Members of the genus Bordetella are small gramnegative bacilli or coccobacilli. All are obligate aerobic bacteria, grow best at 35° to 37° C, do not ferment carbohydrates, oxidize amino acids, and are relatively inactive in biochemical test systems. B. pertussis is fastidious and requires special collection and transport systems as well as culture media. B. pertussis is inhibited by fatty acids, metal ion, sulfides, and peroxides, constituents found in many media. Media for the isolation of B. pertussis requires protective substances like charcoal, blood, or starch. The other Bordetella spp. Are less fastidious and will grow on MacConkey agar or media containing blood. Among the six nonfastidious species, B. holmesii, B. parapertussis, and B. trematum are oxidase negative, while B. avium B. bronchiseptica, and B. hinzii are oxidase positive.

Bordetella pertussis and Bordetella parapertussis

Virulence Factors

Bordetella pertussis possesses a variety of virulence factors that play a role in pathogenesis of disease.

Alamentous hemagglutinin (FHA) and pertactin (a 69-kDa outer membrane protein) are believed to facilitate

attachment to ciliated epithelial cells. Pertussis toxin (P1) is a protein exotoxin that produces a wide variety of responses in vivo. The main activity of PT is modification of host proteins by ADP-ribosyl (adenosine diphosphate) transferase, which interferes with signal transduction. B. parapertussis and B. bronchiseptica

contain the structural gene for PT, but do not express the complete operon. Adenylate cyclase toxin inhibits

host epithelial and immune effector cells by inducing supraphysiologic concentrations of cyclic adenosine monophosphate (cAMP). Tracheal cytotoxin contributes to pathogenesis by causing ciliostastis, inhibiting DNA synthesis, and promoting cell death. Other virulence factors have been proposed, but their current roles in disease remain unclear.


Infections Caused by Bordetella pertussis and Bordetella parapertussis

Classic pertussis or whooping cough resulting from B. pertussis infection occurs following exposure to the organism through the respiratory tract and a 1- to 3-week incubation period, usually 7 to 10 days. The initial catarrhal phase symptoms are insidious and nonspecific. These include sneezing, mild cough, runny nose, and perhaps conjunctivitis, although infants can develop apnea and/or respiratory distress. At this stage, the infection is highly communicable because of the large number of organisms in the respiratory tract. However, cultures are not often performed at this stage because the symptoms are nonspecific.

The catarrhal phase is followed by the paroxysmal phase of the disease. The hallmark of this phase is the sudden attack of severe, repetitive coughing followed by the characteristic “whoop” at the end of the coughing spell. The whooping sound is caused by the rapid gasp for air following the prolonged bout of coughing. Coughing spells may occur many times a day and are sometimes followed by vomiting. Young children may experience apnea and/or pneumonia and require aid in maintaining a patent airway. Many of these symptoms may be either absent or altered in very young infants, partially immunized children, or adolescents and adults. B. parapertussis generally causes a similar disease with milder symptoms. The convalescent phase of disease generally begins within 4 weeks of onset with a decrease in frequency and severity of the coughing spells. Complete recovery may require weeks or months.



Pertussis is a human disease. No known animal reservoir or vector has been found. Infections caused by Bordetella spp. are acquired through the respiratory tract via respiratory droplets or direct contact with infectious secretions. Organisms are uniquely adapted to adhere to and replicate on ciliated respiratory epithelial cells. The organisms remain localized to the respiratory tract, but toxins and other virulence factors are produced and have systemic effects.

Pertussis is one of the most highly communicable diseases of childhood; secondary attack rates of

80% occur among susceptible contacts. In fact, of all childhood diseases for which universal vaccination is recommended, only pertussis incidence has increased in recent decades. During 2003, a total of 11,647 cases of pertussis were reported, compared with only 1730 cases in 1980; that was the largest number of cases since 1964. Approximately 17% of the cases occurred among infants younger than 6 months old, who were too young to receive the first three doses of the vaccine.

In addition, more than 60% of the cases were among adolescents (10 to 19 years of age) and adults

20 years old or older. These are the age groups (infants, adolescents, and adults) when immunity from acellular

pertussis (DTaP) vaccine, given in combination with the diphtheria-tetanus toxoids (DT), is either immature or waning. To counter this trend, two pertussis vaccines were approved in the United States in 2005.

