Elsevier

Vaccine

Volume 28, Issue 7, 17 February 2010, Pages 1677-1683
Vaccine

Review
Haemophilus influenzae type b disease in HIV-infected children: A review of the disease epidemiology and effectiveness of Hib conjugate vaccines

https://doi.org/10.1016/j.vaccine.2009.12.011Get rights and content

Abstract

The paper reviews the literature on the epidemiology of Hib disease and the effectiveness of Hib conjugate vaccine (HibCV) in HIV-infected children. The current three-dose primary Hib conjugate vaccine schedule in low-income settings has had a striking impact on the incidence of Hib disease. However, HIV-infected children have an almost 6-fold higher risk of Haemophilus influenzae type b (Hib) invasive disease than HIV-uninfected children and HibCV effectiveness is lower in this population. HIV-related HibCV failures are difficult to detect without well functioning surveillance systems and HIV testing of cases. Breakthrough Hib cases have been noted in vaccinated HIV-infected children in South Africa. A HibCV booster dose in addition to the three-dose primary schedule is routine in many, but not all, high-income countries. In order to determine whether a booster dose should be given to HIV-infected children in developing countries, well-designed studies need to be conducted to better determine the persistence of protective antibody concentrations, response to booster doses of vaccine as well as timing of and risk factors for vaccine failure in HIV-infected children both treated and naive to antiretroviral drug therapy (ART). Meanwhile, physicians and public health personnel should be especially vigilant at ensuring that HIV-infected infants receive their primary doses of HibCV, ART and co-trimoxazole prophylaxis. Until more definitive evidence is available, physicians may also need to consider a booster dose for such children irrespective of ART status. In any updating of vaccine schedules, HIV-infected children need particular consideration.

Introduction

Haemophilus influenzae type b (Hib), a Gram-negative encapsulated coccobacillus, is a leading cause of meningitis and a common cause of bacterial pneumonia in children under 5 years not vaccinated with Hib conjugate vaccine (HibCV). Invasive Hib disease is caused by invasion of the blood, lung parenchyma or meningeal space by the bacterium, which exists as a commensal of the upper respiratory tract in 3–11% of unvaccinated young children [1], [2]. Other clinical presentations of invasive Hib disease include septicaemia, cellulitis, osteomyelitis and epiglottitis. The World Health Organization (WHO) estimates that Hib disease causes 8.1 million episodes of serious illness and 371,000 childhood deaths annually, of which 8100 are among HIV-infected children [3].

There are an estimated 2.5 million HIV-infected children under 15 years of age and an additional 400,000 children newly infected with HIV each year, mostly in developing countries [4]. Without antiretroviral therapy, one in three HIV-infected children die before the age of 1 year from causes that include serious bacterial diseases from Hib [5]. High HIV prevalence countries are focusing on the provision of universal access to HIV testing, prevention of parent-to-child transmission and, more recently, antiretroviral therapy (ART). However, scale up of these activities has been slow [6]. Less than 10% of HIV-infected children requiring ART are estimated to receive these drugs [6]. Where ART is unavailable, co-trimoxazole prophylaxis has been found to reduce mortality by 43% and hospitalisation by 23% in HIV-infected children [7]; and much of this effect is attributed to prevention of serious bacterial infections. However, the need to attend for ongoing and frequent dosing with co-trimoxazole is a major limitation. Immunisation with bacterial conjugate vaccines, including HibCV, remains an important approach to reducing the risk of bacterial infections in HIV-infected children.

The highly effective HibCV [8] has been included in routine infant vaccination programmes since 1990 in many industrialised countries, and reductions in disease incidence have been striking. Recently, HibCV has been introduced into routine immunisation programmes in developing countries at an increasing rate [9], effectively reducing Hib disease in the general child population [10], [11], [12]. WHO currently recommends the administration of three doses of HibCV in the first year of life without a booster dose [13]. HibCV is commonly administered in a pentavalent formulation with diphtheria, tetanus, pertussis (DTP) and Hepatitis B in low-income countries, and the cost of the vaccine is co-financed by the GAVI Alliance, which does not provide funding for a booster dose.

All routine infant vaccines recommended by WHO are recommended in HIV-infected children with some modifications based on concerns regarding safety (e.g. the risk of disseminated BCG disease in HIV-infected children [14]), or to provide earlier protection (such as an extra measles dose at 6 months of age to account for lower levels of maternal antibodies in HIV-infected infants [15]). Effectiveness of all vaccines may be reduced because of immune suppression [16], and booster doses are often considered in these circumstances, guided by research findings or good practice. Little is known regarding the interaction between HIV infection and HibCV effectiveness.

