- Email: [email protected]
Long-term complications and risk of other serious infections following invasive Haemophilus influenzae serotype b disease in vaccinated children
Vaccine 28 (2010) 2195–2200
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Long-term complications and risk of other serious infections following invasive Haemophilus influenzae serotype b disease in vaccinated children Shamez Ladhani a,∗ , Paul T. Heath b , Rashna J. Aibara c , Mary E. Ramsay d , Mary P.E. Slack d , Martin L. Hibberd e , Andrew J. Pollard f , E. Richard Moxon f , Robert Booy a,g a
Academic Unit of Paediatrics, Barts and The London School of Medicine and Dentistry, London, United Kingdom Vaccine Institute and Division of Child Health, St George’s, University of London, United Kingdom Department of Paediatrics, Central Middlesex Hospital, London, United Kingdom d Centre for Infections, Health Protection Agency, London, United Kingdom e Infectious Diseases Unit, Genome Institute of Singapore, Republic of Singapore f Oxford Vaccine Group, Department of Paediatrics, University of Oxford, United Kingdom g National Centre for Immunisation Research & Surveillance, University of Sydney, Australia b c
a r t i c l e
i n f o
Article history: Received 27 July 2009 Received in revised form 10 December 2009 Accepted 23 December 2009 Available online 5 January 2010 Keywords: Conjugate vaccine Meningitis Sequelae
a b s t r a c t This study describes the long-term complications in children with Haemophilus influenzae serotype b (Hib) vaccine failure and to determine their risk of other serious infections. The families of 323 children with invasive Hib disease after appropriate vaccination (i.e. vaccine failure) were contacted to complete a questionnaire relating to their health and 260 (80.5%) completed the questionnaire. Of the 124 children with meningitis, 18.5% reported serious long-term sequelae and a further 12.1% of parents attributed other problems to Hib meningitis. Overall, 14% (32/231 cases) of otherwise healthy children and 59% (17/29 cases) of children with an underlying condition developed at least one other serious infection requiring hospital admission. In a Poisson regression model, the risk of another serious infection was independently associated with the presence of an underlying medical condition (incidence risk ratio (IRR) 7.6, 95% CI 4.8–12.1; p < 0.0001), both parents having had a serious infection (IRR 4.1, 95% CI 1.6–10.3; p = 0.003), requirement of more than two antibiotic courses per year (IRR 2.3, 95% CI 1.4–3.6; p = 0.001) and the presence of a long-term complication after Hib infection (IRR 1.8, 95% CI 1.1–3.1; p = 0.03). Thus, rates of long-term sequelae in children with vaccine failure who developed Hib meningitis are similar to those in unvaccinated children in the pre-vaccine era. One in seven otherwise healthy children (14%) with Hib vaccine failure will go on to suffer another serious infection requiring hospital admission in childhood, which is higher than would be expected for the UK paediatric population. © 2009 Elsevier Ltd. All rights reserved.
1. Introduction Haemophilus influenzae serotype b (Hib) conjugate vaccines are very effective in preventing invasive Hib disease [1], even in young infants [2,3]. In the United Kingdom, the introduction of the Hib conjugate vaccine as part of an accelerated 2, 3 and 4 month schedule in October 1992, alongside a year-long catch-up campaign targeting children <5 years, resulted in a rapid and dramatic reduction in invasive Hib disease [4]. Unlike other countries, the UK did not introduce a booster dose in the second year of life until September 2006 [5] because initial studies demonstrated only a small decline in protection over the first 5 years of life [6–7].
∗ Corresponding author at: Immunisation Department, Centre for Infections, Health Protection Agency, 61 Colindale Avenue, London NW9 5EQ, United Kingdom. Tel.: +44 208 327 7155. E-mail address: [email protected] (S. Ladhani). 0264-410X/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2009.12.057
The development of invasive Hib disease after appropriate immunisation (i.e. vaccine failure) is rare, with an estimated vaccine failure rate of 2.2 per 100,000 vaccinees (95% confidence interval (CI), 1.8–2.7 per 100,000) in the UK [6]. We have previously reported that up to 44% of children with Hib vaccine failure had an underlying clinical risk factor (20%) and/or immunological deficiency (30%), with around 12% of children being born prematurely [8]. We undertook a long-term follow-up study to determine whether children with Hib vaccine failure were at risk of other serious infections (including recurrent invasive Hib disease) and, if they were, to identify possible risk factors. We also aimed to describe long-term complications of invasive Hib disease in those with vaccine failure.
2. Materials and methods National surveillance of H. influenzae infection was conducted through a combination of routine laboratory reporting
2196
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
and enhanced surveillance, supplemented by clinical reporting schemes. Laboratory reports from microbiologists working in nearly 400 hospital and public health laboratories have been received at the Health Protection Agency Centre for Infections since 1975. In addition, clinicians, microbiologists and public health doctors were encouraged to report all cases of H. influenzae infections to the Haemophilus Reference Unit, and then to refer a clinical isolate for confirmation and serotyping [8]. From October 1992 until September 2000, the Oxford Vaccine Group undertook a national clinical reporting scheme through the British Paediatric Surveillance Unit (BPSU), asking paediatricians to report cases of invasive Hib disease (defined as isolation of Hib from a normally sterile site) on a report card sent to them on a monthly basis [4,8]. The reporting paediatrician was then sent a questionnaire requesting clinical, demographic and laboratory information for each case [7]. The local microbiologist was also contacted to send the clinical isolate to the Haemophilus Reference Unit if they had not already been done so. Eligible cases with isolates referred to the Haemophilus Reference Unit and those reported by laboratories or through enhanced surveillance were entered into a single Dataease database, reconciled and de-duplicated regularly. For each case, the Hib vaccination status as recorded in the child’s computerised medical records was obtained from the child’s general practitioner (GP). Hib vaccine failure was defined as invasive Hib disease occurring any time after three doses of Hib vaccine in the first year of life, or >1 week after two doses in the first year of life, or >2 weeks after one dose given to a child older than 12 months [6,8]. For this study, the GPs for all children with Hib vaccine failure were contacted at least 1 year after invasive Hib disease for permission to approach the family. The families of children who had died were excluded from the study, as were families whom the GP considered inappropriate to contact. Following a positive response from the GP, the families were invited to complete a short questionnaire relating to inter alia: (a) serious infection episodes in the index case; (b) risk factors for childhood infections; (c) serious infections in the immediate family; and (d) parental employment and social status. A serious infection was defined as an episode of bacterial or viral infection that was severe enough to warrant hospital admission. Permission was also requested to contact the family regarding queries arising from their questionnaire response, and the child’s GP and/or paediatrician, to clarify the nature of any illness and to confirm any reported long-term sequelae. Questionnaire data were entered into Microsoft Excel by one author (RA) and independently verified by another (SL), and then transferred into Stata 9.0 for analysis. Continuous data are expressed as mean ± 95% CI if they followed a Normal distribution or as median and interquartile ranges (IQR). Incidence rate ratios, which adjust for the variable duration of follow-up for each case, were used to compare risk factors for another serious infection among children with Hib vaccine failure. A Poisson regression model, which takes into account the number of serious infections per case as well as the duration of follow-up for each case, was used to analyse the relationship between the different explanatory variables and the risk of another serious infection. All parameters in the univariate analysis with a p value of less than 0.2 were incorporated into the model. The least statistically significant parameter was then sequentially removed in a stepwise manner until only those parameters with a p value <0.05 remained. Ethical agreement for this study was obtained from the Thames Valley Multi-Centre Research Ethics Committee (Reference 05/MRE12/50). 3. Results There were 388 Hib vaccine failure cases in the UK over the study period, of whom nine (2.3%) died during the acute episode. After
extensive follow-up, 323 families finally received the questionnaire, of whom 260 (80.5%) returned the completed questionnaire (Fig. 1). The proportion of cases that completed the questionnaire ranged between 73.3% and 100% per year for the years included in the study.
