohort studies in Epi-Info Version 6.04c.
HIV-1 status in the unvaccinated cohort of children was performed according to routine departmental clinical protocol with a third generation HIV enzyme-linked immunosorbent assay (ELISA; Axsym system, HIV 1/2; Abbott Diagnostics, Wiesbaden, Germany), and positive results were confirmed using the Wellcozyme HIV 1+2 enzyme immunoassay (Murex Diagnostics, Kent, Dartford, UK). The HIV status of children <15 months of age who had fewer than 3 signs of CDC Category A AIDS were confirmed using an HIV DNA PCR (AMPLICOR HIV-1 test, Version 1.5; Roche Diagnostics Systems, Nutley, NJ, or HIV-1 RNA PCR (HIV-1 MONITOR Test, Version 1.5; Roche). Clinical categorization of the HIV status was based on the CDC classification for HIV infection in children. 23
The HIV-1 status of children in the vaccinated cohort was assessed by screening with an HIV ELISA (Axsym system, HIV 1/2) and confirmed as with the unvaccinated cohort where the ELISA was positive or if there was any clinical suspicion of HIV infection.
Culture methods and serotyping
Blood was cultured for bacterial growth using the BacT/Alert microbial detection system (Organon Teknika, Durham, NC). Routine methods were used for culture and identification of H. influenzae type b from blood and cerebrospinal fluid (CSF) samples. The capsular type of H. influenzae was identified using specific agglutinating sera (Murex Biotech Ltd., Kent, Dartford, UK).
H. influenzae type b antigen detection was performed on cerebrospinal fluid using the Wellcogen bacterial antigen kit (Murex, Wiesbaden, Germany).
Hib polyribosylribitol phosphate IgG antibody assays
Serum IgG antibody assays to Hib PRP were measured by an enzyme-linked immunosorbent assay in vaccinated children who developed invasive disease. These specimens were obtained at the time of the acute illness and centrifuged and stored at −70°C until processed at the South African Institute for Medical Research laboratory.
The following case definitions were used for the clinical categorization of children: (1) Hib meningitis was diagnosed in children from whom Hib was isolated from the CSF or in a child with >5 neutrophils/mm3 in the CSF with clinical features of meningitis who had a reactive CSF Hib latex test or in whom Hib was isolated from blood; (2) bacteremic Hib pneumonia was diagnosed in children with a clinical and radiologic diagnosis of pneumonia in whom Hib was isolated from blood; (3) Hib-associated septic shock without a focus was diagnosed in children with hemodynamic instability (hypotension and metabolic acidosis) in the absence of any focus of infection in whom Hib had been isolated from blood; and (4) other focal Hib disease was diagnosed where Hib was isolated from a normally sterile site in a child with evidence of focal disease other than that described above.
Epidemiology of invasive Hib disease in the unvaccinated cohort of
children. Fifty-seven children with invasive Hib disease were identified in children in 1997. Fifty-two of these children were tested for HIV-1; the prevalence of HIV-1 infection was 21.2%. Among the five children who were not tested for HIV-1 infection, one child had signs suggestive of AIDS (male, 4.2 months old, CDC Clinical Category B) and another 10-day-old infant had a reactive HIV ELISA test but no HIV-1 PCR result. Both children had bronchopneumonia and died. The remaining three children who were not tested for HIV-1 infection were asymptomatic for HIV-1 disease (two boys and one girl ages 24.8, 25.5 and 23.5 months, respectively) and included one child each with meningitis, bacteremic pneumonia and epiglottitis. These five children in whom the HIV-1 status was unknown were excluded from further analysis.
H. influenzae type b was isolated from blood in 10 HIV-1-infected and 22 uninfected children. In addition Hib was isolated from the CSF in 25 HIV-1-uninfected children (including 1 child with a positive latex test but no growth of Hib), 5 of whom also had a positive blood culture.
