General Characteristics

The genus Haemophilus contains significant genetic diversity. Members of the genus are small, nontitle, pleomorphic gram-negative bacilli. The cells are typically coccobacilli or short rods. Species of the genus Haemophilus require protoporphyrin IX (a metabolic intermediate of the hemin biosynthetic pathway) referred to as X factor and the V factor, nicotine adenine dinucleotide (NAD) or NADP for in vitro growth. Haemophilus are facultative anaerobes enhanced in a 5% to 7% CO₂-enriched atmosphere. The morphologic and physiologic features of individual species are presented in the discussion of laboratory diagnosis. Aggregatibacter aphrophilus and Aggregatibacter paraphrophilus have been reclassified as a single species based on their multilocus sequence analysis (A. aphrophilus).

Epidemiology

As presented in Table 32-1, except for Haemophilus ducreyi, Haemophilus spp. normally inhabit the upper respiratory tract of humans. Asymptomatic colonization with H. influenzae type b is rare. Although H. ducreyi is only found in humans, the organism is not part of our normal flora, and its presence in clinical specimens indicates infection.

Among H. influenzae strains, there are two broad categories: typeable and nontypeable (NTHi). Strains are typed based on capsular characteristics. The capsule is composed of a sugar-alcohol phosphate (i.e., polyribitol phosphate) complex. Differences in this complex are the basis for separating encapsulated strains into one of six groups: type a, b, c, d, e, or f. H. influenzae type b (Hib) is most commonly encountered in serious infections in humans. Nontypeable strains do not produce a capsule and are most commonly encountered as normal inhabitants of the upper respiratory tract.

Although person-to-person transmission plays a key role in infections caused by Haemophilus influenzae and H. ducreyi, infections caused by other Haemophilus strains and species likely arise endogenously as a person’s own flora gains access to a normally sterile site. The colonizing organism invades the mucosa and enters the patient’s bloodstream. Encapsulated strains are protected from clearance from host phagocytes. Once in the circulation, the organism is able to spread to additional sites and tissues including the lungs, pericardium, pleura, and meninges.

Pathogenesis and Spectrum of Disease

Production of a capsule and factors that mediate bacterial attachment to human epithelial cells are the primary virulence factors associated with Haemophilus spp. In general, infections caused by Haemophilus influenzae are often systemic and life threatening, whereas infections caused by nontypeable (do not have a capsule) strains are usually localized (Table 32-2). The majority of serious infections caused by H. influenzae type b are typically biotypes I and II. The development and use of the conjugate vaccine in children since 1993 has reduced the infection rate by 95% in children younger than 5 years old in the United States.

The majority of H. influenzae infections are now caused by nontypeable strains (NTHi). Transmission is often via respiratory secretions. The organism is able to gain access to sterile sites from colonization in the upper respiratory tract. Clinical infections include otitis media (ear infection), sinusitis, bronchitis, pneumonia, and conjunctivitis. Immunodeficiencies and chronic respiratory problems such as chronic obstructive pulmonary disease may predispose an individual to infection with NTHi.

Chancroid is the sexually transmitted disease caused by H. ducreyi (see Table 32-2). The initial symptom is the development of a painful genital ulcer and inguinal lymphadenopathy. Although small outbreaks of this disease have occurred in the United States, this disease is more common among socioeconomically disadvantaged populations inhabiting tropical environments. Epidemics of disease are associated with poor hygiene, prostitution, drug abuse, and poor socioeconomic conditions.

Haemophilus spp. can be isolated from most clinical specimens. The collection and transport of these specimens are outlined in Table 5-1, with emphasis on the following points. First, Haemophilus spp. are susceptible to drying and temperature extremes. Therefore, specimens suspected of containing these organisms should be inoculated to the appropriate media immediately. Specimens susceptible to contamination with normal flora such as a lower respiratory specimen should be collected by bronchioalveolar lavage. In cases of pneumonia or cerebrospinal fluid (CSF) infection or suspected infection of any other normally sterile body fluid, blood cultures should also be collected.

Second, the recovery of H. ducreyi from genital ulcers requires special processing. The ulcer should be cleaned with sterile gauze moistened with sterile saline. A cotton swab moistened with phosphate-buffered saline is then used to collect material from the base of the ulcer. To maximize the chance for recovering the organism, the swab must be plated to special selective media within 10 minutes of collection.

Direct Detection Methods

Direct Observation

Gram stain is generally used for the direct detection of Haemophilus in clinical material (Figure 32-1). However, in some instances the acridine orange stain (AO; see Chapter 6 for more information on this technique) is used to detect smaller numbers of organisms that may be undetectable by gram staining.

To increase the sensitivity of direct Gram stain examination of body fluid specimens, especially CSF, specimens may be centrifuged (2000 rpm for 10 minutes) and the smear is prepared from the pellet deposited in the bottom of the tube. Most laboratories are now equipped with a cytocentrifuge (10,000 × g for 10 minutes) used for concentration of specimens. This is highly recommended over traditional centrifugation for non-turbid specimens. This concentration step can increase the sensitivity of direct microscopic examination from five to tenfold. Moreover, cytocentrifugation of the specimen, in which clinical material is concentrated by centrifugation directly onto microscope slides, reportedly increases sensitivity of the Gram stain by as much as 100-fold (see Chapter 71 for information on infections of the central nervous system).