Considering the importance of routine immunisations to prevent illness and death among HIV-infected children, particularly where access to ART is limited, we decided to examine whether specific recommendations for Hib vaccination in HIV-infected children are required. Developing or altering HibCV recommendations requires a clear understanding of the epidemiology of Hib disease in HIV-infected children, as well as the safety, immunogenicity and effectiveness of HibCV in this population. We review here the recent literature regarding the effect of HIV infection on the epidemiology of Hib disease and the effectiveness of Hib conjugate vaccine in HIV-infected children. The literature reviewed was based on a PubMed search from 1985 (when HIV was first identified) using the MeSH search terms “vaccines, conjugate” AND “HIV” supplemented by: screening of paper titles identified in a systematic review of the burden of disease from Hib and pneumococcal disease [3]; searches on specific topics such as effectiveness of ARTs on bacterial infections; recent relevant WHO and UNAIDs reports on paediatric HIV or vaccines and selected national government reports on guidelines for use of Hib vaccine.

Immune suppression associated with HIV infection increases the risk of serious bacterial infections, especially from encapsulated bacteria such as Hib and S. pneumoniae [17], [18]. In a South African hospital setting, HIV infection increased the risk of Hib-related bacteraemic pneumonia by more than 20 times, and the risk of S. pneumoniae by more than 40 times [19]. Both were also seen as the major causes of invasive disease in HIV-infected children in Zimbabwe [20] and Rwanda [21]. Bacterial disease is the most common presentation of acquired immunodeficiency syndrome (AIDS) in HIV-infected children not on ART [22].

Clinically, HIV-infected children are more likely to present with bacteraemic Hib pneumonia than Hib meningitis [19], [23]. Compared to HIV-uninfected children, the risk of developing Hib meningitis is only slightly higher in HIV-infected children. However, its severity is increased: severe neurological sequelae in survivors were seen in 5/7 (71%) HIV-infected children compared to 7/30 (33%) HIV-uninfected children [23]. HIV-infected children with Hib meningitis are also more likely to have concurrent pneumonia and local infections such as otitis media, mastoiditis or sinusitis together with malnutrition [23], [24]. In a Malawian hospital setting, presentation of any bacterial meningitis was also more likely to include shock if the child was HIV-infected [24].

The only well-defined data on the incidence of Hib disease in HIV-infected children in the absence of vaccination comes from a South African pneumococcal vaccine efficacy trial [25]. The risk of bacteraemic Hib pneumonia was higher in HIV-infected children as compared to uninfected children, with a relative risk of 18 (95% CI 6.9–47.1) in HIV-infected children less than 1 year [26], [19] and an overall 5.9-fold (95% CI 2.7–12.6) increased risk of invasive Hib disease [26]. The additional burden of Hib pneumonia attributable to HIV was 822 per 100,000 among HIV-infected under 1 year olds (870 vs. 48 cases per 100,000 were seen in HIV-infected compared to uninfected children) [19]. The risk of Hib meningitis with HIV infection was 1.74 (0.23–13.2) fold higher in under 1 year olds but with confidence intervals that included one [23].

In unvaccinated developing country populations, over 80% of childhood Hib cases occur before the age of 2 years [27], [11], [28]. In contrast, Hib disease tends to infect HIV-infected children at older ages. In a South African study, 5/19 HIV-infected children with bacteraemic Hib pneumonia were over 2 years of age, compared to none of the HIV-uninfected children [19]. Similar age patterns were observed for Hib meningitis (2/8 children vs. 1/36 children; respectively [23]); possibly because of continuing susceptibility resulting from progressive immunosuppression with age.

Few studies have examined the effect of ART on the incidence and case fatality of bacterial disease in HIV-infected populations in low-income settings [29]. Overall, the incidence of bacteraemia is decreased with ART, as seen for instance in a historical cohort study of ART use in Brazil when HibCV use was not routine [22]. Evaluation of the effect of ART on Hib disease is likely to remain limited given the small number of countries who are yet to adopt HibCV. In the US where HibCV was already included in routine immunisation, the introduction of ART reduced bacteraemia rates among HIV-infected children by 70% compared to the pre ART era but survival in those who had bacteraemia was still poorer than in those who had not developed bacteraemia [30] i.e. even in the presence of ART, risk of Hib-related mortality from bacteraemic diseases is increased by HIV infection. Thus, vaccinations including HibCV to prevent invasive bacterial infection remain important tools in the management of HIV-infected children.

