WENGER, JAY D. MD
Haemophilus influenzae infections have been the major cause of invasive bacterial disease in children in the United States and Canada for several decades. During the 1970s and 1980s a series of studies characterized the burden of disease in the United States.1 Population-based surveillance studies in a variety of locations demonstrated rates of invasive disease (defined as infection with growth of Hib from a normally sterile site, such as blood or cerebrospinal fluid) of between 40 and 100 cases per 100 000 children <5 years old. One child of 200 contracted Hib disease by 5 years of age.2 Peak rates of disease occurred in children 6 to 11 months of age,3 with 60 to 70% of all invasive disease in children occurring in those <18 months of age (Fig. 1).
Hib disease presented primarily as meningitis (up to 60% of all cases) and epiglottitis (15 to 30%), with the remainder presenting as arthritis, facial cellulitis and bacteremia. From 3 to 6% of all children with invasive disease died from their infection, and of the survivors of meningitis up to 25% suffered long term sequelae such as mental retardation or, more commonly, hearing loss. The overall cost of disease to the US health system was estimated to be more than $500 million per year.
Studies performed to evaluate potentially modifiable risk factors showed a variety of associations with disease. Daycare attendance, lower socioeconomic status (including low income, low parental educational levels and household crowding), number of children in a family and black race were associated with risk of disease.4-8 Breast-feeding occasionally was identified as a protective factor. Which of these factors actually were contributing to disease and which were confounders was not clear, however, because most of these studies presented results of univariate analyses only. Cochi et al.9 in 1983 and 1984 evaluated risk factors for Hib disease in metropolitan Atlanta and, using multivariate analysis of a case-control study, demonstrated that when controlling for day care, race, breast-feeding, income and socioeconomic status, crowding and day-care attendance remained the strongest predictors of risk for infection, with the highest risk among the youngest children. In addition household crowding was an independent, although less strong risk factor for infection. Breast-feeding was protective among the youngest age group, but protection did not extend beyond 6 months in this study.
None of these risk factors proved to be amenable to substantial change through routine public health measures. Breast-feeding already had been supported for many years by public health recommendations in the US, and further major improvements in breast-feeding practice were unlikely to occur. There was no practical way to reduce either household crowding or day-care attendance. Efforts were made to identify the factors in attending day care that contributed to disease transmission. Istre et al.8 found that disease was more likely to occur in day-care centers with larger classes than in those with fewer children. A study that compared day-care centers in which cases of Hib disease occurred with centers in which a case had not occurred in the recent past found that reuse of towels or handkerchiefs to wipe noses and admission of children with diarrhea or who were not toilet-trained was associated with disease,10 but recommendations against most of these practices already had been made, and additional efforts in this direction were not likely to be successful. Thus although these studies demonstrated a number of practices with pathophysiologically plausible associations with disease, they did not lead to practical public health recommendations that reduced disease burden substantially. Hope for reduction in disease burden clearly lay in development of vaccines for use in children.
IMPACT OF IMMUNIZATION
Hib vaccines were first introduced in North America in the mid-1980s. Data on the impact of these vaccines are available from both the US and Canada and will be presented separately. For each country the timing of introduction of each type of Hib vaccine, the surveillance systems used and the surveillance findings will be reviewed.
