Diagnosing Pertussis: PCR, Culture, and Serology

Why Pertussis Is So Often Missed
Nasopharyngeal Swab: How to Do It Correctly
PCR: The Test of Choice in the First 3 Weeks
Culture: Gold Standard but Slow
When PCR Becomes Negative
Serology: The Late-Stage Test
The Lymphocyte Count Clue
Chest X-Ray Findings
Clinical Diagnosis vs Laboratory Confirmation
Key Research Papers
Connections
Featured Videos

Why Pertussis Is So Often Missed

Whooping cough is one of the most frequently missed bacterial infections in clinical practice. Several factors combine to make diagnosis difficult:

Early symptoms are indistinguishable from a cold: In the catarrhal stage — the first 1–2 weeks — pertussis looks exactly like a common cold or early bronchitis. No test is ordered; no one thinks of pertussis.
The classic "whoop" is absent in many patients: Vaccinated children, adolescents, and adults often do not produce the distinctive whooping sound. Their pertussis looks like a persistent, unremarkable cough.
Specific tests must be explicitly ordered: Pertussis is not detected by a standard respiratory viral panel or a sputum culture. Clinicians must specifically request a nasopharyngeal pertussis PCR or culture — and many do not think to do so.
The disease runs its natural course before diagnosis: By the time many patients see a doctor for a prolonged cough, they are already 3–6 weeks into the illness. At that point, antibiotic treatment no longer shortens the illness (though it stops contagiousness), and the window for the most sensitive PCR testing has passed.

Population studies suggest that for every reported pertussis case, many more go undiagnosed — particularly in adults. Understanding the available diagnostic tests and their timing is essential to catching the disease when it matters.

Nasopharyngeal Swab: How to Do It Correctly

For both PCR and culture testing, the sample must come from the posterior nasopharynx — the very back of the nasal passage, where it meets the throat. This is not a nasal swab and it is not a throat swab. It specifically targets the area where Bordetella pertussis colonizes the ciliated epithelium of the upper respiratory tract.

Getting the sample right matters enormously:

Use Dacron or calcium alginate swabs — NOT cotton. Cotton contains fatty acids and other compounds that inhibit the PCR reaction. Cotton swab testing will produce false-negative PCR results.
Technique: Insert the swab through the nostril horizontally (not upward), advancing it along the floor of the nasal cavity until it reaches the posterior nasopharynx — typically 5–8 cm in adults. Leave it in place for 10–30 seconds while rotating gently, then remove. The distance is roughly equivalent to half the distance from the tip of the nose to the earlobe.
Both nostrils are ideal: Sampling both nostrils increases the bacterial yield and reduces the chance of a false-negative result.
Patient preparation: The person should blow their nose first if possible. The procedure is uncomfortable — patients should be warned it will feel like a deep tickle or mild pain at the back of the nose, often causing eyes to water.
In young infants: A nasopharyngeal aspirate (using a small suction catheter) may be more practical than a swab and can yield higher bacterial counts.

Samples should be transported to the laboratory promptly. For culture, transport in specialized media (Regan-Lowe transport medium) at room temperature is required; the culture will be killed by cold or excessive heat.

PCR: The Test of Choice in the First 3 Weeks

Polymerase chain reaction (PCR) testing is now the standard of care for pertussis diagnosis in clinical practice, and it has largely replaced culture for this purpose. PCR's advantages are significant:

High sensitivity: Can detect as few as 10–100 organisms per sample — far fewer than culture requires for a positive result.
Speed: Results typically available within 24–48 hours, compared to 7–12 days for culture.
Stability: Does not require the live bacteria that culture depends on; bacterial DNA is detectable even if organisms have been killed.
Antibiotic tolerance: Remains positive for approximately 5 days after starting appropriate antibiotic treatment, whereas culture may become negative within hours of antibiotic exposure.

The most common PCR target for B. pertussis detection is the IS481 insertion sequence, which is present in approximately 238 copies per bacterial chromosome. This multicopy target makes it extremely sensitive. However, IS481 is also present in Bordetella holmesii, a less common pathogen that can cause pertussis-like illness. Some reference laboratories use additional targets (such as ptxP or ptxA for the pertussis toxin promoter) to distinguish between species.