Two studies on the effectiveness of HibCV in HIV-infected children are available in two different settings and one in all children with a moderately high prevalence of maternal HIV infection. The first, a study from South Africa, compared the rates of invasive Hib disease in an unvaccinated cohort born in 1997 to a pneumococcal conjugate vaccine trial study cohort enrolled between 1998 and 2000 who received HibCV. It documented moderately good vaccine effectiveness against Hib disease in HIV-infected populations (55% reduction of invasive Hib disease), but lower effectiveness than in HIV-uninfected children (91%; Table 1). However, the denominators for the HIV infected birth cohorts were imputed based on HIV prevalence in antenatal clinic attendees and the transmission rate from mother to child during that period [31]. Thus vaccine effectiveness may actually be higher in HIV-infected children if the denominator had been underestimated or if incidence of HIV had worsened concurrently with the start of the Hib vaccination programme.

Follow-up of the same cohort also indicated that HIV-infected children had a 29-fold (95% CI 9.9–84.5) higher risk of HibCV failure compared to uninfected children over 4 years of enhanced passive surveillance [31]. The average age of the 17 HIV-infected children with vaccine failure was older (median 9 months; range 2–49 months) than in the 7 uninfected children (median age 6 months; range 2–16). The time from vaccination to failure following the second or third dose, appeared to have a bimodal distribution with most failures (11/17) occurring in HIV-infected children under the age of 1 year, and then a second peak at ages ranging from 26 to 49 months [31]. Additional data are being collated from ongoing surveillance of Hib disease at the national level [12].

The second study examining vaccine effectiveness against clinical disease in HIV-infected children was a case–control study in Malawi. The number of cases was however too few to provide evidence of lower vaccine effectiveness in HIV-infected children (Table 1) [32]. Overall vaccine effectiveness was 88% (95% CI 63–96) despite a 14% prevalence of HIV among pregnant women. In the third study, another before/after study in Kenya, which had a 6.6% prevalence of HIV among pregnant women (and therefore probably <2% HIV infection prevalence in infants), vaccine effectiveness of 87% (95% CI 66–96) against invasive Hib disease was observed in all children under 2 years of age [11].

Other recent studies examining HibCV protection in low-income settings have reported HibCV failures; however, these reports originated from areas with lower HIV prevalence than in South Africa, and the limited data collected indicated few of the failures were related to HIV [33], [32], [11]. In settings, however, where survival among HIV-infected children is very poor, few children may reach the age at which Hib disease is most commonly seen.

Immunogenicity studies have demonstrated lower immune response to HibCV in HIV-infected children. Antibody levels to the capsular polysaccharide polyribosol-ribitol phosphate (PRP) of >1 μg/ml are used as a long term “correlate of protection” against invasive Hib disease [34]. Antibodies to PRP have, only recently, been examined in association with PRP conjugate HibCV in HIV-infected children. The presence of these antibodies in HIV-infected individuals should be interpreted with caution, as they may be functionally impaired [31] and evidence that their presence reflects long term clinical protection is limited. The two available studies have conflicting findings. In a small Spanish case study following the introduction of mass HibCV vaccination in 1996, all five HibCV failures (receiving two or more doses) had antibody levels below 0.17 μg/ml, apart from one HIV-infected child (who had anti-PRP antibodies of 2.29 μg/ml), whereas all three cases with non-typeable strains had levels that were considered protective [35]. Conversely, in a cross-sectional study of children in Soweto following three HibCV doses, a lower proportion with anti-PRP antibodies >1 μg/ml was seen in HIV-infected children compared to uninfected children (51.5% vs. 95.3%), measures which mirror the effectiveness observed against invasive disease in this setting [31].

Based on the results of immunogenicity studies, HIV-infected individuals generally have an attenuated response following a primary series of three doses (Table 2, Table 3). These studies have characteristics that limit their applicability to developing countries. They have been conducted mostly in high-income countries, include small sample sizes, evaluate different thresholds of immunogenicity and/or immunological parameters, use different conjugate vaccines, and slightly different timings of the primary doses. Despite these study limitations, they show that a lower proportion of HIV-infected children have >1 μg/ml anti-PRP antibodies after primary vaccination compared to HIV-uninfected children [36], [37], [31], [38], the exception being Read et al. [39] (Table 2). Children with more severe or symptomatic HIV disease were also less likely to respond [36], [40], [31]. There are limited data on antibody levels and CD 4 counts but no obvious relationship has been reported (Table 3).

More importantly, studies that examined the effect of a booster dose showed improvement in the proportion of HIV-infected children with antibodies >1 μg/ml even if it was at a lower level than that achieved in HIV-uninfected children [37], [40], [39]. A trial is currently underway in South Africa to determine the effect of a HibCV booster allocated randomly to HIV-infected children at 18 months that are treated with ARTs (S. Madhi, personal communication).