The demographic characteristics and clinical diagnosis of the children with invasive Hib disease are shown in Table 1. The spectrum of clinical presentation differed between HIV-1-infected and -uninfected children, with bacteremic pneumonia being more common in HIV-1-infected children and meningitis being more common in HIV-1-uninfected children (P = 0.0012) There were, however, no differences in overall mortality, bacteremic pneumonia-specific mortality or the proportion of beta-lactamase-producing Hib isolates between HIV-1-infected and -uninfected children (Table 1).
The estimated relative annual incidence rates (per 100 000 children) of overall invasive Hib disease in children <1 year of age was greater in HIV-1-infected (1003) than in uninfected children (170; relative risk, 5.9; 95% CI 2.7 to 12.6;P < 0.0002). Further HIV-1-infected children <1 year of age were also at greater risk of developing Hib bacteremic pneumonia (870 per 100 000 per year) than were HIV-1-uninfected children (48 per 100 000 per year; relative risk, 18.0; 95% CI 6.9 to 47.1;P < 0.00001). There was, however, no difference in the relative risk of developing Hib meningitis between HIV-1-infected and -uninfected children <1 year of age (estimated incidence rates per 100 000 per year: 0 vs
106, respectively;P > 0.99).
Invasive Hib disease in children vaccinated with Hib conjugate
vaccine. Nine children of the vaccinated cohort developed culture-confirmed invasive Hib disease, eight of them <1 year of age. Eight isolates were cultured from blood and two from the cerebrospinal fluid, including one child with a positive blood culture. Of these children seven were HIV-1-infected and two were uninfected. Five of the seven HIV-1-infected children presented with pneumonia, and most of these children had high HIV-1 viral loads at the time of developing their illness (Table 2). Both of the HIV-1-uninfected children had other underlying predisposing factors for developing invasive Hib disease; in addition one child had received only one dose of Hib conjugate vaccine (Table 2). All three of the fully vaccinated HIV-1-infected children in whom serum was available for analysis had anti-PRP antibody titers ≥0.15 μg/ml but <1 μg/ml. The antibody titer in the fully vaccinated HIV-1 uninfected child was 11.53 μg/ml (Table 2). Four of the five fully vaccinated HIV-1-infected children developed invasive disease at least 16 weeks after having received their third dose of vaccine.
Table 3 shows the effectiveness of the vaccine in fully vaccinated and partially vaccinated HIV-1-infected and -uninfected children. The vaccine was less effective in fully vaccinated HIV-1-infected (43.9%; 95% CI −76.1 to 82.1) than in HIV-1-uninfected children (96.5%; 95% CI 74.4 to 99.5;P < 0.00001).
This is the first effectiveness study of this particular Hib conjugate vaccine in a developing country and the first effectiveness study of Hib conjugate vaccines in children with a high prevalence of HIV-1 infection. The overall effectiveness of this Hib conjugate vaccine was high (83%), and the effectiveness of the vaccine in preventing invasive Hib disease, including meningitis and bacteremic pneumonia in HIV-1-uninfected children in this study mirrored that observed in developed countries. 3–5 Aside from two vaccine failures, both of which occurred in children with other predisposing causes for immunosuppression and one of which occurred only 15 days after the child received his first dose of Hib conjugate vaccine, the vaccine was estimated to be 96.5% effective in fully vaccinated HIV-1-uninfected children. The overall estimated effectiveness of the vaccine was, however, lower than that observed in developed countries mainly because of the reduced effectiveness of the vaccine in HIV-1-infected children (44%). Unlike the observation in developed countries where primary immunodeficiencies and other non-HIV underlying illness are the most common cause for Hib conjugate vaccine failure, 24, 25 HIV-1 infection is the most common cause of vaccine failures in this area with a high prevalence of pediatric HIV-1 infection.