3.1. Long-term follow-up Of the 260 cases with a completed questionnaire, 47.5% (124 cases) initially had meningitis, 24.2% (63 cases) had epiglottitis, 15.0% (39 cases) had septicaemia and 13.1% (34 cases) had other invasive Hib infections, including pneumonia (14 cases), orbital cellulitis (9 cases), septic arthritis (7 cases), skin and soft tissue infection (3 cases) and tracheitis (1 case). Compared with those that did not respond to the questionnaire, meningitis (48% vs. 37%, p = 0.17) and epiglottitis (24% vs. 16%, p = 0.18) were overrepresented, while septicaemia (15% vs. 24%, p = 0.13) and other infections (13% vs. 22%, p = 0.12) were under-represented among responders. Almost all children were White (250/260 children, 96.2%), while three were Asian, one African, two mixed White/Asian and four mixed White/Afro-Caribbean. Only one child (who did not have recurrent infections) had consanguineous parents. The median age at initial Hib disease was 2.3 years (IQR 1.5–3.6, range 0.5–11.5 years), median age at follow-up was 7.9 years (IQR 5.8–11.4, range 2.7–14.9 years) and median duration of follow-up was 4.0 years (IQR 3.3–7.1, range 1.1–14.5 years) (Table 1). Twenty-five children (9.6%) had been born prematurely (<37 weeks gestation) and 29 (11.1%) had serious underlying conditions, including malignancy (8 cases), Down syndrome (8 cases), congenital abnormalities (3 cases), primary immune deficiency (3 cases; common variable immune deficiency, cyclical neutropenia and primary immunodeficiency of unknown aetiology), metabolic disorders (3 cases; congenital adrenal hyperplasia, diabetes insipidus and Addison’s disease), quadriplegic cerebral palsy (2 cases) and congenital heart disease (2 cases). Children with underlying medical conditions were more likely to have been born prematurely (7/29 (24.1%) vs. 18/231 (7.8%), p = 0.005). In total, 47 children (18.1%) had either been born prematurely, had an underlying condition or both. The long-term complication rate was 10.4% (Table 1) and children with meningitis had highest rate (18.5%). Fourteen children (11.3%) had hearing loss (bilateral and severe in 8 cases; bilateral and moderate in 3 cases; unilateral and severe in 3 cases). Other long-term complications included cerebellar damage affecting balance and mobility (3 cases), and one case each of absence seizures, hydrocephalus requiring a ventriculo-peritoneal shunt, quadriplegic cerebral palsy, right-sided hemiplegia, paraplegia and cranial nerve palsy. The families of a further 15 children (12.1%) with Hib meningitis attributed other complications to the infection because the symptoms began after the illness. These included behavioural problems (4 cases), mild hearing problems with limited loss of certain frequencies (3 cases), hyperactivity requiring medical treatment (3 cases), clumsiness (2 cases), dyslexia/dyspraxia (2 cases) and recurrent severe headaches (1 case). Three children with septicaemia (7.7%) had severe long-term complications, including below-knee amputations, paraplegia and subglottic stenosis following prolonged ventilation. One of 14 children (7.1%) with Hib pneumonia required a lobectomy for lung necrosis. Children with epiglottitis did not have long-term complications. There was no association between long-term sequelae and age at disease, gender, prematurity or presence of an underlying condition.
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
2197
Fig. 1. Recruitment of Hib vaccine failure cases for the long-term follow-up questionnaire study.
3.2. Risk of further infection Almost one-fifth (49/260, 18.8%) of children had another serious infection requiring hospital admission and 20 (7.7%) had at least two serious infections (Table 2), giving an overall incidence of 30 episodes per 1000 child-years. Infections occurred anytime up to 4 years after development of invasive Hib disease, with no evidence of clustering before, during or after Hib infection. It was not possible to reliably distinguish between bacterial
and viral infections for further analysis. However, septicaemia and central line infections were more prevalent among children with underlying medical conditions, while gastrointestinal infections were more common among otherwise healthy children (Table 2). None of the children had a second episode of invasive Hib disease and none reported another Hib infection in the same family. In a univariate analysis, another serious infection was associated with the presence of an underlying medical condition, requirement
Table 1 Initial clinical presentation, median age at disease, underlying conditions, sequelae and occurrence of other serious infections among 260 children with Hib vaccine failure. Diagnosis
Median age at disease (years) (IQR)
Prematurity
Underlying condition
Sequelae
At least 1 other serious infection
Meningitis (n = 124) Epiglottitis (n = 63) Septicaemia (n = 39) Other (n = 34)
2.0 (1.1–3.0) 2.8 (2.1–4.5) 2.9 (1.8–4.0) 1.9 (1.1–4.6)
6 (4.8%) 5 (7.9%) 10 (25.6%) 4 (11.8%)
10 (8.1%) 7 (11.1%) 6 (15.4%) 6 (17.7%)
23 (18.5%) 0 (0.0%) 3 (7.7%) 1 (2.9%)
21 (16.9%) 8 (12.7%) 12 (30.8%) 8 (23.5%)
Total (n = 260)
2.3 (1.4–3.6)
25 (9.6%)
29 (11.2%)
27 (10.4%)
49 (18.9%)
At least 2 other serious infections 8 (6.5%) 1 (1.6%) 4 (10.2%) 7 (20.6%) 20 (7.7%)
2198
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
Table 2 Episodes of serious infection requiring hospital admission among Hib vaccine failure cases with and without underlying conditions. Seventeen children with an underlying medical condition had 49 episodes of infection and 32 otherwise healthy children had 35 episodes of infection. Total number (%) of serious infections
Serious infections among 17 of 29 children with underlying conditions
Serious infections among 32 of 231 children without underlying conditions
Septicaemia Pneumonia Bronchiolitis Urinary tract infection Central line infection Tonsillitis Varicella Bacterial meningitis Skin and soft tissue infection Gastrointestinal infection Bone and joint infection Other infections
19 (22.6%) 17 (20.2%) 8 (9.5%) 7 (8.3%) 5 (6.0%) 5 (6.0%) 4 (4.8%) 4 (4.8%) 4 (4.8%) 4 (4.8%) 2 (2.4%) 5 (6.0%)
17 (34.7%) 11 (22.4%) 3 (6.1%) 2 (4.1%) 5 (10.2%) 2 (4.1%) 3 (6.1%) 2 (4.1%) 1 (2.0%) 0 (0.0%) 1 (2.0%) 2 (4.1%)
2 (5.7%) 6 (17.1%) 5 (14.3%) 5 (14.3%) 0 (0.0%) 3 (8.6%) 1 (2.9%) 2 (5.7%) 3 (8.6%) 4 (11.4%) 1 (2.9%) 3 (8.6%)
Total
84 (100%)
49 (100%)
35 (100%)
p value 0.002 0.55 0.21 0.095 0.05 0.39 0.49 0.73 0.17 0.015 0.81 0.39
Table 3 Univariate analysis for risk of other serious infections among children with Hib vaccine failure. Data are presented as number of cases with percentage in parentheses, unless otherwise stated. Parameter
Total (n = 260)
No other serious infections (n = 211)
At least one other serious infection (n = 49)
Median age at disease (IQR)a Median years followed-up (IQR)a Age at Hib disease <1 year Male sex Prematurity
2.3 (1.5–3.6) 4.0 (3.3–6.7) 34 (13.1) 138 (53.1) 25 (9.6)
2.2 (1.4–3.5) 4.0 (3.4–7.0) 27 (12.8) 105 (49.8) 22 (10.4)
2.7 (1.7–4.8) 3.9 (2.8–6.0) 7 (14.3) 33 (67.4) 3 (6.1)
Underlying condition >2 courses of antibiotics/year
29 (11.2) 74 (28.5)
12 (5.7) 48 (22.8)
Presented with Hib meningitis Presented with Hib epiglottitis Presented with Hib septicaemia Presented with other Hib diagnoses Presence of sequelae
124 (47.7) 63 (24.2) 39 (15.0) 34 (13.1) 27 (10.4)
103 (48.8) 55 (26.1) 27 (12.8) 26 (12.3) 18 (8.5%)
Has at least one sibling Serious infection in a sibling Serious infection in both parentsb
255 (86.5) 21 (8.1) 4 (1.5)
