Background: Surveillance for pertussis is complicated by imperfect diagnostic tests. Culture is the gold standard for confirming B. pertussis, yet use of PCR has increased since it was included in the CSTE case definition in 1997. Serology is not an accepted confirmatory test for pertussis outside of Massachusetts; however, serologic assays for pertussis are commercially available and being used for diagnosis and notification. Since both culture and PCR have low sensitivity after 14 days of cough, including serology in the case definition could improve case confirmation, provided validated assays are used.
Objectives: To evaluate the trends in use of culture, PCR and serology among pertussis cases reported to the National Notifiable Diseases Surveillance System (NNDSS) from 1990 to 2008.
Methods: Annual proportions of cases with each test performed were calculated by dividing the total number of cases with any reported result (positive, negative, indeterminate) for each diagnostic test by the total number of confirmed and probable cases reported. Regression was used to detect significant trends in testing patterns over time.
Results: Culture increased from 1990 to its peak in 1993, when it was used in 59% of all reported cases. Between 1993 and 2008, the use of culture declined 47% (slope:-3.38; p < 0.001). Since its inclusion in the CSTE case definition in 1997 and 2008, the proportion of cases tested by PCR has increased 39% (slope: 3.47; p < 0.001). The proportion of cases tested by serology fluctuated between 1995 and 2008 (median proportion: 11%; Range: 1% - 22%. Slope: 0.542; p=0.119). In 1995, 16 states including Massachusetts reported serology results (median number tests: 3; Range: 1-15). As of 2008, Massachusetts reported 35% of all serology results (n=527); however, 34 other states also reported serology (median number tests: 13; Range: 1-156).
Conclusions: There has been a nationwide shift toward PCR as the primary diagnostic method. Concurrently, serological testing has increased among states outside