The increased burden of invasive Hib disease in HIV-1-infected children noted in the unvaccinated cohort is understandable because of the impairment of humoral and cell-mediated immunity in these children. 26, 27 The reason for the increase in bacteremic Hib pneumonia but not in Hib meningitis in HIV-1-infected children is unclear. If the pathogenesis of meningitis included a transient bacteremic phase, HIV-1-infected children may be unable to contain this initial bacteremia, and they may present for care before the localization and multiplication of organisms in the CSF. HIV-1-infected children with Hib pneumonia may be at greater risk of being bacteremic as was observed in HIV-1-infected adults with pneumococcal pneumonia (60 to 80%) compared with HIV-1-uninfected adults (10%). 28
All three fully vaccinated HIV-1-infected children in whom anti-PRP antibody titers were performed had titers ≥0.15 μg/ml at the time of their illness. This antibody concentration has been suggested to correlate with short term immunity against Hib invasion. 29 The development of invasive Hib disease despite antibody concentrations ≥0.15 μg/ml in HIV-1-infected children may be because of their having had low antibody titers (<0.15 μg/ml) before being infected by Hib, with the resulting measured antibody concentrations being a consequence of an acute booster response after Hib infection. Alternatively the functionality of the produced antibody in HIV-1-infected children may have been impaired, resulting in invasive disease despite antibody concentrations greater than the surrogate level associated with short term protection. 29 Our data therefore support the need to perform functional antibody assays in HIV-1-infected children who are recognized to have impaired cell-mediated and B lymphocyte activity. 27, 30 Concurrent administration of the Hib conjugate vaccine with the nonavalent pneumococcal conjugate vaccine in some of the vaccinated children is unlikely to have contributed to the vaccine failures, because simultaneous administration of these two vaccines has been shown to be associated with a greater antibody response to Hib-PRP than when the Hib conjugate vaccine was administered with placebo (geometric mean antibody titers of 11.62 vs
4.58, respectively). 14
This analysis of the effectiveness of Hib conjugate vaccine in an area with a high prevalence of HIV-1 infection has a number of limitations. Because Hib conjugate vaccination was already recommended for HIV-1-infected children, it was not ethically feasible to conduct a randomized placebo-controlled trial to evaluate the efficacy of this vaccine in HIV-1-infected children, hence the current study design. In addition to being an open label trial evaluating two cohorts at different time periods, the prevalence of HIV-1 infection in the cohorts was based on estimates derived from the prevalence of HIV-1 infection among women attending antenatal clinics, coupled with an estimated vertical transmission rate and suboptimal HIV-1 categorization in the unvaccinated cohort of children. The latter would have resulted in an underestimation of the disease burden in the unvaccinated HIV-1-infected cohort and consequently overestimated vaccine effectiveness amongst HIV-1-infected children. A further limitation was that the sample size for the HIV-1-infected children was inadequate to detect any vaccine effectiveness of <90%. Despite these limitations the results clearly reflect the reduced effectiveness of Hib conjugate vaccine in HIV-1-infected than in HIV-1-uninfected children.
The period after vaccination when disease occurred in fully vaccinated HIV-1-infected children is earlier than that observed in the UK. In South Africa as in the UK, no booster dose is given after the primary immunization series. 25 Even if a booster dose had been planned, this is unlikely to have had any impact on the outcomes because all the failures in fully vaccinated HIV-1-infected children occurred before a booster dose would have been scheduled. The occurrence of invasive Hib disease in vaccinated HIV-1-infected children also suggests that the vaccine failed to prevent nasopharyngeal colonization in these children that was probably related to an inadequate antibody response to vaccine in these children. 6–9, 31 Consequently HIV-1-infected children may remain potential sources for the transmission of Hib to unvaccinated children, and the herd immunity benefits of this vaccine may not be realized in communities where there is a high prevalence of pediatric HIV-1 infection.
The results of this trial highlight the importance of evaluating both the immunogenicity and effectiveness of the newer generation of vaccines, including pneumococcal conjugate vaccines, in HIV-1-infected children, particularly in areas where the prevalence of pediatric HIV-1 infection is high. Finally despite the fact that this Hib conjugate vaccine is less effective in HIV-1-infected children, the benefits of using the Hib conjugate vaccine in areas with a high prevalence of pediatric HIV-1 infection, as in much of sub-Saharan Africa, are confirmed by its effectiveness in reducing the overall burden of invasive Hib disease.