179 (84.3) 16 (7.6) 1 (0.47)
a b
17 (34.7) 26 (53.1)
Incidence risk ratio (95% CI)
p value
– – 0.70 (0.36–1.36) 1.59 (0.99–2.57) 0.53 (0.17–1.67)
0.24 0.18 0.30 0.05 0.28
10.9 (7.13–16.8) 3.72 (2.39–5.80)
<0.0001 <0.0001
21 (42.9) 8 (16.3) 12 (24.5) 8 (16.3) 9 (18.4%)
0.67 (0.43–1.03) 0.48 (0.24–1.07) 2.31 (1.39–3.85) 1.64 (0.99–2.71) 2.62 (1.57–4.36)
0.07 0.15 0.001 0.06 <0.0001
46 (93.9) 5 (10.2) 3 (6.1)
1.92 (0.70–5.24) 1.09 (0.52–2.25) 7.13 (2.89–17.6)
0.20 0.82 <0.0001
Medians were compared statistically using the Mann–Whitney U test. Serious infections reported individually by either mother or father were not significantly associated with an increased risk of another serious infection.
of more than two courses of antibiotics per year, original presentation with Hib septicaemia, development of long-term sequelae, and the reporting of both parents having at least one serious infection requiring hospital admission (Table 3). These variables remained statistically significant (p < 0.05) after correcting for multiple comparisons (Bonferroni’s Correction, n = 30). There was no association with attending a nursery, breastfeeding (for any period), having a sibling, number of siblings, parental smoking or employment status, or parental ownership of a car and/or accommodation (data not shown). The proportion of children with an underlying condition increased with the number of serious infections: only 6% (12/199 cases) of children with no further infection had an underlying condition compared with 10% (3/29 cases) of those with one infection, 40% (2/5 cases) of those with two infections and 80% (12/15 cases) of those with three or more infections (p < 0.0001 for trend). Expressed another way, 86.2% (199/231 cases) of cases with no underlying condition did not develop another serious infection, while 58.6% (17/29 cases) of children with an underlying condition had at least one other serious infection. In a Poisson regression model, the presence of an underlying condition, the reporting of both parents having at least one serious infection requiring hospital admission (4 cases), requirement of more than two courses of antibiotics per year and the presence of long-term sequelae all remained independently associated with an increased risk of another serious infection (Table 4).
Table 4 Poisson regression for risk of other serious infections among children with Hib vaccine failure. Risk factor
Incidence risk ratio
p value
Presence of an underlying condition Both parents reporting at least one serious infection requiring hospital admission Requirement of more than two courses of antibiotics per year Presence of a long-term complication after the Hib infection
7.6 (4.8–12.1) 4.1 (1.6–10.3)
<0.0001 0.003
2.3 (1.4–3.6)
0.001
1.8 (1.1–3.1)
0.03
4. Discussion The development of invasive Hib disease in spite of vaccination is rare and raises the possibility that such children may have an underlying susceptibility to other serious infections, although it is well-documented that the lack of a 12 month booster dose in the UK immunisation schedule and the use of a less immunogenic whole-cell pertussis-containing Hib vaccine between 2000 and 2003 are both likely to have contributed to the increased Hib vaccine failure cases after 1999 [5]. To our knowledge, this is the first study to determine the risk of other serious infections in children with Hib vaccine failure. We believe that our long-term follow-up cohort is representative of children with Hib vaccine
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
failure as a whole because of the high questionnaire response rate and, although the proportion of meningitis and epiglottitis cases were over-represented, this difference was not statistically significant. Moreover, our cohort is similar to our previous shortterm follow-up study on UK children with Hib vaccine failure, in terms of clinical presentation and underlying medical conditions [8]. Because only Hib vaccine failure cases were included in this study, it was not possible to assess whether unvaccinated children who developed invasive Hib disease would also be at risk of other serious infections. Because we relied on parental recall for episodes of serious infections in respect of themselves and their children, we do not have the results of confirmatory microbiological tests for the reported episodes. However, we would suggest that serious infections requiring hospital admission are uncommon and parents are likely to remember if they or their children had been admitted to the hospital for a serious infection. It is perhaps not surprising that children with underlying conditions were more likely to develop multiple episodes of serious infection – children with malignancies undergoing chemotherapy and those with primary immunodeficiencies, for example, would be immunosuppressed and, therefore, at higher risk of serious infections. Children with long-term complications due to their initial Hib infection were also more likely to develop another serious infection. The data did not show any clustering of hospital admissions in the period after the Hib infection episode, which might have indicated lower threshold by clinicians to admit such children to hospital with minor illnesses because of their recent history of severe invasive Hib disease. However, around 14% of otherwise healthy children (i.e. one in seven children) with Hib vaccine failure also had another serious infection. In England, it is estimated that 5% of children <15 years (10% of children <5 years and 2% of 5–14 year-olds) are admitted to hospital [9]. Only a proportion of these admissions will be due to an infection. In addition, the number of children admitted with a second episode of serious infection would be expected to be much lower. Thus, the finding of a 14% admission rate for another serious infection was substantially higher than the 5% expected for the general paediatric population. Another risk factor independently associated with the risk of another serious infection was frequent use of antibiotics in the early years of life. Frequent antibiotic use in early childhood was also reported as an independent risk factor for invasive Hib disease in a recent UK case–control study [10], raising the possibility that these children may be genetically and/or environmentally more susceptible to recurrent infections that require oral antibiotic treatment by GPs. Finally, the reporting of a serious infection by both parents was also independently associated with an increased risk of multiple serious infections in the child suggesting that, at least in a small proportion of families (3/49, 6.1%), a genetic predisposition to serious infections may be present. Interestingly, none of the other family members developed invasive Hib disease and none of the cases had a second episode of Hib infection; this may be explained by the low background rates of Hib disease in the pre-vaccine era and the even lower rates now that routine vaccination has reduced carriage and transmission [11]. We were unable to demonstrate any significant association between the risk of another serious infection and social and/or environmental factors. Historically, risk factors for invasive Hib disease included lack of breastfeeding [12–14], having an older sibling [15,16], household crowding [17,18], single parent families [10], smoking in the household [18], nursery attendance [12,15–17,19–21] and low socio-economic status [18,19] – including non-owner household occupancy and lack of access to a car [22], which have been shown to be reliable markers of income [23]. In our study, nine children are known to have died of invasive Hib disease (2.3%), a figure higher than previously reported (1.6%) [8], but lower than case fatality rates reported in England in the