Gram stains of the smears from clinical specimens must be examined carefully. Haemophilus spp. stain a pale pink and may be difficult to detect in the pink background of proteinaceous material often found in clinical specimens. Underdecolorization may result in misidentification of H. influenzae as either Streptococcus spp. or Listeria monocytogenes.

H. influenzae appears as pleomorphic coccobacilli or small rods, whereas the cells usually appear as long, slender rods. H. haemolyticus are small coccobacilli or short rods with occasional cells appearing as tangled filaments.

H. parainfluenzae produce either small pleomorphic rods or long filamentous forms, whereas H. parahaemolyticus usually are short to medium-length bacilli. Aggregatibacter aphrophilus is a very short bacillum but occasionally are seen as filamentous forms. H. ducreyi may be either slender or coccobacillary. Traditionally, H. ducreyi cells are described as appearing as “schools of fish.” However, this morphology is rarely seen in clinical specimens.

Table 32-3 presents Haemophilus influenzae and H. parainfluenzae biotypes.

TABLE 32-3

Differentiation of Haemophilus influenzae and H. parainfluenzae Biotypes

Organism and Biotype	Indole	Ornithine Decarboxylase	Urease
H. influenzae
I	pos	pos	pos
II	pos	pos	neg
III	neg	pos	neg
IV	neg	pos	pos
V	pos	neg	pos
VI	neg	pos	neg
VII	pos	neg	neg
VIII	neg	neg	neg
H. parainfluenzae
I	neg	neg	pos
II	neg	pos	pos
III	neg	pos	neg
IV	pos	pos	pos
V	neg	neg	neg
VI	pos	neg	pos
VII	pos	pos	neg
VIII	pos	neg	neg

Modified from Versalovic J: Manual of clinical microbiology, ed 10, Washington, DC, 2011, ASM Press.

Antigen Detection

Haemophilus influenzae type b capsular polysaccharide in clinical specimens, such as CSF and urine, can be detected directly using commercially available particle agglutination assays (see Chapter 9). Organisms in clinical infections are usually present at a sufficiently high concentration to be visualized by Gram stain. Therefore, most clinical laboratories no longer perform the latex test for the identification of Haemophilus spp. Latex tests are sensitive and specific for detection of H. influenzae type b, especially in patients treated with antimicrobial therapy prior to specimen collection. However, false positives have been reported in CSF and urine of patients who have recently immunized with the Hib vaccine.

Molecular Testing

Rapid screening procedures are very useful for patient therapy and evaluating outbreaks and have been developed for detection from CSF, plasma, serum, and whole blood. A PCR method for Haemophilus influenzae capsular types a and f has been developed. PCR product was amplified for the specific capsular type for which the primer was designed. PCR has its advantages over serotyping in that problems of cross-reaction and autoagglutination are gone. Detection from some clinical samples has been problematic based on the presence of small numbers of organisms in the sample increasing the need for large samples and concentration procedures.

Diagnosis of chancroid and the identification of H. ducreyi have been successfully completed using a variety of molecular targets. Amplification of the 16SrRNA, the rrs (16S)-rri (23S) spacer region, or the heat shock protein gene groEL has been used in molecular assays. Molecular methods have demonstrated improved sensitivity over traditional methods.

In addition to molecular methods for the identification, pulsed-field gel electrophoresis is considered the gold standard for typing Haemophilus isolates. Additional amplification methods such as repetitive-element sequence-based PCR, ribotyping, restriction fragment length polymorphism (RFLP), multilocus enzyme electrophoresis, and rapidly amplified polymorphic DNA (RAPD) have also been used.

Cultivation

Media of Choice

Haemophilus spp. typically grows on chocolate agar as smooth, flat or convex buff, or slightly yellow colonies. Chocolate agar provides hemin (X factor) and NAD (V factor), necessary for the growth of Haemophilus spp. Most strains will not grow on 5% sheep blood agar, which contains protoporphyrin IX but not NAD. Several bacterial species, including Staphylococcus aureus, produce NAD as a metabolic byproduct. Therefore, tiny colonies of Haemophilus spp. may be seen growing on sheep blood agar very close to colonies of bacteria capable of producing V factor; this is known as the satellite phenomenon (Figure 32-2). The satellite phenomenon has become important in this era of needing to rapidly identify potential agents of a bioterrorist attack. To examine an isolate for the satellite phenomenon, place a single streak of a hemolysin-producing strain of Staphylococcus spp. on a sheep blood agar plate that has been inoculated with a suspected Haemophilus spp. The Staphylococcus lyses the red blood cells adjacent to the streak line, releasing hemin (x factor) and NAD (v factor), providing the necessary components for growth of Haemophilus spp. Haemophilus spp. will grow adjacent to the streak line where the nutrients are available.