PCR sensitivity is highest in the first 1–3 weeks of cough and in unvaccinated patients with higher bacterial loads. Sensitivity is lower in vaccinated patients (who have lower bacterial burdens), in patients who have received even a single dose of antibiotics, and in the later stages of illness.

Culture: Gold Standard but Slow

Before PCR became widely available, bacterial culture was the only laboratory confirmation of pertussis. It remains the definitive "gold standard" in the sense that a positive culture is unambiguous confirmation of B. pertussis infection — there are no cross-reactive false positives. It also yields a live isolate that can be characterized for antibiotic susceptibility, vaccine strain matching, and genomic epidemiology.

However, culture has major practical disadvantages:

Specialized media required: B. pertussis will not grow on standard blood agar or chocolate agar. It requires specialized media such as Regan-Lowe charcoal agar or the original Bordet-Gengou medium. Most routine hospital laboratories do not maintain these media.
Long incubation: Growth requires 7–12 days of incubation at 35–37°C in a humidified environment. Results cannot guide early management decisions.
Fastidious transport: The organism is sensitive to cold, heat, drying, and chemical inhibitors. Samples must reach the laboratory within hours of collection using appropriate transport media.
Lower sensitivity overall: Even with optimal technique, culture is less sensitive than PCR and becomes negative rapidly after antibiotic treatment begins.

Today, culture for B. pertussis is most commonly performed at public health reference laboratories for surveillance purposes — tracking strain characteristics, monitoring for antibiotic resistance, and confirming outbreak cases. In clinical practice, PCR has taken over.

When PCR Becomes Negative

One of the most important practical points about pertussis diagnosis is that PCR sensitivity declines sharply as the illness progresses. This timing limitation catches many patients off-guard:

Weeks 1–2: PCR sensitivity is highest — approximately 80–95% in well-performed studies
Week 3: Sensitivity begins declining, approximately 60–80%
Weeks 4–5: Sensitivity drops to 30–60% or lower
After week 6: PCR is often negative even in true pertussis cases

The critical implication: a negative PCR result does not rule out pertussis in a patient who has had a cough for more than 3–4 weeks. This is when serology should take over as the primary diagnostic tool.

Similarly, starting antibiotic treatment — even a single dose of azithromycin — can render both PCR and culture negative within days. If pertussis is suspected and antibiotics have already been started, PCR may still be positive for the first 5 days of treatment, but after that, serology becomes the main option.

Serology: The Late-Stage Test

Serological testing measures the body's antibody response to B. pertussis antigens in a blood sample. The most standardized and validated serological test is measurement of anti-pertussis toxin IgG antibodies (anti-PT IgG) by ELISA (enzyme-linked immunosorbent assay).

Serology is the right test when:

The cough has lasted more than 3–4 weeks and PCR sensitivity is declining
The patient received antibiotic treatment before a swab was taken
Retrospective confirmation of pertussis is needed (e.g., for epidemiological investigation)

Key points about interpreting pertussis serology:

Single high titer: A single anti-PT IgG level above 100 IU/mL (by WHO reference standard) in a person who has not received a pertussis vaccine in the past year is strongly suggestive of recent pertussis infection.
Paired acute/convalescent samples: A 4-fold rise in anti-PT IgG between an acute sample and a convalescent sample drawn 3–6 weeks later provides the most convincing evidence of recent infection.
Vaccination confounds interpretation: Recent Tdap vaccination produces antibody levels that overlap with infection levels, making serological interpretation unreliable for 12 months after vaccination.
Lack of standardization: Assays vary between laboratories and reference standards differ internationally, making cross-laboratory comparison difficult.

Serology is not routinely available in most clinical laboratories and is generally sent to public health or reference laboratories. Results may take 1–2 weeks.

The Lymphocyte Count Clue

While not a specific test for pertussis, a routine complete blood count (CBC) can provide an important diagnostic clue. B. pertussis toxin blocks lymphocytes from leaving the bloodstream, causing them to accumulate and driving up the total white blood cell count — a pattern called lymphocytosis.

