Abstract
The use of a multiplex respiratory real-time PCR in patients clinically suspected of pertussis increases the number of pathogens detected.
Keywords: Real-time PCR, Pertussis, Respiratory virus, Diagnosis
Bordetella pertussis is a major etiological agent of pertussis (whooping cough). The disease can be atypical and pertussis-like coughing can be caused by other respiratory pathogens [4]. The diagnosis of pertussis is usually suspected clinically, but only proven by specific polymerase chain reaction (PCR) or culture.
PCR methods are more sensitive for B. pertussis identification and other respiratory pathogens. Multiplex real-time PCR could be applied as broad-range respiratory PCR.
All patients in whom a diagnostic test for B. pertussis was requested from April 2001 to February 2002 were included. The records of the patients were analyzed retrospectively and the patients were assigned to two groups as described in Table 1 [6].
Table 1.
Two groups of rettrospectively analyzed patients.
| WHO recommended case definition |
|---|
| Clinical case definition |
| A case diagnosed as pertussis by a physician or |
| A person with a cough lasting at least 2 weeks with at least one of the following symptoms: |
| Paroxysms (i.e., fits) of coughing |
| Inspiratory whooping |
| Post-tussive vomiting (i.e., vomiting immediately after coughing) without other apparent cause |
| Criteria for laboratory confirmation |
| Isolation of Bordetella pertussis or |
| Detection of genomic sequences by means of the PCR |
| Case classification: |
| Suspected: A case that meets the clinical case definition. |
| Confirmed: A person with a cough that is laboratory-confirmed or epidemiologically linked to a laboratory confirmed case. |
| Group I - Case that meets the definition for suspected pertussis |
| Group II - Case that does not meet the definition for suspected pertussis. |
Nasopharyngeal swabs (15 samples), throat swabs (39 samples), sputum (three samples) and nasopharyngeal aspirates (NPAs) (five samples) were received. The PCR assay was performed by a single extraction. The extracted nucleic acid was thereafter added to a four-tube multiplex RNA real-time PCR for 11 RNA respiratory viruses [4] and a four-tube DNA real-time PCR [4, 5]. All PCR products were detected by specific fluorophore-labelled probes that can be distinguished without post-PCR analysis.
Sixty-two patients were seen in the study period from which a diagnostic test for B. pertussis was requested. Sufficient clinical information could be obtained on 59/62 (95%) patients and assigned Group I or Group II. The clinical information for 59 patients is shown in Table 2, the symptoms of paroxysmal cough and whoops were significantly associated with Group I.
Table 2.
Clinical characteristics of patients in study group
| Pertussisa | Non-pertussisb | Whole group | P c | |
|---|---|---|---|---|
| (n = 38) | (n = 21) | (n = 59) | ||
| Group I | Group II | (%) | ||
| Male | 16 | 9 | 27 | NS |
| Age | ||||
| <1 | 18 (13) | 10 (3) | 28 | NS |
| 1–12 | 16 | 8 | 26 | NS |
| >12 | 4 | 3 | 8 | NS |
| Pertussis vaccination | 25 | 18 | 42 | 0.018 |
| Immunocompromised | 3 | 2 | 5 | NS |
| Stage of diseased | ||||
| Catarrhal | 17 | 9 | 26 | NS |
| Paroxysmal | 12 | 5 | 17 | NS |
| Convalescent | 3 | 4 | 7 | NS |
| Clinical detailsd | ||||
| Cough | 32 | 18 | 50 | NS |
| Paroxysmal | 26 | 8 | 34 | <0.005 |
| Whoop | 5 | 0 | 5 | 0.045 |
| Vomiting | 16 | 9 | 25 | NS |
| Fever | 5 | 5 | 11 | 0.014 |
| GP | 16 | 12 | 28 | NS |
| Out-patient clinic | 8 | 4 | 12 | NS |
| Hospital admission | 14 | 8 | 22 | NS |
aPatients that meet the clinical definition for suspected pertussis
bPatients that did not meet the clinical definition for suspected pertussis
cStatistical differences between Group I and Group II
dIncomplete clinical details for six cases in the pertussis group and three cases in the non-pertussis group
Numbers in parentheses indicate no pertussis vaccination
No clinical information was available for three cases and these were not assigned to either group
Analysis of the samples by real-time PCR showed that B. pertussis was detected in 17 of the 38 patients in Group I and none of the patients in Group II (P < 0.005). Using the WHO definition for suspected pertussis used in surveillance as the ’gold standard’ for B. pertussis, the PCR has a sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 45, 100, 100 and 48%, respectively. Other pathogens were detected in group I as shown in Table 3. Some respiratory viruses were detected in Group II. All other respiratory pathogens had sensitivity and specificity lower than B. pertussis. Pathogens were detected by PCR in 31 out of 38 (82%) cases and 10 out of 21 (48%) cases in Group I and Group II, respectively. This difference was significant (p = 0.01).