A selective medium, such as horse blood–bacitracin agar, may be used for isolation of H. influenzae from respiratory secretions of patients with cystic fibrosis. This medium is designed to prevent overgrowth of H. influenzae by mucoid Pseudomonas aeruginosa. Haemophilus spp. are unable to grow on MacConkey agar.

H. ducreyi requires additional growth factors and special media for cultivation in the laboratory. Two types of media utilized within the laboratory include (1) Mueller-Hinton–based chocolate agar supplemented with 1% IsoVitaleX and 3 µg/mL vancomycin and (2) heart infusion–based agar supplemented with 10% fetal bovine serum and 3 µg/mL vancomycin. The vancomycin inhibits gram-positive colonizing organisms of the genital tract.

Haemophilus spp. will grow in commercial blood culture broth systems and in common nutrient broths such as thioglycollate and brain-heart infusion. However, the growth is often slower, produces weakly turbid suspensions, and may not be readily visible in broth cultures. For this reason, blind subcultures to chocolate agar or examination of smears by AO or Gram stain have been used to enhance detection. Subcultures have not demonstrated a clinically significant effect on the isolation and detection of Haemophilus spp. from blood culture systems.

Rabbit or horse blood agars are commonly used for detecting hemolysis by hemolysin-producing strains of Haemophilus strains unable to grow on 5% sheep blood.

Incubation Conditions and Duration

Most strains of Haemophilus spp. are able to grow aerobically and anaerobically (facultative anaerobes). Growth is stimulated by 5% to 10% carbon dioxide (CO₂). It is recommended that cultures be incubated in a candle extinction jar, CO₂ pouch, or CO₂ incubator. These organisms usually grow within 24 hours, but cultures are routinely held 72 hours before being discarded as negative. An exception is H. ducreyi, which may require as long as 7 days to grow.

Optimal growth of all Haemophilus spp., except H. ducreyi, occurs at 35° to 37° C. Cultures for H. ducreyi should be incubated at 33° C. In addition, H. ducreyi requires high humidity, which may be established by placing a sterile gauze pad moistened with sterile water inside the candle jar or CO₂ pouch.

Colonial Appearance

Table 32-4 describes the colonial appearance and other distinguishing characteristics (e.g., odor and hemolysis) of each species.

TABLE 32-4

Colonial Appearance and Characteristics

Organism	Medium	Appearance
Aggregatibacter aphrophilus	CHOC	Round; convex with opaque zone near center
H. ducreyi	Selective medium	Small, flat, smooth, and translucent to opaque at 48-72 hours; colonies can be pushed intact across agar surface
H. haemolyticus	CHOC	Resembles H. influenzae except beta-hemolytic on rabbit or horse blood agar
H. influenzae	CHOC	Unencapsulated strains are small, smooth, and translucent at 24 hours; encapsulated strains form larger, more mucoid colonies; mouse nest odor; nonhemolytic on rabbit or horse blood agar
H. influenzae biotype aegyptius	CHOC	Resembles H. influenzae except colonies are smaller at 48 hours
H. parahaemolyticus	CHOC	Resembles H. parainfluenzae beta-hemolytic on rabbit or horse blood agar
H. parainfluenzae	CHOC	Medium to large, smooth, and translucent; nonhemolytic on rabbit or horse blood agar
Aggregatibacter segnis	CHOC	Convex, grayish white, smooth or granular at 48 hours

CHOC, Chocolate agar.

Antimicrobial Susceptibility Testing and Therapy

Standard methods have been established for performing in vitro susceptibility testing with clinically relevant isolates of Haemophilus spp. (see Chapter 12 for details on these methods). In addition, various agents may be considered for testing and therapeutic use (Table 32-6). Although widespread H. influenzae is capable of producing beta-lactamase (penicillin resistance), third-generation cephalosporins are not notably affected by the enzyme (i.e., ceftriaxone and cefotaxime) and may be effective therapeutic agents. Therefore, routine susceptibility testing of clinical isolates as a guide to therapy may not be necessary. Care should be taken when preparing inoculum concentrations (0.5 McFarland) for Haemophilus spp.; in particular, beta-lactamase-producing strains of H. influenzae, as higher suspensions may lead to false-resistant results.

Prevention

Several multiple-dose protein-polysaccharide conjugate vaccines are licensed in the United States for H. influenzae type b. These vaccines have substantially reduced the incidence of severe invasive infections caused by type b organisms, and vaccination of children starting at 2 months of age is strongly recommended. Antibody to the Hib capsule and activation of the complement pathway within the host play a primary role in clearance and protection from infection. Newborns are protected for a short period following birth due to the presence of maternal antibodies.

Rifampin chemoprophylaxis is recommended for all household contacts of index cases of Hib meningitis in which there is at least one unvaccinated household member younger than 4 years of age. Children and staff of daycare centers should also receive rifampin prophylaxis if at least two cases have occurred among the children.