Pediatric Haemophilus Influenzae Infection

Updated: Apr 04, 2016
  • Author: Mobeen H Rathore, MD, CPE, FAAP, FIDSA; Chief Editor: Russell W Steele, MD  more...
  • Print


Although the type of infectious diseases caused by Haemophilus influenzae has changed considerably in recent years because of the widespread and routine immunization of children against type b organisms, H influenzae remains a significant pathogen. The recent vaccine shortage in the United States from December 2007 to September 2009 is a reminder that despite the decline in the incidence of invasive disease, Hib immunization remains critical in the control of the disease. [1]

First isolated in 1892 by Robert Pfeiffer from the sputum of patients with pandemic influenza infection, H influenzae was thought to be the infectious agent responsible for flu. In 1920, the organism was named Haemophilus influenzae (from the Greek haemophilus, meaning "blood-loving") to reflect the fastidious growth requirement of the organism, as well as its apparent association with influenza. In 1933, the discovery of the viral etiology of influenza eventually refuted this erroneous association. Nevertheless, subsequent findings revealed that H influenzae was responsible for a wide spectrum of clinical diseases.

In the 1930s, Margaret Pittman defined 2 major categories of H influenzae: the unencapsulated strains and the encapsulated strains. The unencapsulated strains were chiefly responsible for infections at mucosal surfaces, including otitis media, conjunctivitis, bronchitis, and sinusitis. In contrast, one of the 6 antigenically distinct encapsulated strains, strain type b, was associated with invasive diseases (eg, septicemia, meningitis, cellulitis, septic arthritis, epiglottitis, pneumonia).

Prior to the availability of an effective vaccine, H influenzae type b (Hib) was the most common cause of pediatric bacterial meningitis in the United States.




A major virulence factor of H influenzae is its polysaccharide capsule, which plays a central role in molecular pathogenesis and the immune response. Six antigenically and biochemically distinct capsular polysaccharide subtypes (a-f) have been identified (see Laboratory Studies). Although, historically, type b encapsulated strains have been of primary clinical and immunologic importance because of their association with invasive infection, including meningitis, the other encapsulated strains also can cause invasive disease.

The type b capsular polysaccharide is well characterized at the molecular level. It is composed of repeating heteropolymers of ribosyl and ribitol phosphate. Rapid diagnostic latex agglutination tests are available for the identification of this polyribosyl ribitol phosphate (PRP) after its release in infected patients. This polysaccharide structure is unique to type; the other capsular Haemophilus serotypes are composed of hexose rather than pentose sugars.

Another important component of the H influenzae cell wall that contributes to pathogenesis is lipopolysaccharide (LPS). Although chemically different from the LPS of the Enterobacteriaceae, the biological activity of Hib LPS appears to be similar to that of other gram-negative endotoxins. Finally, numerous outer membrane proteins have recently been identified as important components of pathogenesis and immunity. Pili (or fimbriae), fibrils, and a protein called Hia mediate adherence of Hib to cells of the human respiratory tract. The multiple adhesins target specific cells of the airway and provide redundancy for adherence to respiratory tissues. H influenzae encodes 3 distinct immunoglobulin A (IgA) proteases that may be involved as virulence factors by interfering with host mucosal defenses.

Because the entire genome of a laboratory strain of H influenzae has been sequenced, additional insights into its molecular mechanisms of pathogenesis will undoubtedly be gained in the near future.

Age-related susceptibility and immune response

In pioneering experiments in the 1930s, Fothergill and Wright demonstrated that blood obtained from children aged 3 months to 3 years lacked bactericidal activity against type b strains, whereas the blood of neonates, older children, and adults was bactericidal. Eventually, this effect was shown to depend on the presence of a specific antibody against the type b capsule. In 1942, Alexander proposed that the polyribosyl ribitol capsular polysaccharide was intrinsically antiphagocytic and that efficient ingestion by phagocytes was facilitated by opsonization with type-specific antibodies.

Subsequent evidence has demonstrated that antibody to PRP protects individuals against invasive disease. It activates complement, is opsonophagocytic, and protects animals from experimental challenge. Indeed, passive transfer of type b antiserum was shown to be effective in the treatment of Hib disease in the preantibiotic era. Ultimately, the most compelling evidence for the protective properties of anticapsular antibody is the observation that purified PRP vaccines confer protection. Induction of the antibody to the type b capsular polysaccharide is the means by which all Hib vaccines result in protection.

The age-dependent susceptibility to infections is correlated with the age-dependent nature of the immune response to Hib capsular polysaccharide. When infants are at maximal risk of infection (ie, at the nadir of protective transplacental immunity), their serum anti-PRP antibody levels are low or absent. Even after they recover from illness, their antibody levels are low and their immune responses are poor. As a consequence, second or third episodes of invasive Hib disease are described, and a previous episode of invasive infection does not obviate Hib immunization. Infants' failure to make serum anti-PRP antibodies is typical of the natural delay in the development of their immune response to polysaccharide antigens.