Table 3.
Pertussis-related symptoms in relation to pathogen detected
| Pathogens | Group Ia (n = 38) | Group IIb (n = 21) | P c |
|---|---|---|---|
| Bordetella Pertussis | 17 | 0 | <0.005 |
| Other atypical bacteriad | 2 | 0 | NS |
| Human rhinoviruses | 16 | 4 | NS |
| Respiratory syncytial virus | 1 | 5 | 0.02 |
| Parainfluenza viruses | 2 | 3 | NS |
| Adenovirus | 6 | 1 | NS |
| Human coronavirus –229E | 1 | 0 | NS |
| Human metapneumovirus | 0 | 1 | NS |
| Mixed infections | 14 | 4 | NS |
| No infection | 7 | 11 | 0.01 |
aPatients that meet the clinical definition for suspected pertussis
bPatients that did not meet the clinical definition for suspected pertussis
cStatistical differences between Group I and Group II
d Mycoplasma pneumoniae (1) and Chlamydophila pneumoniae (1)
No influenza A&B virus, human coronavirus OC43, B. parapertussis or B. holmesii were detected
Detection of B. pertussis by PCR was indicated on the basis of clinical presentation and PPV of 100% was obtained. However, 21 out of 38 cases that met the clinical definition of pertussis were negative for B. pertussis. Therefore, either the B. pertussis PCR gave false-negative results or the definition gave a false-positive result. Different sample types have different sensitivities, with NPAs being the most sensitive. PCR can also give false-negative results and fewer diagnoses in those patients with pertussis could be due to the quality of the sample.
Other pathogens were detected in 31 out of 38 patients with clinical diagnosis of pertussis. Other pathogens that cause a pertussis-like disease have previously been described using serological assays and Cherry et al. describe that pertussis can be misdiagnosed, as there are a number of viral and bacterial pathogens other than B. pertussis that can cause a paroxysmal cough [1, 2]. It has been well described that PCR is more sensitive than culture at detecting B. pertussis in the acute phase [3]. The use of PCR for diagnosis would help improve microbiological diagnosis, improve the speed of results, and reduce the need for the reliance on imperfect clinical criteria for diagnosis. This would prompt more appropriate treatment of B. pertussis and other atypical bacteria, differentiating them from viral infections as well as use of vaccination to prevent spread of pertussis.
In conclusion, multiplex PCR in a combined approach identifies B. pertussis and other pathogens causing pertussis-like symptoms and use of this form of diagnosis might help in treatment and improve management of patients.
Abbreviations
- PCR
polymerase chain reaction
- NPAs
nasopharyngeal aspirates
- PPV
positive predictive value
- NPV
negative predictive value
Footnotes
Alida M. van Kruijssen and Kate E. Templeton, contributed equally to this work.
References
- 1.Cherry JD, Grimprel E, Guiso N, Heininger U, Mertsola J. Defining pertussis epidemiology: clinical, microbiologic and serologic perspectives. Pediatr Infect Dis J. 2005;24(suppl):S25–S34. doi: 10.1097/01.inf.0000160926.89577.3b. [DOI] [PubMed] [Google Scholar]
- 2.Jackson LA, Cherry JD, Wang SP, Grayston JT. Frequency of serological evidence of Bordetella infections and mixed infections with other respiratory pathogens in university students with cough illnesses. Clin Infect Dis. 2000;31:3–6. doi: 10.1086/313911. [DOI] [PubMed] [Google Scholar]
- 3.Loeffelholz MJ, Thompson CJ, Long KS, Gilchrist MJ. Comparison of PCR, culture, and direct fluorescent-antibody testing for detection of Bordetella pertussis. J Clin Microbiol. 1999;37:2872–2876. doi: 10.1128/jcm.37.9.2872-2876.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Templeton KE, Scheltinga SA, van den Eden WCJFM, Graffelman AW, van den Broek PJ, Claas EC. Improved diagnosis of etiology of community-acquired pneumonia using real-time PCR. Clin Infect Dis. 2005;41:345–351. doi: 10.1086/431588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Templeton KE, Scheltinga SA, van der Zee, Diederen BM, van Kuijssen AM, Goossens H, Kuijper EJ, Claas EC. Evaluation of real-time PCR for detection of and discrimination between Bordetella pertussis, Bordetella parapertussis, and Bordetella holmesii for clinical diagnosis. J Clin Microbiol. 2003;41:4121–4126. doi: 10.1128/JCM.41.9.4121-4126.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.World Health Organisation (2005) Vaccines and biologicals. WHO-recommended standards for surveillance of selected vaccine-preventable diseases. WHO/V&B/03.01 (http://www.who.int/vaccines-documents/DocsPDF03/www742.pdf)