2199
pre-vaccine era (4.3%) [24]. However, the 2.3% may still be an underestimate because it is possible that at least some of the 32 children that we were unable to trace had also died. Of the 260 children that completed the questionnaire, 18.5% of children with Hib meningitis reported significant long-term sequelae (including moderate-tosevere deafness in 11%), which were confirmed by the children’s GP, and a further 12% of parents attributed other problems to Hib meningitis. Although our data was collected using a parental questionnaire and, hence, could be subject to error or bias, previous studies in the pre-vaccine era have reported similar results. A systematic review addressing hearing impairment following bacterial meningitis, for example, reported permanent sensorineural hearing impairment in 11.4% of patients following Hib meningitis [25]. Similarly, one large study reported that 27% of children surviving meningitis had major (IQ < 70, seizures, hydrocephalus, spasticity, blindness, or severe to profound hearing loss, 8.5%) or minor (IQ 70–80, inability to read, mild to moderate hearing loss, abnormalities in speech discrimination, or school behaviour problems, 18.5%) sequelae that affected their academic performance [26]. Thus, the rate of long-term sequelae in UK children with Hib vaccine failure is comparable to that in unvaccinated children in the pre-vaccine era. Clinicians should be aware that one in seven otherwise healthy children (14%) and up to two-thirds of the children with an underlying condition (59%) with Hib vaccine failure will have at least one other serious infection requiring hospital admission in childhood. However, it is reassuring to note that none had a second episode of Hib disease and none of the close family contacts developed invasive Hib disease either. Our findings may also be relevant for children with other conjugate vaccine failure. Acknowledgements The authors would like to thank the General Practitioners, paediatricians and families of all children with Hib vaccine failure who participated in this study. The authors are also grateful to Dr Nick Andrews, Departmental Statistician at the Centre for Infections, Health Protection Agency, for helping with the statistical analyses. SL was awarded a two-year European Society for Paediatric Infectious Diseases (ESPID) to complete this study. AJP is a Jenner Institute Investigator and is funded by the NIHR Oxford Biomedical Research Centre. References [1] Swingler G, Fransman D, Hussey G. Conjugate vaccines for preventing Haemophilus influenzae type B infections. Cochrane Database Syst Rev 2007. CD001729. [2] Booy R, Hodgson S, Carpenter L, Mayon-White RT, Slack MP, Macfarlane JA, et al. Efficacy of Haemophilus influenzae type b conjugate vaccine PRP-T. Lancet 1994;344:362–6. [3] Booy R, Taylor SA, Dobson SR, Isaacs D, Sleight G, Aitken S, et al. Immunogenicity and safety of PRP-T conjugate vaccine given according to the British accelerated immunisation schedule. Arch Dis Child 1992;67:475–8. [4] Heath PT, McVernon J. The UK Hib vaccine experience. Arch Dis Child 2002;86:396–9. [5] Ladhani S, Slack MP, Heys M, White J, Ramsay ME. Fall in Haemophilus influenzae serotype b (Hib) disease following implementation of a booster campaign. Arch Dis Child 2008;93:665–9. [6] Heath PT, Booy R, Azzopardi HJ, Slack MP, Bowen-Morris J, Griffiths H, et al. Antibody concentration and clinical protection after Hib conjugate vaccination in the United Kingdom. JAMA 2000;284:2334–40. [7] Booy R, Heath PT, Slack MP, Begg N, Moxon ER. Vaccine failures after primary immunisation with Haemophilus influenzae type-b conjugate vaccine without booster. Lancet 1997;349:1197–202. [8] Heath PT, Booy R, Griffiths H, Clutterbuck E, Azzopardi HJ, Slack MP, 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. [9] MacFaul R, Werneke U. Recent trends in hospital use by children in England. Arch Dis Child 2001;85:203–7. [10] McVernon J, Andrews N, Slack M, Moxon R, Ramsay M. Host and environmental factors associated with Hib in England, 1998–2002. Arch Dis Child 2008;93:670–5.
2200
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
[11] McVernon J, Ramsay ME, McLean AR. Understanding the impact of Hib conjugate vaccine on transmission, immunity and disease in the United Kingdom. Epidemiol Infect 2008;136:800–12. [12] Takala AK, Eskola J, Palmgren J, Ronnberg PR, Kela E, Rekola P, et al. Risk factors of invasive Haemophilus influenzae type b disease among children in Finland. J Pediatr 1989;115:694–701. [13] Silfverdal SA, Bodin L, Hugosson S, Garpenholt O, Werner B, Esbjorner E, et al. Protective effect of breastfeeding on invasive Haemophilus influenzae infection: a case-control study in Swedish preschool children. Int J Epidemiol 1997;26:443–50. [14] Wolff MC, Moulton LH, Newcomer W, Reid R, Santosham M. A case-control study of risk factors for Haemophilus influenzae type B disease in Navajo children. Am J Trop Med Hyg 1999;60:263–6. [15] Istre GR, Conner JS, Broome CV, Hightower A, Hopkins RS. Risk factors for primary invasive Haemophilus influenzae disease: increased risk from day care attendance and school-aged household members. J Pediatr 1985;106: 190–5. [16] Zielinski A, Kwon CB, Tomaszunas-Blaszczyk J, Magdzik W, Bennett JV. Risk of Haemophilus influenzae type b meningitis in Polish children varies directly with number of siblings: possible implications for vaccination strategies. Eur J Epidemiol 2003;18:917–22. [17] Cochi SL, Fleming DW, Hightower AW, Limpakarnjanarat K, Facklam RR, Smith JD, et al. Primary invasive Haemophilus influenzae type b disease: a populationbased assessment of risk factors. J Pediatr 1986;108:887–96.
[18] Vadheim CM, Greenberg DP, Bordenave N, Ziontz L, Christenson P, Waterman SH, et al. Risk factors for invasive Haemophilus influenzae type b in Los Angeles County children 18–60 months of age. Am J Epidemiol 1992;136:221–35. [19] Broome CV. Epidemiology of Haemophilus influenzae type b infections in the United States. Pediatr Infect Dis J 1987;6:779–82. [20] Fogarty J, Moloney AC, Newell JB. The epidemiology of Haemophilus influenzae type b disease in the Republic of Ireland. Epidemiol Infect 1995;114:451–63. [21] Redmond SR, Pichichero ME. Hemophilus influenzae type b disease. An epidemiologic study with special reference to day-care centers. JAMA 1984;252:2581–4. [22] Olowokure B, Spencer NJ, Hawker JI, Blair I, Smith RL. Changing socioeconomic risk factors for invasive H. influenzae disease after the introduction of conjugate vaccine. J Infect 2003;46:46–8. [23] Macintyre S, Ellaway A, Der G, Ford G, Hunt K. Do housing tenure and car access predict health because they are simply markers of income or self esteem? A Scottish study. J Epidemiol Community Health 1998;52:657–64. [24] Booy R, Hodgson SA, Slack MP, Anderson EC, Mayon-White RT, Moxon ER. Invasive Haemophilus influenzae type b disease in the Oxford region (1985–91). Arch Dis Child 1993;69:225–8. [25] Fortnum HM. Hearing impairment after bacterial meningitis: a review. Arch Dis Child 1992;67:1128–33. [26] Grimwood K, Anderson VA, Bond L, Catroppa C, Hore RL, Keir EH, et al. Adverse outcomes of bacterial meningitis in school-age survivors. Pediatrics 1995;95:646–56.