The first vaccine was approved for adolescents 10 to 18 years of age, and the second was approved for

adolescents and adults age 11 to 54 years of age. Both vaccines are administered as a single dose in combination with diphtheria and tetanus toxoids at the reduced booster formulation.

The increase in reported cases is also due to expansion of reporting into adolescents and adults, as well as additional diagnostic tools such as the polymerase chain reaction (PCR). Even in well-immunized populations

such as the United States, periodic outbreaks occur every 3 to 5 years, and isolated cases occur at all times. The vaccination series consists of five doses from age 6 weeks to 6 years. After decreases in immunization

rates, increased attack rates or major epidemics have emerged. For example, from 1990 to 1996,54% of 10,650

children ages 3 months to 4 years with pertussis and with known vaccine status were not age-appropriately vaccinated. Immunity is short-lived, and B. pertussis appears to be maintained in the human population by

adults who become transiently colonized. Adults mayor may not experience respiratory symptoms, which

typically manifests as persistent cough. Other recent developments in vaccine formulations have included the addition of Haemophilus influenzae type b (Hib) to DTaP for the fourth dose at greater than 12 months of age, and in 2002 the first pentavalent immunization was licensed in the United States, which combined DTaP with hepatitis B virus and inactivated polio for use between 6 weeks and 6 years of age. If a child has recovered from confirmed pertussis, additional doses of pertussis vaccine are not necessary.


Miscellaneous Species

The remaining Bordetella spp. are either opportunists or not primary human pathogens. B. bronchiseptica is a respiratory tract pathogen of a number of animals including dogs, in whom it causes kennel cough.

Symptomatic B. bronchiseptica infections in humans generally present with a nonspecific cough or bronchitis,

and often the patients have underlying conditions such as immunosuppression or contact with animals.

B. bronchiseptica and B. holmesii have been infrequently associated with pertussis syndrome and other respiratory tract infections.

Contemporary laboratory diagnosis of pertussis generally employs culture isolation with or without more rapid PCR or direct fluorescent antibody (DFA) testing.

Organisms do not survive well outside the host, so culture can lack sensitivity. Rapid detection with DFA is about 60% to 70% sensitive when compared with culture and can lack specificity. Detection of B. pertussis DNA by PCR amplification was added to case

inclusion criteria in 1995, but testing is typically only available from reference or public health laboratories.

Likewise, serologic testing can identify more cases but is not generally used because it has not been standardized and is not widely available.


Specimen Collection and Handling

Nasopharyngeal aspirates or swabs (calcium alginate or Dacron polyester with a flexible wire shaft) are the specimen of choice for culture, DFA, and PCR testing for Bordetella. Throat cultures have a much lower sensitivity and should be discouraged. Generally, two swabs are collected, one through each external nares;

the swabs should be inserted as far back as possible into the nasopharynx, rotated, held a few seconds, and then gently withdrawn. In practice, swabs are used more commonly than aspiration methods.

Nasopharyngeal swab specimens should be plated directly onto culture media or transferred to an appropriate transport system at the bedside. Fresh media or transport systems and swabs must be available to medical providers on short notice. Transport systems should be selected based on anticipated transport time and

test desired. If the transit time is less than 2 hours or DFAis the desired test, the swab can be expressed into

a solution of 1% casein hydrolysate (casamino acids) broth. Amies transport medium with charcoal is appropriate for up to 24 hours. Specimens should be transported at room temperature and transferred to culture

media as soon as they arrive at the laboratory.

In situations requiring overnight or several-day transport, half-strength charcoal agar containing 10′}{, horse blood and 40 mg/L cephalexin (Regan-Lowe transport medium) should be used. This medium is prepared in screw-capped containers and is inoculated by streaking the surface and then submerging the swab in the agar and leaving it in place. The inoculated Regan-Lowe medium may be sent to the processing laboratory immediately or incubated at the local laboratory for 1 or 2 days at 35° C. The processing laboratory can test any growth on the transport medium directly for Bordetella spp. and use it to subculture for isolation. The CDC Pertussis Laboratory does not recommend incubation of specimens in the transport medium due to overgrowth of commensal flora and decreased yield of B. pertussis.


