Performance of 2 automated real time PCR methods for the detection of Bordetella pertussis and Bordetella parapertussis

The aetiological agent of pertussis is B. pertussis, but B. parapertussis and B. holmesii can cause very similar clinical manifestations.¹B. bronchiseptica infection is rare, clinically distinct and affects debilitated patients.² The most commonly used targets for the diagnosis of Bordetella spp. by PCR have been the IS481 and IS1001 sequences. IS481 is present in B. pertussis and B. holmesii, and may also be present in B. bronchiseptica. IS1001 is found in B. parapertussis and occasionally in B. bronchiseptica.^3–5 One way to distinguish species is based on the use of specific primers for the promoter region of the pertussis toxin gene (ptxA-pr), which is specific to B. pertussis.^3,5,6 An alternative is the B. pertussis porin gene (BPTD_0837).⁴ The BP283 gene has also been used to identify B. pertussis.⁷ Numerous kits are now available for the molecular diagnosis of pertussis. However, their interpretation can be complex due to the possibility of detecting the same sequences in different species. The aim of this study was to evaluate the performance of two real-time PCR methods, RealCycler® BORD-T (Progenie Molecular), and Simplexa™ Bordetella Direct, (DiaSorin Molecular LLC) for the diagnosis of pertussis.

Fifty nasopharyngeal swab/nasopharyngeal wash samples obtained from patients with clinical suspicion of pertussis between July 2018 and January 2020 were studied, and 44 nasopharyngeal swab samples received in February 2022 for diagnosis of SARS-CoV-2 by RT-PCR were studied as a control group. Samples were kept frozen at −80°C until studied. All samples were from epidemiological surveillance studies and were processed simultaneously with the RealCycler® BORD-T assay after nucleic acid extraction and Simplexa™ Bordetella Direct assay directly from the sample without prior extraction. The results of the two techniques were interpreted according to the manufacturers' respective recommendations.

For the 50 suspected pertussis samples, the results with RealCycler® BORD-T were: 28 (56%) B. pertussis, 12 (24%) Bordetella spp., six (12%) B. parapertussis or B. bronchiseptica, three (6%) co-infection with different Bordetella spp. and one (2%) negative. With Simplexa™ Bordetella Direct, the results were: 39 (78%) B. pertussis, six (12%) B. parapertussis, two (4%) co-infection by different Bordetella spp., two (4%) negative and one (2%) invalid due to amplification inhibition (Table 1). In 47 samples with suspicion of pertussis (94%), the two methods matched regarding the identification at the genus level or co-infection with Bordetella spp. In 27 (54%) of these cases the two techniques identified the species as B. pertussis. In 11 cases (22%) where Simplexa™ Bordetella Direct classified the result as B. pertussis, RealCycler® BORD-T classified it as Bordetella spp. In the six cases (12%) identified as B. parapertussis by Simplexa™ Bordetella Direct, the RealCycler® BORD-T result was either B. parapertussis or B. bronchiseptica. All 44 samples collected for SARS-CoV-2 detection were negative for Bordetella spp. in both techniques.

Table 1.

Distribution of RealCycler® BORD-T and Simplexa™ Bordetella Direct results in terms of the total number of samples of suspected pertussis studied.

RealCycler® BORD-T				Simplexa™ Bordetella Direct			Total
	IS481	IS1001	BP283		IS481	IS1001	N	%
B. pertussis	Positive	Negative	Positive	B. pertussis	Positive^a	Negative^a	27	54
Bordetella spp.	Positive	Negative	Negative	B. pertussis	Positive^a	Negative^a	11	22
B. parapertussis or B. bronchiseptica	Negative	Positive	Negative	B. parapertussis	Negative^b	Positive^b	6	12
[1,0]Co-infection by different Bordetella spp.	Positive	Positive	Positive	[1,0]B. pertussis and B. parapertussis	Positive^b	Positive^b	1	4
	Positive	Positive	Negative		Positive^b	Positive^b	1
Co-infection by different Bordetella spp.	Positive	Positive	Positive	B. pertussis	Positive	Negative	1	2
Bordetella spp.	Positive	Negative	Negative	Negative	Negative	Negative	1	2
Negative	Negative	Negative	Negative	Negative	Negative	Negative	1	2
B. pertussis	Positive	Negative	Positive	Invalid^c	Invalid	Invalid	1	2
							50	100

The results in this table are only applicable when the internal quality specifications of each of the techniques evaluated (e.g., amplification of internal controls and/or Ct within range) are met.

a

This combination of results from Simplexa™ Bordetella Direct would not rule out infection by B. holmesii.

b

This combination of results from Simplexa™ Bordetella Direct would not rule out infection by B. bronchiseptica.

c

Invalid due to amplification inhibition.

The RealCycler® BORD-T assay kit detects both IS481 and the BP283 region. This combination incorporates high sensitivity and high specificity for pertussis diagnosis. This kit also includes IS1001 to detect B. parapertussis.

Simplexa™ Bordetella includes IS481 for the detection of B. pertussis and IS1001 for B. parapertussis.⁸ Although the combination of positivity and negativity for each of these targets does not definitively exclude others such as B. holmesii or B. bronchiseptica, for practical purposes, in samples with clinical suspicion of pertussis the result could be considered "probable infection by" B. pertussis or B. parapertussis, as appropriate.³ This kit has shown very good sensitivity and specificity⁹ and excellent overall percent agreement values.¹⁰

In this study the number of samples studied is small. Furthermore, there was no "gold standard" to assess the sensitivity of the techniques, no control strains were available and discordant cases were not confirmed with a third alternative technique. However, the results of the two techniques matched well for the detection of Bordetella spp. Each technique offers its own advantages: RealCycler® BORD-T would theoretically be highly specific for B. pertussis, while Simplexa™ Bordetella Direct does not require nucleic acid extraction, making it a simple and rapid alternative.

Conflicts of interest

In this study the Simplexa™ Bordetella Direct (DiaSorin Molecular LLC) reagents were provided by DiaSorin Iberia S.A. Authors V.B. and E.M. are employees of DiaSorin Iberia S.A.

This study has been partially submitted and accepted as a submission to the XXVI National Congress of the Sociedad Española de Enfermedades Infecciosas y Microbiología Clínica [Spanish Society of Infectious Diseases and Clinical Microbiology].