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Long-term complications and risk of other serious infections following invasive Haemophilus influenzae serotype b disease in vaccinated children Shamez Ladhani a,∗ , Paul T. Heath b , Rashna J. Aibara c , Mary E. Ramsay d , Mary P.E. Slack d , Martin L. Hibberd e , Andrew J. Pollard f , E. Richard Moxon f , Robert Booy a,g a
Academic Unit of Paediatrics, Barts and The London School of Medicine and Dentistry, London, United Kingdom Vaccine Institute and Division of Child Health, St George’s, University of London, United Kingdom Department of Paediatrics, Central Middlesex Hospital, London, United Kingdom d Centre for Infections, Health Protection Agency, London, United Kingdom e Infectious Diseases Unit, Genome Institute of Singapore, Republic of Singapore f Oxford Vaccine Group, Department of Paediatrics, University of Oxford, United Kingdom g National Centre for Immunisation Research & Surveillance, University of Sydney, Australia b c
a r t i c l e
i n f o
Article history: Received 27 July 2009 Received in revised form 10 December 2009 Accepted 23 December 2009 Available online 5 January 2010 Keywords: Conjugate vaccine Meningitis Sequelae
a b s t r a c t This study describes the long-term complications in children with Haemophilus influenzae serotype b (Hib) vaccine failure and to determine their risk of other serious infections. The families of 323 children with invasive Hib disease after appropriate vaccination (i.e. vaccine failure) were contacted to complete a questionnaire relating to their health and 260 (80.5%) completed the questionnaire. Of the 124 children with meningitis, 18.5% reported serious long-term sequelae and a further 12.1% of parents attributed other problems to Hib meningitis. Overall, 14% (32/231 cases) of otherwise healthy children and 59% (17/29 cases) of children with an underlying condition developed at least one other serious infection requiring hospital admission. In a Poisson regression model, the risk of another serious infection was independently associated with the presence of an underlying medical condition (incidence risk ratio (IRR) 7.6, 95% CI 4.8–12.1; p < 0.0001), both parents having had a serious infection (IRR 4.1, 95% CI 1.6–10.3; p = 0.003), requirement of more than two antibiotic courses per year (IRR 2.3, 95% CI 1.4–3.6; p = 0.001) and the presence of a long-term complication after Hib infection (IRR 1.8, 95% CI 1.1–3.1; p = 0.03). Thus, rates of long-term sequelae in children with vaccine failure who developed Hib meningitis are similar to those in unvaccinated children in the pre-vaccine era. One in seven otherwise healthy children (14%) with Hib vaccine failure will go on to suffer another serious infection requiring hospital admission in childhood, which is higher than would be expected for the UK paediatric population. © 2009 Elsevier Ltd. All rights reserved.
1. Introduction Haemophilus influenzae serotype b (Hib) conjugate vaccines are very effective in preventing invasive Hib disease [1], even in young infants [2,3]. In the United Kingdom, the introduction of the Hib conjugate vaccine as part of an accelerated 2, 3 and 4 month schedule in October 1992, alongside a year-long catch-up campaign targeting children <5 years, resulted in a rapid and dramatic reduction in invasive Hib disease [4]. Unlike other countries, the UK did not introduce a booster dose in the second year of life until September 2006 [5] because initial studies demonstrated only a small decline in protection over the first 5 years of life [6–7].
∗ Corresponding author at: Immunisation Department, Centre for Infections, Health Protection Agency, 61 Colindale Avenue, London NW9 5EQ, United Kingdom. Tel.: +44 208 327 7155. E-mail address: [email protected] (S. Ladhani). 0264-410X/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2009.12.057
The development of invasive Hib disease after appropriate immunisation (i.e. vaccine failure) is rare, with an estimated vaccine failure rate of 2.2 per 100,000 vaccinees (95% confidence interval (CI), 1.8–2.7 per 100,000) in the UK [6]. We have previously reported that up to 44% of children with Hib vaccine failure had an underlying clinical risk factor (20%) and/or immunological deficiency (30%), with around 12% of children being born prematurely [8]. We undertook a long-term follow-up study to determine whether children with Hib vaccine failure were at risk of other serious infections (including recurrent invasive Hib disease) and, if they were, to identify possible risk factors. We also aimed to describe long-term complications of invasive Hib disease in those with vaccine failure.
2. Materials and methods National surveillance of H. influenzae infection was conducted through a combination of routine laboratory reporting
2196
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
and enhanced surveillance, supplemented by clinical reporting schemes. Laboratory reports from microbiologists working in nearly 400 hospital and public health laboratories have been received at the Health Protection Agency Centre for Infections since 1975. In addition, clinicians, microbiologists and public health doctors were encouraged to report all cases of H. influenzae infections to the Haemophilus Reference Unit, and then to refer a clinical isolate for confirmation and serotyping [8]. From October 1992 until September 2000, the Oxford Vaccine Group undertook a national clinical reporting scheme through the British Paediatric Surveillance Unit (BPSU), asking paediatricians to report cases of invasive Hib disease (defined as isolation of Hib from a normally sterile site) on a report card sent to them on a monthly basis [4,8]. The reporting paediatrician was then sent a questionnaire requesting clinical, demographic and laboratory information for each case [7]. The local microbiologist was also contacted to send the clinical isolate to the Haemophilus Reference Unit if they had not already been done so. Eligible cases with isolates referred to the Haemophilus Reference Unit and those reported by laboratories or through enhanced surveillance were entered into a single Dataease database, reconciled and de-duplicated regularly. For each case, the Hib vaccination status as recorded in the child’s computerised medical records was obtained from the child’s general practitioner (GP). Hib vaccine failure was defined as invasive Hib disease occurring any time after three doses of Hib vaccine in the first year of life, or >1 week after two doses in the first year of life, or >2 weeks after one dose given to a child older than 12 months [6,8]. For this study, the GPs for all children with Hib vaccine failure were contacted at least 1 year after invasive Hib disease for permission to approach the family. The families of children who had died were excluded from the study, as were families whom the GP considered inappropriate to contact. Following a positive response from the GP, the families were invited to complete a short questionnaire relating to inter alia: (a) serious infection episodes in the index case; (b) risk factors for childhood infections; (c) serious infections in the immediate family; and (d) parental employment and social status. A serious infection was defined as an episode of bacterial or viral infection that was severe enough to warrant hospital admission. Permission was also requested to contact the family regarding queries arising from their questionnaire response, and the child’s GP and/or paediatrician, to clarify the nature of any illness and to confirm any reported long-term sequelae. Questionnaire data were entered into Microsoft Excel by one author (RA) and independently verified by another (SL), and then transferred into Stata 9.0 for analysis. Continuous data are expressed as mean ± 95% CI if they followed a Normal distribution or as median and interquartile ranges (IQR). Incidence rate ratios, which adjust for the variable duration of follow-up for each case, were used to compare risk factors for another serious infection among children with Hib vaccine failure. A Poisson regression model, which takes into account the number of serious infections per case as well as the duration of follow-up for each case, was used to analyse the relationship between the different explanatory variables and the risk of another serious infection. All parameters in the univariate analysis with a p value of less than 0.2 were incorporated into the model. The least statistically significant parameter was then sequentially removed in a stepwise manner until only those parameters with a p value <0.05 remained. Ethical agreement for this study was obtained from the Thames Valley Multi-Centre Research Ethics Committee (Reference 05/MRE12/50). 3. Results There were 388 Hib vaccine failure cases in the UK over the study period, of whom nine (2.3%) died during the acute episode. After
extensive follow-up, 323 families finally received the questionnaire, of whom 260 (80.5%) returned the completed questionnaire (Fig. 1). The proportion of cases that completed the questionnaire ranged between 73.3% and 100% per year for the years included in the study.