We thank Helena Käyhty, KTL-Finland, for collaborating in the setting up of the anti-PRP antibody assays, and the staff of the Department of Paediatrics, Chris Hani-Baragwanath Hospital for caring for the children and allowing the study to take place. The study was supported through funding by Wyeth-Lederle Vaccines and Pediatrics.
1. Lagos R, Horwitz I, Toro J, et al. Large scale, post licensure, selective vaccination of Chilean infants with PRP-T conjugate vaccine: practicality and effectiveness in preventing invasive Haemophilus influenzae
type b infections. Pediatr Infect Dis J 1996; 15: 216–22.
2. Mulholland K, Hilton S, Adegbola R, et al. A randomised trial of Haemophilus influenzae
type b tetanus protein conjugate for the prevention of pneumonia and meningitis in Gambian infants. Lancet 1997; 349: 1191–7.
3. Anonymous. Progress toward eliminating Haemophilus influenzae
type b disease among infants and children: United States, 1987–1997. MMWR 1998; 47: 993–8.
4. Shinefield HR, Black S. Postlicensure surveillance for Haemophilus influenzae
type b invasive disease after use of Haemophilus influenzae
type b oligosaccharide CRM197 conjugate vaccine in a large defined United States population: a four-year eight-month follow-up. Pediatr Infect Dis J 1995; 14: 978–81.
5. Peltola H. Worldwide Haemophilus influenzae
type b disease at the beginning of the 21st century: global analysis of the disease burden 25 years after the use of the polysaccharide vaccine and a decade after the advent of conjugates. Clin Microbiol Rev 2000; 13: 302–17.
6. Gibb D, Giacomelli A, Masters J, et al. Persistence of antibody responses to Haemophilus influenzae
type b polysaccharide conjugate vaccine in children with vertically acquired human immunodeficiency virus infection. Pediatr Infect Dis J 1996; 15: 1097–101.
7. Read JS, Frasch CE, Rich K, et al. The immunogenicity of Haemophilus influenzae
type b conjugate vaccines
in children born to human immunodeficiency virus-infected women: Women and Infants Transmission Study Group. Pediatr Infect Dis J 1998; 17: 391–7.
8. Rutstein RM, Rudy BJ, Cnaan A. Response of human immunodeficiency virus-exposed and -infected infants to Haemophilus influenzae
type b conjugate vaccine. Arch Pediatr Adolesc Med 1996; 150: 838–41.
9. Kale KL, King JC Jr., Farley JJ, Vink PE, Cimino CO, Paradiso PR. The immunogenicity of Haemophilus influenzae
type b conjugate (HbOC) vaccine in human immunodeficiency virus-infected and uninfected infants. Pediatr Infect Dis J 1995; 14: 350–4.
10. Wilfert CM. Invasive bacterial infections in children with HIV infection. In: Pizzo PA, Wilfert CM, eds. Pediatric AIDS: the challenge of HIV infection in infants, children, and adolescents. 3rd ed. Baltimore: Williams & Wilkins, 1998: 117–25.
11. Madhi SA, Petersen K, Madhi A, Khoosal M, Klugman KP. Increased disease burden and antibiotic resistance of bacteria causing severe community-acquired lower respiratory tract infections in human immunodeficiency virus type 1-infected children. Clin Infect Dis 2000; 31: 170–6.
12. Madhi SA, Madhi A, Peterson K, Khoosal M, Klugman KP. The impact of human immunodeficiency virus type-1 infection on the epidemiology and outcome of bacterial meningitis in South African children. Int J Infect Dis 2001; 5: 119–25.
13. Madhi SA, Petersen K, Madhi A, Wasas A, Klugman KP. Impact of human immunodeficiency virus type-1 on the disease spectrum of Streptococcus pneumoniae
in South African children. Pediatr Infect Dis J 2000; 19: 1141–7.