First generation Hib vaccines comprised of the capsular polysaccharide polyribosylribitol phosphate (PRP) were introduced in the US in mid-1985. Although recommendations for routine use for children at 2 years of age (and at 18 months if children were in day care) were published by both ACIP and AAP, coverage with this vaccine was suboptimal and in many areas never exceeded 35%. The first PRP-protein conjugate vaccine, PRP-D, was licensed in December, 1987, for use in all children at 18 months of age; it came into wide use in 1988. Coverage during this period (1988 to 1990) ranged from 20 to 70%, depending on the age group evaluated.11, 12 HbOC (also termed PRP-CRM) was licensed in October, 1990, and was followed shortly thereafter by PRP-OMP. Both were recommended for use in infants starting at 2 months of age, and widespread infant immunization occurred during 1991. However, it took several years to achieve immunization coverage similar to that of diphtheria-tetanus-pertussis vaccine, and it was not until 1995 that coverage was estimated to have reached 90%.13
Several surveillance systems were available to evaluate the impact of Hib vaccine introduction. The National Bacterial Meningitis and Bacteremia Reporting System was a passive surveillance system developed by the National Center for Infectious Diseases at the CDC in Atlanta.14 It was started in 1977 to collect case reports of episodes of bacterial meningitis and included information on case demographics, organism characteristics, disease syndromes and outcome. Between 1980 and 1991, 20 states consistently reported data, and these states form the basis of the data included in this report. Although the data system was slow and reporting was insensitive (∼40 to 50% of all meningitis cases were reported), the system provided a relatively stable baseline against which temporal changes could be assessed.
In late 1988 the National Center for Infectious Diseases developed a system of active, laboratory-based surveillance projects to supplement the passive surveillance system.15 In each area, state or locality investigators performed population-based surveillance. The population of each surveillance area was between 2 and 5 million persons. Surveillance included routine contact with all hospital laboratories in the surveillance areas to identify all positive sterile-site cultures of Hib and other H. influenzae strains. Extensive data, similar to those collected for the national passive system, were collected for each case. Isolates were typed and usually sent to CDC for confirmation or additional analysis. Laboratory records were audited routinely to ensure complete reporting of all cases. Thus this system, although expensive, provided the most accurate, complete data on Hib disease. Its primary drawbacks were the cost and the relatively small population under surveillance (between 10 and 15 million persons each year).
In addition to the national disease surveillance systems noted above, the National Center for Health Statistics collects data on hospitalization and causes of death.16 Although the time required to collect and collate these data make this system untimely, it may be used to evaluate major trends in disease activity, with a lag period of 2 to 3 years.
Figure 2 demonstrates the striking decline in H. influenzae meningitis in children <5 years of age after distribution of the conjugate Hib vaccines (and not, as can be seen, after introduction of unconjugated PRP vaccine), based on data from the National Bacterial Meningitis and Bacteremia Reporting System.15 The decline began soon after widespread use of vaccine for children 18 months and older in 1988, with decreased rates noted in 1989. Between 1980 and 1991, during which the decline in meningitis occurred, no change occurred in meningitis rates recorded in this system for children 12 years of age and older. Figure 3 evaluates further the decline in young children by stratifying rates by the age of cases. There was no substantial effect of unconjugated PRP vaccine on any age group, although a slight decline may have occurred in children ages 2 to 4 years. However, licensure of PRP-D in 1987 and its wide use in 1988 was followed by declines in incidence not only in the 1-year-old children and those ages 2 to 4 years, some of whom were receiving the vaccine, but also in children <1 year old. Substantial declines were seen in 1989 and 1990, before conjugate vaccines were used in this age group. Declines in all age groups continued after immunization of infants began in 1991.
One reason for the impressive effect of the vaccine in populations of children, some of whom had not received the vaccine but nonetheless were protected (herd immunity), was first identified in Finland by Takala et al.,17 where a series of carriage studies documented disappearance of the organism from throat cultures among a well-vaccinated population. Murphy et al.18 showed that vaccinated children in day care were less likely to carry Hib than unvaccinated classmates. In a convenience sample of children in metropolitan Atlanta, where immunization coverage was 75%, Mohle-Boetani et al.19 found a carriage rate of 0.2%, contrasting sharply with previous studies showing carriage of 3 to 7% in general pediatric populations.