3.1. Long-term follow-up Of the 260 cases with a completed questionnaire, 47.5% (124 cases) initially had meningitis, 24.2% (63 cases) had epiglottitis, 15.0% (39 cases) had septicaemia and 13.1% (34 cases) had other invasive Hib infections, including pneumonia (14 cases), orbital cellulitis (9 cases), septic arthritis (7 cases), skin and soft tissue infection (3 cases) and tracheitis (1 case). Compared with those that did not respond to the questionnaire, meningitis (48% vs. 37%, p = 0.17) and epiglottitis (24% vs. 16%, p = 0.18) were overrepresented, while septicaemia (15% vs. 24%, p = 0.13) and other infections (13% vs. 22%, p = 0.12) were under-represented among responders. Almost all children were White (250/260 children, 96.2%), while three were Asian, one African, two mixed White/Asian and four mixed White/Afro-Caribbean. Only one child (who did not have recurrent infections) had consanguineous parents. The median age at initial Hib disease was 2.3 years (IQR 1.5–3.6, range 0.5–11.5 years), median age at follow-up was 7.9 years (IQR 5.8–11.4, range 2.7–14.9 years) and median duration of follow-up was 4.0 years (IQR 3.3–7.1, range 1.1–14.5 years) (Table 1). Twenty-five children (9.6%) had been born prematurely (<37 weeks gestation) and 29 (11.1%) had serious underlying conditions, including malignancy (8 cases), Down syndrome (8 cases), congenital abnormalities (3 cases), primary immune deficiency (3 cases; common variable immune deficiency, cyclical neutropenia and primary immunodeficiency of unknown aetiology), metabolic disorders (3 cases; congenital adrenal hyperplasia, diabetes insipidus and Addison’s disease), quadriplegic cerebral palsy (2 cases) and congenital heart disease (2 cases). Children with underlying medical conditions were more likely to have been born prematurely (7/29 (24.1%) vs. 18/231 (7.8%), p = 0.005). In total, 47 children (18.1%) had either been born prematurely, had an underlying condition or both. The long-term complication rate was 10.4% (Table 1) and children with meningitis had highest rate (18.5%). Fourteen children (11.3%) had hearing loss (bilateral and severe in 8 cases; bilateral and moderate in 3 cases; unilateral and severe in 3 cases). Other long-term complications included cerebellar damage affecting balance and mobility (3 cases), and one case each of absence seizures, hydrocephalus requiring a ventriculo-peritoneal shunt, quadriplegic cerebral palsy, right-sided hemiplegia, paraplegia and cranial nerve palsy. The families of a further 15 children (12.1%) with Hib meningitis attributed other complications to the infection because the symptoms began after the illness. These included behavioural problems (4 cases), mild hearing problems with limited loss of certain frequencies (3 cases), hyperactivity requiring medical treatment (3 cases), clumsiness (2 cases), dyslexia/dyspraxia (2 cases) and recurrent severe headaches (1 case). Three children with septicaemia (7.7%) had severe long-term complications, including below-knee amputations, paraplegia and subglottic stenosis following prolonged ventilation. One of 14 children (7.1%) with Hib pneumonia required a lobectomy for lung necrosis. Children with epiglottitis did not have long-term complications. There was no association between long-term sequelae and age at disease, gender, prematurity or presence of an underlying condition.
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
2197
Fig. 1. Recruitment of Hib vaccine failure cases for the long-term follow-up questionnaire study.
3.2. Risk of further infection Almost one-fifth (49/260, 18.8%) of children had another serious infection requiring hospital admission and 20 (7.7%) had at least two serious infections (Table 2), giving an overall incidence of 30 episodes per 1000 child-years. Infections occurred anytime up to 4 years after development of invasive Hib disease, with no evidence of clustering before, during or after Hib infection. It was not possible to reliably distinguish between bacterial
and viral infections for further analysis. However, septicaemia and central line infections were more prevalent among children with underlying medical conditions, while gastrointestinal infections were more common among otherwise healthy children (Table 2). None of the children had a second episode of invasive Hib disease and none reported another Hib infection in the same family. In a univariate analysis, another serious infection was associated with the presence of an underlying medical condition, requirement
Table 1 Initial clinical presentation, median age at disease, underlying conditions, sequelae and occurrence of other serious infections among 260 children with Hib vaccine failure. Diagnosis
Median age at disease (years) (IQR)
Prematurity
Underlying condition
Sequelae
At least 1 other serious infection
Meningitis (n = 124) Epiglottitis (n = 63) Septicaemia (n = 39) Other (n = 34)
2.0 (1.1–3.0) 2.8 (2.1–4.5) 2.9 (1.8–4.0) 1.9 (1.1–4.6)
6 (4.8%) 5 (7.9%) 10 (25.6%) 4 (11.8%)
10 (8.1%) 7 (11.1%) 6 (15.4%) 6 (17.7%)
23 (18.5%) 0 (0.0%) 3 (7.7%) 1 (2.9%)
21 (16.9%) 8 (12.7%) 12 (30.8%) 8 (23.5%)
Total (n = 260)
2.3 (1.4–3.6)
25 (9.6%)
29 (11.2%)
27 (10.4%)
49 (18.9%)
At least 2 other serious infections 8 (6.5%) 1 (1.6%) 4 (10.2%) 7 (20.6%) 20 (7.7%)
2198
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
Table 2 Episodes of serious infection requiring hospital admission among Hib vaccine failure cases with and without underlying conditions. Seventeen children with an underlying medical condition had 49 episodes of infection and 32 otherwise healthy children had 35 episodes of infection. Total number (%) of serious infections
Serious infections among 17 of 29 children with underlying conditions
Serious infections among 32 of 231 children without underlying conditions
Septicaemia Pneumonia Bronchiolitis Urinary tract infection Central line infection Tonsillitis Varicella Bacterial meningitis Skin and soft tissue infection Gastrointestinal infection Bone and joint infection Other infections
19 (22.6%) 17 (20.2%) 8 (9.5%) 7 (8.3%) 5 (6.0%) 5 (6.0%) 4 (4.8%) 4 (4.8%) 4 (4.8%) 4 (4.8%) 2 (2.4%) 5 (6.0%)
17 (34.7%) 11 (22.4%) 3 (6.1%) 2 (4.1%) 5 (10.2%) 2 (4.1%) 3 (6.1%) 2 (4.1%) 1 (2.0%) 0 (0.0%) 1 (2.0%) 2 (4.1%)
2 (5.7%) 6 (17.1%) 5 (14.3%) 5 (14.3%) 0 (0.0%) 3 (8.6%) 1 (2.9%) 2 (5.7%) 3 (8.6%) 4 (11.4%) 1 (2.9%) 3 (8.6%)
Total
84 (100%)
49 (100%)
35 (100%)
p value 0.002 0.55 0.21 0.095 0.05 0.39 0.49 0.73 0.17 0.015 0.81 0.39
Table 3 Univariate analysis for risk of other serious infections among children with Hib vaccine failure. Data are presented as number of cases with percentage in parentheses, unless otherwise stated. Parameter
Total (n = 260)
No other serious infections (n = 211)
At least one other serious infection (n = 49)
Median age at disease (IQR)a Median years followed-up (IQR)a Age at Hib disease <1 year Male sex Prematurity
2.3 (1.5–3.6) 4.0 (3.3–6.7) 34 (13.1) 138 (53.1) 25 (9.6)
2.2 (1.4–3.5) 4.0 (3.4–7.0) 27 (12.8) 105 (49.8) 22 (10.4)
2.7 (1.7–4.8) 3.9 (2.8–6.0) 7 (14.3) 33 (67.4) 3 (6.1)
Underlying condition >2 courses of antibiotics/year
29 (11.2) 74 (28.5)
12 (5.7) 48 (22.8)
Presented with Hib meningitis Presented with Hib epiglottitis Presented with Hib septicaemia Presented with other Hib diagnoses Presence of sequelae
124 (47.7) 63 (24.2) 39 (15.0) 34 (13.1) 27 (10.4)
103 (48.8) 55 (26.1) 27 (12.8) 26 (12.3) 18 (8.5%)
Has at least one sibling Serious infection in a sibling Serious infection in both parentsb
255 (86.5) 21 (8.1) 4 (1.5)
179 (84.3) 16 (7.6) 1 (0.47)
a b
17 (34.7) 26 (53.1)
Incidence risk ratio (95% CI)
p value
– – 0.70 (0.36–1.36) 1.59 (0.99–2.57) 0.53 (0.17–1.67)
0.24 0.18 0.30 0.05 0.28
10.9 (7.13–16.8) 3.72 (2.39–5.80)
<0.0001 <0.0001
21 (42.9) 8 (16.3) 12 (24.5) 8 (16.3) 9 (18.4%)
0.67 (0.43–1.03) 0.48 (0.24–1.07) 2.31 (1.39–3.85) 1.64 (0.99–2.71) 2.62 (1.57–4.36)
0.07 0.15 0.001 0.06 <0.0001
46 (93.9) 5 (10.2) 3 (6.1)
1.92 (0.70–5.24) 1.09 (0.52–2.25) 7.13 (2.89–17.6)
0.20 0.82 <0.0001
Medians were compared statistically using the Mann–Whitney U test. Serious infections reported individually by either mother or father were not significantly associated with an increased risk of another serious infection.