14. Mbelle N, Huebner RE, Wasas AD, Kimura A, Chang I, Klugman KP. Immunogenicity and impact on nasopharyngeal carriage of a nonavalent pneumococcal conjugate vaccine. J Infect Dis 1999; 180: 1171–6.
15. Health Systems Research and Epidemiology. Summary report: 1998 National HIV sero-prevalence survey of women attending public antenatal clinics in South Africa. Department of Health, Pretoria, South Africa, February, 1999.
16. Health Systems Research and Epidemiology. Summary report: National HIV sero-prevalence survey of women attending public antenatal clinics in South Africa. Department of Health, Pretoria, South Africa, April, 2000.
17. Daponte BO. Results from the 1995 Soweto population survey: estimates of the population of towns in Soweto. Report to the Greater Johannesburg Metropolitan Council. Human Rights Institute of South Africa, Pretoria, South Africa, July, 1995.
18. Gray GE, McIntyre JA, Lyons SF. Effect of breastfeeding on the vertical transmission of HIV-1 in Soweto, South Africa [Abstract Th. C.413]. In: Proceedings of the XIth International Conference on AIDS, Vancouver, July 12 to 17, 1997.
19. Ndaye K. Immunization coverage and factors affecting acceptance of childhood immunization in Zola and Chiawelo township: The Greater Soweto Maternal and Child Health Project. (Thesis). Johannesburg: University of the Witwatersrand, 1997.
20. Department of Health, Medical Research Council, Macro International Inc. South Africa demographic and health survey 1998: preliminary report. Department of Health, Pretoria, South Africa, August, 1999:7.
21. Epi-Info Version 6.04c statistical package. Atlanta, GA: Centers for Disease Control and Prevention, 1997.
22. Friedland IR, Klugman KP, Karstaedt AS, Patel J, McIntyre JA, Allwood CW. AIDS- the Baragwanath experience, Part 1: epidemiology of HIV infection at Baragwanath Hospital; 1988–1990. S Afr Med J 1991; 82: 86–90.
23. Centers for Disease Control and Prevention. 1994 Revised classification system for human immunodeficiency virus infection in children <13 years of age. MMWR 1994; 43: 1–19.
24. Holmes SJ, Lucas AH, Osterholm MT, Froeschle JE, Granoff DM. Immunoglobulin deficiency and idiotype expression in children developing Haemophilus influenzae
type b disease after vaccination with conjugate vaccine: The Collaborative Study Group. JAMA 1991; 266: 1960–5.
25. Heath PT, Booy R, Griffiths, et al. Clinical and immunological risk factors associated with Haemophilus influenzae
type b conjugate vaccine failure in childhood. Clin Infect Dis 2000; 31: 973–80.
26. Lane HC, Depper JM, Greene WC, Whalen G, Waldmann TA, Fauci AS. Qualitative analysis of immune function in patients with the acquired immunodeficiency syndrome. N Engl J Med 1985; 313: 79–84.
27. Pahwa S, Fikrig S, Menez R, Pahwa R. Pediatric acquired immunodeficiency syndrome: demonstration of B lymphocyte defects in vitro
. Diagn Immunol 1986; 4: 24–30.
28. Janoff EN, Rubins JB. Invasive pneumococcal disease in the immunocompromised host. In: Tomasz A, ed. Streptococcus pneumoniae
: molecular biology and mechanisms of disease. 1st ed. Larchmont, NY: Mary Ann Liebert, 2000: 321–41.
29. Käythy H, Peltola H, Karanko V, Mäkelä; PH. The protective level of serum antibodies to the capsular polysaccharide of Haemophilus influenzae
type b. J Infect Dis 1983; 147: 1100.
30. Schlesinger Y, Granoff DM and the Vaccine Study Group. Avidity and bactericidal activity of antibody elicited by different Haemophilus influenzae
type b conjugate vaccines
. JAMA 1992; 267: 1489–94.
31. Fernandez J, Levine OS, Sanchez J, et al. Prevention of Haemophilus influenzae
type b colonization by vaccination: correlation with serum anti-capsular IgG concentration. J Infect Dis 2000; 182: 1553–6.