English abstract
Objectives: To evaluate real-time polymerase chain reaction (PCR) as a diagnostic tool
for pertussis (whooping cough), vis-à-vis culture, the long-time gold standard, with
emphasis on the importance of clinical correlation in determining specificity. Methods:
Nasopharyngeal/throat swab samples in charcoal media received from all over Oman
at Central Public Health Laboratories from January 2014 to December 2016 were
included in this study. These samples were tested using both culture and real-time
PCR. PCR was compared with culture to calculate its sensitivity and specificity.
Further clinical correlation was conducted for the discrepant cases using different case
definitions. Results: A total of 590 nasopharyngeal/throat samples were included in
the study. Out of the 590 samples, 73 were positive by PCR compared with 26 positive
samples by culture (which were also positive by PCR). The sensitivity and specificity
of PCR compared with those of culture were 100% (95% CI: 86.77−100) and 91.7%
(95% CI: 89.07−93.81), respectively. To rule out false-positive results by PCR, clinical
correlation was performed. Out of 47 cases that were positive by PCR but negative by
culture, 44 cases were clinically evaluated by accessing clinical details. Out of these 44
cases, 21 (47.7%) met the pertussis clinical criteria according to the CDC-2014 case
definition, 41 (93.2%) according to the Europe-2008 case definition, and 44 (100%)
according to the Canada-2009 and Australia-2014 case definitions. With only positive
PCR, these cases were classified as confirmed according to these case definitions, which
increased the specificity of PCR to 95.7–100%. The mean turnaround time was 3.5
days for PCR compared to 6.2 days for culture. Conclusions: Real-time PCR is a highly
sensitive and specific test for the diagnosis of Bordetella pertussis infection. Based on our
results, we recommend setting up PCR diagnostic facilities in regional hospitals in Oman
as this will lead to a timely and accurate diagnosis of pertussis