of more than two courses of antibiotics per year, original presentation with Hib septicaemia, development of long-term sequelae, and the reporting of both parents having at least one serious infection requiring hospital admission (Table 3). These variables remained statistically significant (p < 0.05) after correcting for multiple comparisons (Bonferroni’s Correction, n = 30). There was no association with attending a nursery, breastfeeding (for any period), having a sibling, number of siblings, parental smoking or employment status, or parental ownership of a car and/or accommodation (data not shown). The proportion of children with an underlying condition increased with the number of serious infections: only 6% (12/199 cases) of children with no further infection had an underlying condition compared with 10% (3/29 cases) of those with one infection, 40% (2/5 cases) of those with two infections and 80% (12/15 cases) of those with three or more infections (p < 0.0001 for trend). Expressed another way, 86.2% (199/231 cases) of cases with no underlying condition did not develop another serious infection, while 58.6% (17/29 cases) of children with an underlying condition had at least one other serious infection. In a Poisson regression model, the presence of an underlying condition, the reporting of both parents having at least one serious infection requiring hospital admission (4 cases), requirement of more than two courses of antibiotics per year and the presence of long-term sequelae all remained independently associated with an increased risk of another serious infection (Table 4).
Table 4 Poisson regression for risk of other serious infections among children with Hib vaccine failure. Risk factor
Incidence risk ratio
p value
Presence of an underlying condition Both parents reporting at least one serious infection requiring hospital admission Requirement of more than two courses of antibiotics per year Presence of a long-term complication after the Hib infection
7.6 (4.8–12.1) 4.1 (1.6–10.3)
<0.0001 0.003
2.3 (1.4–3.6)
0.001
1.8 (1.1–3.1)
0.03
4. Discussion The development of invasive Hib disease in spite of vaccination is rare and raises the possibility that such children may have an underlying susceptibility to other serious infections, although it is well-documented that the lack of a 12 month booster dose in the UK immunisation schedule and the use of a less immunogenic whole-cell pertussis-containing Hib vaccine between 2000 and 2003 are both likely to have contributed to the increased Hib vaccine failure cases after 1999 [5]. To our knowledge, this is the first study to determine the risk of other serious infections in children with Hib vaccine failure. We believe that our long-term follow-up cohort is representative of children with Hib vaccine
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
failure as a whole because of the high questionnaire response rate and, although the proportion of meningitis and epiglottitis cases were over-represented, this difference was not statistically significant. Moreover, our cohort is similar to our previous shortterm follow-up study on UK children with Hib vaccine failure, in terms of clinical presentation and underlying medical conditions [8]. Because only Hib vaccine failure cases were included in this study, it was not possible to assess whether unvaccinated children who developed invasive Hib disease would also be at risk of other serious infections. Because we relied on parental recall for episodes of serious infections in respect of themselves and their children, we do not have the results of confirmatory microbiological tests for the reported episodes. However, we would suggest that serious infections requiring hospital admission are uncommon and parents are likely to remember if they or their children had been admitted to the hospital for a serious infection. It is perhaps not surprising that children with underlying conditions were more likely to develop multiple episodes of serious infection – children with malignancies undergoing chemotherapy and those with primary immunodeficiencies, for example, would be immunosuppressed and, therefore, at higher risk of serious infections. Children with long-term complications due to their initial Hib infection were also more likely to develop another serious infection. The data did not show any clustering of hospital admissions in the period after the Hib infection episode, which might have indicated lower threshold by clinicians to admit such children to hospital with minor illnesses because of their recent history of severe invasive Hib disease. However, around 14% of otherwise healthy children (i.e. one in seven children) with Hib vaccine failure also had another serious infection. In England, it is estimated that 5% of children <15 years (10% of children <5 years and 2% of 5–14 year-olds) are admitted to hospital [9]. Only a proportion of these admissions will be due to an infection. In addition, the number of children admitted with a second episode of serious infection would be expected to be much lower. Thus, the finding of a 14% admission rate for another serious infection was substantially higher than the 5% expected for the general paediatric population. Another risk factor independently associated with the risk of another serious infection was frequent use of antibiotics in the early years of life. Frequent antibiotic use in early childhood was also reported as an independent risk factor for invasive Hib disease in a recent UK case–control study [10], raising the possibility that these children may be genetically and/or environmentally more susceptible to recurrent infections that require oral antibiotic treatment by GPs. Finally, the reporting of a serious infection by both parents was also independently associated with an increased risk of multiple serious infections in the child suggesting that, at least in a small proportion of families (3/49, 6.1%), a genetic predisposition to serious infections may be present. Interestingly, none of the other family members developed invasive Hib disease and none of the cases had a second episode of Hib infection; this may be explained by the low background rates of Hib disease in the pre-vaccine era and the even lower rates now that routine vaccination has reduced carriage and transmission [11]. We were unable to demonstrate any significant association between the risk of another serious infection and social and/or environmental factors. Historically, risk factors for invasive Hib disease included lack of breastfeeding [12–14], having an older sibling [15,16], household crowding [17,18], single parent families [10], smoking in the household [18], nursery attendance [12,15–17,19–21] and low socio-economic status [18,19] – including non-owner household occupancy and lack of access to a car [22], which have been shown to be reliable markers of income [23]. In our study, nine children are known to have died of invasive Hib disease (2.3%), a figure higher than previously reported (1.6%) [8], but lower than case fatality rates reported in England in the
2199