Active surveillance data confirmed that Hib disease declined rapidly from 1989 through 1991 (Fig. 4).13 However, it also was suggested that the rate of decline slowed after 1991. Through 1995 Hib disease still occurred in the surveillance areas at very low levels. A case-control study suggested that most of the children who now were contracting Hib disease were underprivileged and undervaccinated.20 The active surveillance data also demonstrated that there was no increase in non-Hib H. influenzae disease, as had been feared might occur if the ecologic niche previously occupied by Hib in the oropharynx were now to be occupied by a similar pathogen.13
Analysis of national mortality data confirmed several implications of the above work. Although mortality from pneumococcal, meningococcal and Hib meningitis was declining slowly during the 1980s (in the face of stable national hospitalization rates), mortality associated with Hib began to decline steeply in 1989, 1 year after widespread use of conjugate vaccine began in 18-month-old children.16 Also, although mortality declined dramatically both for children ages 1 to 4 years and for those younger than 1 year old after introduction of conjugate Hib vaccine for children 1 to 4 years old, it declined more rapidly for the 1- to 4-year-old children, the immunization target population, than for children <1 year of age.
In Canada PRP vaccine was in use between 1986 and 1988 for children ages 2 years and older; PRP-D was used between 1988 and 1992 for children 18 months and older; and PRP-T and HbOC have been used for infants since 1992. Rates of Hib disease from passive national reporting are shown in Figure 5.21 Similar to the pattern seen in national US data, unconjugated PRP vaccine had no effect on disease, and the major decline in disease incidence occurred during the time PRP-D was used in older infants. Introduction of infant immunization with HbOC and PRP-T continued this downward trend.
To evaluate more accurately the impact of immunization programs, a consortium of major Canadian pediatric hospitals, the Immunization Monitoring Program, Active (IM-PACT), identified the number of Hib cases admitted since 1985, using a combination of retrospective laboratory and discharge diagnosis reviews and prospective laboratory-based surveillance.22 These data show a continuous decline in number of cases since 1985, with most of the decrease occurring during the years of conjugate vaccine use, beginning in 1988. Figure 6, an age-stratified presentation of the number of cases of Hib disease since 1991 identified in the IMPACT system, shows the result of implementing the infant immunization program. In 1991 and 1992, although disease in each age group was declining year by year, the largest number of cases continued to be in the 6- to 18-month-old age group. However, in 1993, within 1 year after introduction of infant immunization, the number of cases in this age group dropped below that of any other age group. Similar to the US study these data also show a leveling off of the rate of decline in 1994 and 1995 as disease reached very low levels.
Hib disease, once the most common cause of bacterial meningitis in the US and Canada, has been brought under control with unexpected rapidity by use of the Hib conjugate vaccines. This is in part because of the reduction of nasopharyngeal carriage induced by widespread immunization. Although precise coverage data are not available, this effect occurred in a totally unimmunized age group, children <18 months of age, even though <70% of the 18- to 59-month-old target age group, and perhaps as little as 40 to 50%, were immunized. The postlicensure surveillance information also has laid to rest fears that "replacement" disease would occur with other strains of H. influenzae.
Despite these successes data from both the US and Canada show that a small amount of Hib disease still is occurring, primarily in underprivileged populations with inadequate immunization coverage. Hib disease has been eliminated in countries in which high immunization rates with Hib conjugate vaccines have been achieved. Additional efforts to immunize underserved populations are necessary in North America.
1. Ward J, Lieberman JM, Cochi SL. Haemophilus influenzae
vaccines. In: Plotkin SA, Mortimer EA, eds. Vaccines. 2nd ed. Philadelphia: Saunders, 1994.
2. Broome CV. Epidemiology of Haemophilus influenzae
type b infections in the United States. Pediatr Infect Dis J 1987;6:779-82.
3. Wenger JD, Hightower AW, Facklam RR, Gaventa S, Broome CV, Bacterial Meningitis Study Group. Bacterial meningitis in the United States, 1986: report of a multistate surveillance study. J Infect Dis 1991;162:1316-23.
4. Granoff DM, Basden M. Haemophilus influenzae
infections in Fresno County, California: a prospective study of the effects of age, race, and contact with a case on incidence of disease. J Infect Dis 1980;141:40-6.
5. Parke JC Jr, Schneerson R, Robbins JB. The attack rate, age incidence, racial distribution, and case fatality rate of Haemophilus influenzae
type b meningitis in Mecklenberg County, North Carolina. J Pediatr 1972;81:765-9.