PRP stimulates B cells but does not adequately activate macrophages and appropriate helper T cells. Therefore, it is considered to be a T-cell–independent antigen. Characteristics of T-cell–independent antigens include the following: immune responses to them are limited, particularly in young infants; no booster response occurs with repeated antigenic stimulation; and the antibodies have low affinity to the antigen and mostly consist of immunoglobulin M (IgM). The development of a Hib vaccine that was more immunogenic and protective in young infants required the conversion of PRP from a T-cell–independent antigen to a T-cell–dependent antigen; the principles of carrier-hapten linkage were used to accomplish this.

The role of other host immune responses in limiting Hib disease is poorly understood. PRP-specific IgA antibodies have been described. This finding suggests a possible role for secretory antibodies that can block Hib attachment to the respiratory tract mucosa, but the clinical relevance of this proposed mechanism remains to be demonstrated. Complement also appears to be involved in the host defense against Hib. Both encapsulated and unencapsulated strains of H influenzae activate the classical and alternative complement pathways in vitro. The clinical correlates of this observation are the results suggesting that individuals with C2, C3, and C4b deficiencies are more susceptible to invasive Hib disease.

Antimicrobial resistance

Another clinically important aspect of the molecular microbiology of Hib is the identification of the genes responsible for its antimicrobial resistance. Resistance to ampicillin has become extremely common; 5-50% of isolates in various parts of the world are resistant. The mechanism of resistance typically is due to the presence of a plasmid-encoded enzyme, beta-lactamase. Plasmid-encoded resistance to chloramphenicol, tetracycline, and sulfonamides can exist independently; in some countries, multiple resistance determinants reside on the same plasmid. Chromosomally mediated antibiotic resistance, due to the accumulation of point mutations, most commonly involves amoxicillin. This resistance is low level and can coexist with plasmid-mediated resistance. Trimethoprim resistance most commonly is due to mutations in chromosomal dihydrofolate reductase. Susceptibility testing should be performed with all isolates identified in invasive infections.

Transmission and infection

Humans are the only natural host for H influenzae. Therefore, maintenance of the organism in the human population depends on person-to-person transmission, which efficiently occurs via respiratory droplet spread. Although both nontypeable strains and Hib easily are spread via person-to-person transmission, Hib strains have historically been associated with invasive disease in children. Before effective vaccines were available, nasopharyngeal acquisition of Hib occurred in most children aged 5 years or younger. Although nasopharyngeal Hib colonization may not produce symptoms, breakthrough bacteremia with subsequent focal infection was common at one time, and it was a major public health problem in children in the United States.

The molecular determinants responsible for the nasopharyngeal colonization and subsequent bacteremic invasiveness of Hib remain poorly understood. Invasive disease requires the spread of bacteria from the upper respiratory tract to the bloodstream and, subsequently, to other body sites. The organism must first colonize the respiratory mucosal epithelium, and several bacterial surface proteins appear to play an important role in the attachment process. The organism then invades the nasal mucosa. The exact mode of entry of the organism into the blood vessel is unknown; it may enter via the lymphatics. The size of the bacterial inoculum and the intercurrent presence of a viral respiratory tract infection potentiate the risk of invasive disease.

It has been suggested that H influenzae can invade and enter respiratory epithelial cells by means of transcytosis. Strains that are able to resist lysis by the complement system (ie, those with a capsule) or opsonophagocytosis (due to the lack of a natural antibody) then can replicate in the bloodstream, causing invasive disease. As organisms divide, bacteremia increases steadily over hours. When bacteremia occurs, its magnitude and duration are determined by the dynamics of bacterial proliferation and clearance by anti-PRP antibodies and phagocytes. When the bacterial concentration exceeds 104 organisms per milliliter, metastatic seeding occurs, especially to the meninges via the choroid plexus. Although the meninges are involved in more than half of recognized cases of invasive Hib disease, other potential metastatic sites include the lungs, joint synovium, pleura, peritoneum, and pericardium.

Noninvasive or mucosal infections are much more common than invasive disease, particularly in the postvaccine era. These generally are due to nontypeable strains of H influenzae, and bacteremia seldom is present. Therefore, these infections are presumed to represent extensions of H influenzae from the respiratory mucosa to contiguous body sites. Noninvasive infections include otitis media, sinusitis, bronchitis, and pneumonia. Local extension of nontypeable H influenzae can occur via the eustachian tubes, bronchi, or sinus passages. Disease is more likely if normal clearance mechanisms or immune function is impaired, for example, after viral infection, sinus obstruction, or eustachian tube dysfunction.

Risk factors and epidemiology

Chronic illnesses associated with an increased risk of invasive Hib disease include sickle-cell disease, asplenia, agammaglobulinemia, Hodgkin disease, and complement deficiencies. Risk is also increased in those who attend daycare and in those who have young siblings, a crowded household, a lower socioeconomic status, or exposure to cigarette smoke. Breastfeeding confers some protection against disease.