pre-vaccine era (4.3%) [24]. However, the 2.3% may still be an underestimate because it is possible that at least some of the 32 children that we were unable to trace had also died. Of the 260 children that completed the questionnaire, 18.5% of children with Hib meningitis reported significant long-term sequelae (including moderate-tosevere deafness in 11%), which were confirmed by the children’s GP, and a further 12% of parents attributed other problems to Hib meningitis. Although our data was collected using a parental questionnaire and, hence, could be subject to error or bias, previous studies in the pre-vaccine era have reported similar results. A systematic review addressing hearing impairment following bacterial meningitis, for example, reported permanent sensorineural hearing impairment in 11.4% of patients following Hib meningitis [25]. Similarly, one large study reported that 27% of children surviving meningitis had major (IQ < 70, seizures, hydrocephalus, spasticity, blindness, or severe to profound hearing loss, 8.5%) or minor (IQ 70–80, inability to read, mild to moderate hearing loss, abnormalities in speech discrimination, or school behaviour problems, 18.5%) sequelae that affected their academic performance [26]. Thus, the rate of long-term sequelae in UK children with Hib vaccine failure is comparable to that in unvaccinated children in the pre-vaccine era. Clinicians should be aware that one in seven otherwise healthy children (14%) and up to two-thirds of the children with an underlying condition (59%) with Hib vaccine failure will have at least one other serious infection requiring hospital admission in childhood. However, it is reassuring to note that none had a second episode of Hib disease and none of the close family contacts developed invasive Hib disease either. Our findings may also be relevant for children with other conjugate vaccine failure. Acknowledgements The authors would like to thank the General Practitioners, paediatricians and families of all children with Hib vaccine failure who participated in this study. The authors are also grateful to Dr Nick Andrews, Departmental Statistician at the Centre for Infections, Health Protection Agency, for helping with the statistical analyses. SL was awarded a two-year European Society for Paediatric Infectious Diseases (ESPID) to complete this study. AJP is a Jenner Institute Investigator and is funded by the NIHR Oxford Biomedical Research Centre. References [1] Swingler G, Fransman D, Hussey G. Conjugate vaccines for preventing Haemophilus influenzae type B infections. Cochrane Database Syst Rev 2007. CD001729. [2] Booy R, Hodgson S, Carpenter L, Mayon-White RT, Slack MP, Macfarlane JA, et al. Efficacy of Haemophilus influenzae type b conjugate vaccine PRP-T. Lancet 1994;344:362–6. [3] Booy R, Taylor SA, Dobson SR, Isaacs D, Sleight G, Aitken S, et al. Immunogenicity and safety of PRP-T conjugate vaccine given according to the British accelerated immunisation schedule. Arch Dis Child 1992;67:475–8. [4] Heath PT, McVernon J. The UK Hib vaccine experience. Arch Dis Child 2002;86:396–9. [5] Ladhani S, Slack MP, Heys M, White J, Ramsay ME. Fall in Haemophilus influenzae serotype b (Hib) disease following implementation of a booster campaign. Arch Dis Child 2008;93:665–9. [6] Heath PT, Booy R, Azzopardi HJ, Slack MP, Bowen-Morris J, Griffiths H, et al. Antibody concentration and clinical protection after Hib conjugate vaccination in the United Kingdom. JAMA 2000;284:2334–40. [7] Booy R, Heath PT, Slack MP, Begg N, Moxon ER. Vaccine failures after primary immunisation with Haemophilus influenzae type-b conjugate vaccine without booster. Lancet 1997;349:1197–202. [8] Heath PT, Booy R, Griffiths H, Clutterbuck E, Azzopardi HJ, Slack MP, 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. [9] MacFaul R, Werneke U. Recent trends in hospital use by children in England. Arch Dis Child 2001;85:203–7. [10] McVernon J, Andrews N, Slack M, Moxon R, Ramsay M. Host and environmental factors associated with Hib in England, 1998–2002. Arch Dis Child 2008;93:670–5.
2200
S. Ladhani et al. / Vaccine 28 (2010) 2195–2200
[11] McVernon J, Ramsay ME, McLean AR. Understanding the impact of Hib conjugate vaccine on transmission, immunity and disease in the United Kingdom. Epidemiol Infect 2008;136:800–12. [12] Takala AK, Eskola J, Palmgren J, Ronnberg PR, Kela E, Rekola P, et al. Risk factors of invasive Haemophilus influenzae type b disease among children in Finland. J Pediatr 1989;115:694–701. [13] Silfverdal SA, Bodin L, Hugosson S, Garpenholt O, Werner B, Esbjorner E, et al. Protective effect of breastfeeding on invasive Haemophilus influenzae infection: a case-control study in Swedish preschool children. Int J Epidemiol 1997;26:443–50. [14] Wolff MC, Moulton LH, Newcomer W, Reid R, Santosham M. A case-control study of risk factors for Haemophilus influenzae type B disease in Navajo children. Am J Trop Med Hyg 1999;60:263–6. [15] Istre GR, Conner JS, Broome CV, Hightower A, Hopkins RS. Risk factors for primary invasive Haemophilus influenzae disease: increased risk from day care attendance and school-aged household members. J Pediatr 1985;106: 190–5. [16] Zielinski A, Kwon CB, Tomaszunas-Blaszczyk J, Magdzik W, Bennett JV. Risk of Haemophilus influenzae type b meningitis in Polish children varies directly with number of siblings: possible implications for vaccination strategies. Eur J Epidemiol 2003;18:917–22. [17] Cochi SL, Fleming DW, Hightower AW, Limpakarnjanarat K, Facklam RR, Smith JD, et al. Primary invasive Haemophilus influenzae type b disease: a populationbased assessment of risk factors. J Pediatr 1986;108:887–96.
[18] Vadheim CM, Greenberg DP, Bordenave N, Ziontz L, Christenson P, Waterman SH, et al. Risk factors for invasive Haemophilus influenzae type b in Los Angeles County children 18–60 months of age. Am J Epidemiol 1992;136:221–35. [19] Broome CV. Epidemiology of Haemophilus influenzae type b infections in the United States. Pediatr Infect Dis J 1987;6:779–82. [20] Fogarty J, Moloney AC, Newell JB. The epidemiology of Haemophilus influenzae type b disease in the Republic of Ireland. Epidemiol Infect 1995;114:451–63. [21] Redmond SR, Pichichero ME. Hemophilus influenzae type b disease. An epidemiologic study with special reference to day-care centers. JAMA 1984;252:2581–4. [22] Olowokure B, Spencer NJ, Hawker JI, Blair I, Smith RL. Changing socioeconomic risk factors for invasive H. influenzae disease after the introduction of conjugate vaccine. J Infect 2003;46:46–8. [23] Macintyre S, Ellaway A, Der G, Ford G, Hunt K. Do housing tenure and car access predict health because they are simply markers of income or self esteem? A Scottish study. J Epidemiol Community Health 1998;52:657–64. [24] Booy R, Hodgson SA, Slack MP, Anderson EC, Mayon-White RT, Moxon ER. Invasive Haemophilus influenzae type b disease in the Oxford region (1985–91). Arch Dis Child 1993;69:225–8. [25] Fortnum HM. Hearing impairment after bacterial meningitis: a review. Arch Dis Child 1992;67:1128–33. [26] Grimwood K, Anderson VA, Bond L, Catroppa C, Hore RL, Keir EH, et al. Adverse outcomes of bacterial meningitis in school-age survivors. Pediatrics 1995;95:646–56.