6. Redmond SR, Pichichero ME. Haemophilus influenzae
type b disease: an epidemiologic study with special reference to day-care centers. JAMA 1984;252:2581-4.
7. Ward JI, Lum MKW, Hall DB, et al. Invasive H. influenzae
type b disease in Alaska: background epidemiology for a vaccine efficacy trial. J Infect Dis 1986;108:887-96.
8. Istre GR, Conner JS, Broome CV, Hightower AW, 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.
9. Cochi SL, Fleming DW, Hightower AW, et al. Primary invasive Haemophilus influenzae
type b disease: a population-based assessment of risk factors. J Pediatr 1986;108:887-6.
10. Wenger JD, Harrison LH, Hightower AW, Broome CV, Haemophilus influenzae
study group: day care characteristics associated with Haemophilus influenzae
disease. Am J Public Health 1990;80:1455-8.
11. Vadheim CM, Greenberg DP, Eriksen E, et al. Eradication of Haemophilus influenzae
type b disease in Southern California. Arch Pediatr Adolesc Med 1994;148:51-6.
12. Wenger JD, Pierce R, Deaver KA, Plikaytis BD, Facklam RR, Broome CV. Efficacy of Haemophilus influenzae
type b polysaccharide-diphtheria toxoid conjugate vaccine in US children aged 18-59 months. Lancet 1991;338:395-8.
13. Centers for Disease Control and Prevention. Progress toward elimination of Haemophilus influenzae
type b disease among infants and children: United States, 1987-1995. MMWR 1996;45:901-6.
14. Schlech WF, Ward JI, Band JD, Hightower AW, Fraser DW, Broome CV. Bacterial meningitis in the United States, 1978 through 1981: the National Bacterial Meningitis Surveillance Study. JAMA 1985;253:1749-54.
15. Adams WG, Deaver KA, Cochi SL, et al. Decline of childhood Haemophilus influenzae
type b (Hib) disease in the Hib vaccine era. JAMA 1993;269:221-6.
16. Schoendorf KC, Adams WG, Kiely JL, Wenger JD. National trends in Haemophilus influenzae
meningitis mortality and hospitalization among children, 1980 through 1991. Pediatrics 1994;93:663-8.
17. Takala AK, Eskola J, Leinonen M, et al. Reduction of oropharyngeal carriage of Haemophilus influenzae
type b (Hib) in children immunized with Hib conjugate vaccine. J Infect Dis 1991;164:982-6.
18. Murphy TV, Pastor P, Medley R, Osterholm MT, Granoff DM. Decreased Haemophilus
colonization in children vaccinated with Haemophilus influenzae
type b conjugate vaccine. J Pediatr 1993;122:517-23.
19. Mohle-Boetani JC, Ajello G, Breneman E, et al. Carriage of Haemophilus influenzae
type b in children after widespread vaccination with Haemophilus influenzae
type b vaccines. Pediatr Infect Dis J 1993;12:589-93.
20. Jafari H, Adams W, Deaver KA, et al. Efficacy of Haemophilus influenzae
type b (Hib) conjugate vaccines, risk factors for disease and under-vaccination in the United States [Abstract G-12]. Presented at the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, CA, September 17 to 20, 1995.
21. Laboratory Centre for Disease Control. Notifiable diseases annual summary: Supplement, 1994. Can Comm Dis Rep 1996;22S2:63-4.
22. Immunization monitoring program, active (IMPACT) of the Canadian Paediatric Society and the Laboratory Centre for Disease Control. Recent trends in pediatric Haemophilus influenzae
type b infections in Canada. Can Med Assoc J 1996;154:1041-7.
FIRST INTERNATIONAL CONFERENCE ON HAEMOPHILUS INFLUENZAE TYPE b INFECTION IN ASIA
The Editors thank the Association pur l'Aide à la Médicine Préventive, the Foundation Mérieux, and the World Health Organization for supporting publication of these proceedsings, and Jennifer Wells for her editorial assistance.
© Williams & Wilkins 1998. All Rights Reserved.