Microscopic Examination

Microscopic Morphology On Gram stain, the organisms stain as tiny gramnegative coccobacilli and may become elongated if recovered from media containing cephalexin. It may be necessary to increase the safranin counterstaining time to 2 minutes to see typical morphology.

Clinical specimens can only be examined for Bordetella microscopically using DFA staining. The DFA test should only be used along with culture because the lack of sensitivity diminishes the clinical utility of a negative result, and false-positive results can occur even in experienced hands. Thus DFA

should not be used as a replacement for culture. Slides for DFAtesting may be prepared directly from swab specimens or following expression of the material from the swab to a solution of 1%casein hydrolysate. At least two slides should be prepared; slides are dried, heat-fixed, and stained on the same day of collection or stored at -70° C and heat-fixed immediately before staining. Polyclonal fluorescent labeled conjugates for both B. pertussis and B. parapertussis are normally used so that each reagent can serve as a negative control for the other. On microscopic examination, the organisms appear as small, fat rods or coccobacilli with intense peripheral yellow-green fluorescence and darker centers.


Isolation and Identification

Isolation Methods

Since the original development of Bordet-Gengou potato infusion agar with glycerol and horse or sheep blood, few alternative formulations have been as successful.

The most successful of these is charcoal agar supplemented with 10% horse blood and 40 mg/L cephalexin. This medium is identical in composition to the transport medium of Regan and Lowe except that it contains agar at full strength. The medium has a shelf life of up to 8 weeks and is commercially available. Care should be taken to ensure the appropriate concentration of cephalexin in the medium. Some strains of B. pertussis have been reported to be inhibited at 40 mg/L or above. For this reason, it may be advisable to also plate a medium without cephalexin.

Plates for the recovery of Bordetella spp. should be incubated at 35° C in ambient air for a minimum of 7 days. It is important to ensure adequate moisture during this period to prevent plates from drying out.

Most isolates of B. pertussis are detected in 3 to 5 days, whereas B. parapertussis is detected a day or so sooner.

A stereomicroscope should be used to detect the colonies before they become visible to the unaided eye.

Colonial Morphology On charcoal-horse blood and Regan-Lowemedia, young colonies are smooth, glistening, and silver, resembling mercury droplets (Figure 19-32).Colonies turn whitishgray as they age. On Bordet-Gengou agar, colonies of B. pertussis and B. parapertussis are hemolytic.

Identification Methods

Suspicious colonies should be further screened for B. pertussis and B. parapertussis using agglutinating or fluorescein-labeled antisera. For the fluorescent antibody test from a plate isolate, slides should be carefully prepared so that organisms are well dispersed on the slide. This ensures that individual cells show characteristic peripheral staining. The agglutination test requires a larger quantity of bacteria, so subculture from the primary isolation plate is sometimes necessary.

When the results of fluorescent staining or agglutination are clear, confirmatory testing is not required.

B. pertussis and B. parapertussis can be adequately identified and separated with these serologic reagents alone. If these tests yield equivocal results, suspicious organisms should be subcultured to charcoal-horse blood agar, SBA, and CHOC (the latter to assess for Haemophilus sp. that may have broken through on selective charcoal-horse blood). Growth patterns should be observed, serologic tests repeated, and additional biochemical tests performed (Table 19-8) to confirm the identification. Box 19-3 summarizes the laboratory diagnosis of pertussis by culture.


Nucleic Acid Detection

Detection of Bordetella nucleic acid by PCR from nasopharyngeal swabs is now a primary rapid diagnostic strategy. This methodology can circumvent many of the problems associated with specimen transport and bacterial cultivation, although it too has limitations. Presently, laboratories that offer PCR must conduct in-house verification and validation studies because there are no FDA-approved kits available.

External proficiency testing is available from the College of American Pathologists to assist with the laboratory’s overall quality assurance program.


Serologic Testing

Serologic diagnosis of pertussis can be used to study outbreaks and to document seroconversion following immunization or infection. Unfortunately, the lack of association between antibody levels and immunity to

pertussis makes results of serologic testing difficult to interpret, so tests for routine purposes are not approved for diagnostic use. If serologic assays are to be performed, reference sera available from the Laboratory of Pertussis (Food and Drug Administration, Bethesda, Md) should be used for quality control.

