To determine how much bacterial culturing would aid the diagnosis of pertussis in China, a study was conducted in a 35-bed ward for respiratory diseases at Beijing Children’s Hospital from June 2000 to May 2001. This facility is the largest children’s hospital in China; it has 3,000–4,000 visits daily to its outpatient department. Nasopharyngeal (NP) swabs were taken from children who had been admitted to the hospital because their cough, with or without paroxysms, had persisted for >2 weeks and was worsening. A total of 55 children (age range 35 days to 13 years) were enrolled during the study period. In addition, NP swabs were obtained from two children (ages 10 weeks and 13 years) with paroxysmal cough who were seen at the outpatient department. Information about disease history, immunization status, and cough characteristics was obtained. Physical examination, chest X-ray, and blood tests were performed. At admission, all participants were diagnosed as having bronchitis, bronchopneumonia, or pneumonia.

After collection, NP swabs were immediately spread onto charcoal agar plates supplemented with cephalexin. Details of Bordetella pertussis culture have been described previously (3). In brief, after inoculation, the plates were incubated in a humid atmosphere at 35°C and inspected daily for 7 days to determine pertussis-like colony growth. Suspected colonies were Gram stained and tested by slide agglutination with antisera to B. pertussis and B. parapertussis (Murex Diagnostics, Dartford, England).

Serum immunoglobulin (Ig) G antibodies to purified pertussis toxin (PT) were tested by enzyme immunoassay, as described (3). Seropositivity was determined by comparing the antibody results in patient serum samples with those in 460 healthy Chinese persons. Results exceeding the mean of the controls by two standard deviations were considered to be seropositive.

Six infants <4 months of age were culture positive for B. pertussis (Table). Five of these patients were in the study group of 55 hospitalized children; the other was one of two children seen at the outpatient department. Before they went to the hospital, all six infants had taken broad-spectrum antibiotics but not erythromycin. None had received any doses of pertussis vaccine.

The immediate family members or other relatives of five infants (cases 1, 2, 4, 5, and 6) had concurrent and persistent cough (Table). NP swabs and serum samples were obtained from family members of cases 1, 5, and 6 (data not shown). Case-patient 1’s grandmother was culture positive for B. pertussis. She, as well as the patient’s mother and father, had been coughing for several weeks, and they all had IgG diagnostic antibodies to PT in their sera. Patient 5’s mother, aunt, and 10-year-old cousin had diagnostic serum IgG antibodies to PT, and another, 8-year-old cousin was culture positive. The grandmother of case-patient 6 had been coughing for 1 month and had diagnostic serum IgG antibodies to PT.

Because antigenic divergence, with respect to PT and pertactin (PRN), has been recently found between B. pertussis vaccine strains and circulating strains, the PRN and PT types of eight clinical strains isolated in this study and two vaccine strains were examined. The methods used for this genotyping were LightCycler (Roche Applied Science, Mannheim, Germany) real-time PCR and fluorescence resonance energy transfer hybridization probes (4,5). The two vaccine strains and seven clinical strains harbored prn1, and one clinical strain contained prn2. For PT types, the vaccine strains harbored ptxS1B or ptxS1D, and all clinical strains had ptxS1A.

To our knowledge, this is the first report of culture-confirmed pertussis cases from China during the last 30 years. Of the 55 patients hospitalized for persistent cough, 5 (9%) were culture positive for B. pertussis. These results indicate that pertussis is not uncommon in the Chinese population and still causes substantial illness in infants and young children, although the pertussis vaccination coverage for the first three doses is >85% (1).

Immunity from immunization wanes with time; thus, older children and adults become susceptible to pertussis (6–8). The role of older children and adults in transmitting B. pertussis to unvaccinated infants has been well documented. More than 80% of hospitalized infants are not the index cases in their families (9). Our results agree with these reports. The family members and other relatives of five infants with culture-confirmed pertussis started to cough first. Evidence of laboratory-confirmed pertussis was obtained from the family members and other relatives of three infants.

Although the most serious effects from pertussis occur in young infants (10,11), all six ill infants in our study recovered. Pierce et al. reported 13 critically ill infants with confirmed pertussis, all <3 months of age (11); 4 had leukocyte counts >100 X 10⁹/L; all these infants died. Nine had leukocyte counts <100 X 10⁹/L (11). In the six ill infants in our study, the highest leukocyte count was far lower, 37 X 10⁹/L. However, our six patients had taken antibiotics before consultation.

In this study, pertussis was not initially suspected in these ill infants. None had the characteristic whoop. Pertussis at this age group is likely to be atypical, and symptoms resemble those of other respiratory tract infection, apnea, or cyanosis. Consequently, the diagnosis of pertussis is not considered and the treatment is delayed (10). These six infants were initially diagnosed as having bronchitis, bronchopneumonia, or pneumonia. The fact that pertussis was not considered in the differential diagnosis may indicate that clinicians were not aware that B. pertussis was circulating in the community.

Early and correct diagnosis of pertussis is important for effective therapy and prevention of transmission of the disease. Culture of B. pertussis from NP samples usually takes 3–7 days. In comparison with culture, PCR is a more sensitive and specific method for diagnosing pertussis (3). Measurement of specific serum antibodies to B. pertussis antigens by enzyme immunoassay can also facilitate this diagnosis, and results are helpful for epidemiologic studies (3,7,12). The use of culture for the diagnosis of pertussis is now being considered in Beijing Children’s Hospital.

Our results suggest that a number of pertussis cases are likely being misdiagnosed in China and that the incidence of the diseases is underestimated.