The extent to which immune reconstitution with ART prevents vaccine failure is an important research question that requires further investigation [41]. In a case series in Washington, USA, 18 children with a median age of 7 years and median times on ART of 20 months were periodically checked for antibody levels to administered vaccines. Four required re-vaccination with a single dose for Hib because of undetectable antibody levels (defined as <0.075 μg/ml), three developed detectable antibodies and, of the two retested, both had persistent Hib IgG levels at 1 year after vaccination [42].

Following the results of the trial currently underway in South Africa, more information will become available to assess the immunogenicity and quality of antibody response to HibCV in HIV-infected children randomised to early or late ART use [43].

HibCV is a non-replicating vaccine and therefore suggests minimal risk of complications in immunocompromised children [16]. However, limited data exist on HibCV safety specific to HIV-infected children. In the general population, HibCV is very well tolerated with few adverse events in trial data [8]. In an immunogenicity study of 67 HIV-infected and 57 uninfected children, only one HIV-infected child was reported to have had a local reaction. There were no other reactions [44]. There is a theoretical concern that HIV disease progression is accelerated due to transiently increased viraemia with T cell dependant vaccines such as HibCV, which stimulates and hence can deplete T cells. However, no prolonged effect is seen in practice [45].

Section snippets

Discussion and recommendations

HIV-infected children are at significantly higher risk for invasive Hib disease than uninfected children. HibCV is used routinely in most countries with high HIV prevalence, following the recent increase in HibCV adoption as a three-dose primary schedule in low-income countries. Public health personnel should exercise vigilance in ensuring that all children, especially those who are exposed to HIV, are vaccinated. Routine HibCV use markedly reduces the risk of invasive Hib disease in

Acknowledgement

Our thanks to Jessica Shearer for editorial assistance.

References (56)

  • K. Mulholland et al.

    Randomised trial of Haemophilus influenzae type-b tetanus protein conjugate vaccine for prevention of pneumonia and meningitis in Gambian infants

    Lancet

    (1997)
  • V.B. Peters et al.

    Immunity to Haemophilus influenzae type b polysaccharide capsule in children with human immunodeficiency virus infection immunized with a single dose of Haemophilus vaccine

    J Pediatr

    (1994)
  • K. Chokephaibulkit et al.

    Catch-up vaccination against Haemophilus influenzae type b in human immunodeficiency virus-infected Thai children older than 2 years old

    Vaccine

    (2004)
  • R.A. Adegbola et al.

    Vaccination with a Haemophilus influenzae type b conjugate vaccine reduces oropharyngeal carriage of H-influenzae type b among Gambian children

    J Infect Dis

    (1998)
  • A.K. Takala et al.

    Reduction of oropharyngeal carriage of Haemophilus-influenzae type-B (Hib) in children immunized with an Hib conjugate vaccine

    J Infect Dis

    (1991)
  • Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO)

    AIDS epidemic update

    (2007)
  • UNICEF

    Children and AIDs a stocktaking report

    (2007)
  • V. Mulenga et al.

    Effect of cotrimoxazole on causes of death, hospital admissions and antibiotic use in HIV-infected children

    AIDS

    (2007)
  • C.O. Obonyo et al.

    Efficacy of Haemophilus influenzae type b vaccination of children: a meta-analysis

    Eur J Clin Microbiol Infect Dis

    (2006)
  • Centers for Disease Control and Prevention

    Progress toward introduction of Haemophilus influenzae type b vaccine in low-income countries—worldwide, 2004–2007

    MMWR

    (2008)
  • K.D. Cowgill et al.

    Effectiveness of Haemophilus influenzae type b conjugate vaccine introduction into routine childhood immunization in Kenya

    JAMA

    (2006)
  • A. von Gottberg et al.

    Impact of conjugate Haemophilus influenzae type b (Hib) vaccine introduction in South Africa

    Bull World Health Organ

    (2006)
  • WHO position paper on Haemophilus influenzae type b conjugate vaccines (replaces WHO position paper on Hib vaccines previously published in the Weekly Epidemiological Record)

    Wkly Epidemiol Rec

    (2006)
  • WHO

    Revised BCG vaccination guidelines for infants at risk for HIV infection

    Wkly Epidemiol Rec

    (2007)
  • World Health Organization

    Measles vaccines: WHO position paper

    Bull World Health Organ

    (2004)
  • W.J. Moss et al.

    Immunization of children at risk of infection with human immunodeficiency virus

    Bull World Health Organ

    (2003)
  • S.A. Spector et al.

    A controlled trial of intravenous immune globulin for the prevention of serious bacterial infections in children receiving zidovudine for advanced human immunodeficiency virus infection. Pediatric AIDS Clinical Trials Group

    N Engl J Med

    (1994)
  • S.A. Madhi et al.

    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)
  • Cited by (0)

    View full text