Optimal diagnostic sensitivity requires paired sera and testing for multiple immunoglobulin class-antigen combinations (e.g., IgG to FHA and PT, along with IgA to FHA). Testing for IgM as an acute phase antibody

has been disappointing and currently is of little assistance.

Serology also tends to be retrospective; several weeks are often required to demonstrate a diagnostic response. Enzyme immunoassay for IgG or IgA antibody to FHA from a single serum specimen or from nasopharyngeal aspirates have been developed, but have not gained widespread acceptance.

Although a few strains of erythromycin-resistant B.pertussis have been reported, erythromycin is still the

drug of choice for treatment and prophylaxis of pertussis. Erythromycin is important for eradication of  the organism and prevention of secondary cases, but only has clinical efficacy if treatment is started during

the catarrhal phase of disease. Azithromycin has fewer and milder side effects, a longer half-life, and requires fewer daily doses and better patient compliance.

Trimethoprim-sulfamethoxazole is also an alternate for treatment or prophylaxis. Routine antimicrobial susceptibility testing of B. pertussis or B. parapertussis is not necessary because of predictable sensitivity to macrolides. The susceptibility of B. bronchiseptica is not predictable, and tests should be performed, although the organism is usually susceptible to the aminoglycosides.



Bordetella pertussis and B. parapertussis cause pertussis, or whooping cough, and are primary human pathogens of the respiratory tract. Humans are the only host. Laboratory diagnosis is made by culture of the fastidious gramnegative bacilli along with rapid methods such as PCR and DFA. The disease is preventable by vaccination and treatable with antimicrobials. B. bronchiseptica can be part of the normal oral flora of dogs and cats, and infections in humans typically follows bites by these animals. B. parapertussis and B. bronchiseptica grow on most  laboratory media including MacConkey agar.


Baughman AL et al: Establishment of diagnostic cutoff  points for levels of serum antibodies to pertussis toxin,

filamentous hemagglutinin, and fimbriae in adolescents and adults in the United States, C/in Diagn Lab Immunol 11:1045,2004.

Centers for Disease Control and Prevention: Summary of notifiable diseases-United States, 2003, MMWR 52:1,


Centers for Disease Control and Prevention: Outbreak of pertussis associated with hospitals-Kentucky, Pennsylvania, and Oregon, 2003,MMWR 54:67,2005.

Centers for Disease Control and Prevention: Recommended childhood and adolescent immunization schedule-

United States, 2005,MMWR 53:Q1, 2005.

Centers for Disease Control and Prevention: Epidemiology and prevention of vaccine-preventable diseases, ed 8

(2nd printing), 2005. Available at: publications/pink/default.htm. Accessed March 7, 2006.

Cherry JD: Epidemiological, clinical, and laboratory aspects of pertussis in adults, C/inInfect Dis 28(suppl 2):112, 1999. Gilchrist MJR: Pertussis: pathophysiology and prevention, C/in Microbiol NewsI12:17, 1990.

Hallander HOet al: Comparison of nasopharyngeal aspirates with swabs for culture of Bordetella pertussis, J Clin Microbiol 31:50, 1993.

HewlettEL: Acommentary on the pathogenesis of pertussis, C/in Infect Dis 28(suppl 2):S94, 1999.

Hoppe J: Update on epidemiology, diagnosis, and treatment of pertussis, Eur J C/in Microbiollnfect Dis 15:189, 1996.

Koneman EWet ai, editors: Color atlas and textbook of diagnostic microbiology, ed 5, Philadelphia, 1997, Lippincott.

Langley JM et al: Azithromycin is as effective and better tolerated than erythromycin estolate for the treatment of pertussis, Pediatrics 114:e96, 2004.

Lind-Brandberg L et al: Evaluation of PCR for diagnosis of Bordetella pertussis and Bordetella parapertussis infections, J Clin Microbiol 36:679, 1998.

Loeffelholz MJ: Bordetella. In Murray PR et ai, editors: Manual of clinical microbiology, ed 8, Washington, DC, 2003, ASM.

Muller FMCet al: Laboratory diagnosis of pertussis: state of the art in 1997,J C/in MicrobioI35:2435, 1997.

Woolfrey BF, Moody JA: Human infections associated with Bordetella bronchiseptica, C/in Microbiol Rev 4:243, 1991 .

figure 19-32 table 19-8 box 19-3