Abstract

BACKGROUND:

From October 2005 to March 2006, a laboratory-confirmed outbreak of pertussis occurred in preschool-aged children (1-4 years) in Toronto, Canada. A case-control study in children was done to identify the risk factors for being positive for Bordetella pertussis by polymerase chain reaction (PCR).

METHODS:

A case was defined as an individual 0 to 18 years of age who tested positive for B. pertussis by PCR in Toronto from October 2005 to March 2006. Each case was matched to 2 controls (negative for B. pertussis by PCR) by: (1) age, (2) geographic region of testing, and (3) date of testing.

RESULTS:

One hundred eighty-nine cases and 296 controls were enrolled. Only 42% of cases (vs. 25% controls, P < 0.0001) met a clinical definition of pertussis, 3 cases (4 controls) were hospitalized, and the secondary household attack rate was 8%. One-third of cases were positive for another respiratory pathogen and >90% of cases and controls were up-to-date with pertussis immunization. Children attending school (matched odds ratio [ORm] = 5.2; 95% confidence intervals [CI]: 1.4-19.0), day care (ORm = 2.2; 95% CI: 1.2-4.0), visiting a doctor's office (ORm = 1.6; 95% CI: 1.0-2.5), or exposed to a household member (median age: 4-5 years) with a pertussis-like cough (ORm = 1.9; 95% CI: 1.0-3.6) were significantly more likely to have a positive PCR test for B. pertussis.

CONCLUSIONS:

The main risk factors for PCR positivity for B. pertussis were school or day care attendance. Atypical symptoms were likely moderated by high immunization rates or may have been caused by other respiratory pathogens. In some cases, a positive PCR result might simply have reflected transient nasopharyngeal carriage of B. pertussis.