The epidemiology of invasive Hib disease has changed dramatically in recent years because of the widespread use of conjugate vaccines. In 1987, the first Hib vaccine (purified PRP) was licensed in the United States for use in children aged 18 months and older. Over the next few years, the incidence of invasive disease dramatically decreased in older children. However, because Hib meningitis was a greater problem in infants younger than 1 year, the most significant decline was not observed until late 1990, when protein-PRP conjugate vaccines were approved for use in infants aged 2 months or older. In populations with high rates of vaccination, the incidence of Hib disease has decreased by more than 95%. While the vaccine shortage between 2007-2009 did not result in an overall increased incidence of Hib invasive disease, cases in infants whose vaccinations were deferred were reported. [2] The trends in Hib diseases epidemiology need closemonitoring.

The protective effectiveness of these vaccines exceeded initial expectations. This result was attributed to their direct effect on nasopharyngeal carriage, which ultimately decreased the environmental burden of Hib and protected even unimmunized children because of the effect of herd immunity. The conjugate vaccines are so effective in preventing Hib infection that the finding of invasive disease in a fully immunized child should suggest the possibility of an underlying immunodeficiency and prompt further diagnostic evaluation.

An important epidemiologic aspect of Hib disease is the risk that it poses to contacts of the affected person. Although the direct contagiousness of invasive Hib infection is limited, household contacts have a significant risk of secondary disease, particularly in the 30 days after their exposure to the index patient. This risk is related to droplet spread under conditions of continuous household exposure. Colonization rates of more than 70% have been noted after exposure in closed populations, such as those in families or daycare centers. This is the rationale for chemoprophylaxis after exposure to invasive Hib disease.

Another less common but recently recognized route of exposure is the vertical transmission of H influenzae via the maternal birth canal. In recent years, this phenomenon has been evidenced by an increase in cases of neonatal bacteremia and meningitis caused by nontypeable strains acquired from the mother's genital tract. These strains are genetically distinct from those that colonize the upper respiratory tract.

A bimodal seasonal disease pattern has been described; one peak occurs in autumn between September and December, and a second occurs in spring between March and May. Historically, invasive Hib infection has been uncommon in adults (apparently because of the gradual development of protective antibodies over time in the context of asymptomatic nasopharyngeal colonization), but Hib occasionally can cause invasive infection in adults. Remarkably, in the postvaccine era, Hib meningitis is more common in adults than in children. The effect of routine childhood Hib immunization on the epidemiologic characteristics of adult infections, if any, remains to be seen.




United States

Nontypeable strains colonize the upper respiratory tract in as many as three fourths of healthy adults. Hib strains colonize the nasopharynx of 3-5% of children; the effectiveness of vaccines is related, in part, to their ability to dramatically diminish the prevalence of nasopharyngeal colonization.

In the prevaccine era, invasive Hib disease had a characteristic and striking age-related prevalence; approximately 85% of cases occurred in children younger than 5 years. The peak prevalence of the most serious form of invasive disease, meningitis, occurred in infants aged 6-12 months. Hib epiglottitis was, in contrast, predominantly a disease of older children, with more than 80% of the infections occurring in children older than 2 years.

In the prevaccine era, approximately 20,000 instances of invasive Hib disease occurred annually in the United States, affecting approximately 1 child in every 200 younger than 5 years. [3] Certain high-risk groups, including Navajo and Alaskan Native Americans, had an even higher incidence of disease.


Hib continues to be a major cause of sepsis and meningitis in the developing world, where resources for widespread immunization programs are not available. In addition, a nonserotypeable H influenzae biogroup III (which is identical to the Haemophilus aegyptius group) has been shown to be the cause of a disease called Brazilian purpuric fever (BPF), which was discovered in children in southern Brazil.

A population-based observational study in England found the incidence of H influenzae meningitis per 100,000 children decreased from 6·72 admissions in 1992 to 0·39 admissions in 1994, after the introduction of routine H influenzae type b vaccination. The study also saw a small rise in admissions in the early 2000s which decreased by 2008 after the introduction of catch-up (2003) and routine (2006) booster programs for young children. [4]


Even with prompt diagnosis and supportive care, the mortality rate of Hib meningitis is approximately 5%. Long-term sequelae occur in 15-30% of survivors and include sensorineural hearing loss, language disorders, mental retardation, and developmental disorders. In epiglottis, the mortality rate of 5-10% invariably is related to poor early airway control.


In the prevaccine era, invasive Hib disease had a characteristic and striking age-related prevalence; approximately 85% of the cases occurred in children younger than 5 years. Hib was the most common cause of pediatric bacterial meningitis, which had a peak prevalence in infants aged 6-12 months. More than 80% of cases of Hib epiglottitis occurred in children older than 2 years. Hib was the leading cause of septic arthritis in children younger than 2 years. Currently, epiglottitis occurs primarily in older children (aged 2-7 y). The vast majority of cases of cellulitis occur in children aged 2 years or younger.