In a patient with a prolonged paroxysmal cough, finding any of the following on CBC should immediately raise suspicion for pertussis:

Total white blood cell count above 15,000–20,000/μL in an adult
Absolute lymphocyte count above 10,000/μL
CBC differential showing 60–80% lymphocytes (instead of the normal 20–40%)

In infants with severe disease, these numbers can be dramatically higher — white cell counts above 50,000/μL with predominantly lymphocytes should prompt immediate consideration of pertussis and urgent evaluation for complications.

Important caveat: lymphocytosis is not specific to pertussis. It also occurs with viral infections (especially EBV/mononucleosis), other bacterial infections, and some medications. It is a supportive clue, not a standalone diagnosis. But in the right clinical context — a paroxysmal cough of 1–4 weeks duration — an unexpected lymphocytosis on CBC should trigger specific pertussis testing.

Chest X-Ray Findings

Chest X-ray is not a diagnostic test for pertussis itself, but it is commonly ordered in patients with significant respiratory illness to look for complications, particularly pneumonia. Several findings are associated with pertussis:

"Shaggy heart" sign: The classic description of pertussis on chest X-ray is irregular, frayed-looking borders around the cardiac silhouette, caused by inflammation and infiltrates along the bronchi and peribronchial tissues. The heart itself is normal — it is the surrounding inflammation that creates this appearance.
Peribronchial cuffing: Thickening of bronchial walls visible as "tram-track" or cuffed shadows around airways on the X-ray.
Patchy infiltrates: Areas of consolidation (bright white on X-ray) from pneumonia, typically in the lower lobes and perihilar regions.
Atelectasis: Collapsed lung segments from mucus plugging of airways, appearing as dense triangular shadows typically at the bases.
Air trapping and hyperinflation: From partial airway obstruction by mucus, the lungs may appear overinflated with flattened diaphragms.

None of these findings are specific to pertussis — other causes of bronchopneumonia can look identical. A normal chest X-ray does not rule out pertussis; many cases, especially in the early paroxysmal phase, will have normal or near-normal films.

Clinical Diagnosis vs Laboratory Confirmation

In some settings — particularly during known local outbreaks or in households with a laboratory-confirmed case — a clinical diagnosis of pertussis is reasonable and treatment should not be delayed waiting for laboratory results.

Clinical criteria for pertussis (as used by many public health authorities) typically require:

A cough lasting 14 days or more, AND
At least one of: paroxysms of coughing, inspiratory whoop, or post-cough vomiting

For a clinical case that is part of a laboratory-confirmed outbreak, the diagnosis is accepted without individual testing. For sporadic cases without a known source, laboratory confirmation is preferred and is required for public health reporting.

In practice, clinical diagnosis of adults and adolescents with pertussis is made more often than laboratory confirmation suggests, because many clinicians recognize the characteristic prolonged cough pattern and treat empirically. This is generally appropriate — the antibiotics used for pertussis (azithromycin, clarithromycin) are safe, and early treatment ends contagiousness and may shorten illness.

However, laboratory confirmation matters for public health: identifying a case triggers contact tracing, post-exposure prophylaxis for vulnerable household members (especially infants), and may identify the source of an outbreak that is spreading through a community. Always report confirmed and suspected cases to local public health authorities.

Key Research Papers

Faulkner A, et al. Comparison of molecular and serological methods for diagnosis of pertussis. J Clin Microbiol. 2016;54(12):2996–3000. PMID 27683540
Metzger K, et al. Pertussis diagnosis in infants — PCR versus serology and effect of prior antibiotic therapy. Pediatr Infect Dis J. 2018;37(5):422–427. PMID 29116998
Riffelmann M, et al. Nucleic acid amplification tests for diagnosis of Bordetella infections. J Clin Microbiol. 2005;43(10):4925–9. PMID 16207951
Tatti KM, et al. Novel multitarget real-time PCR assay for rapid detection of Bordetella species in clinical specimens. J Clin Microbiol. 2011;49(12):4059–66. PMID 21998424
Guiso N, et al. What to do and what not to do in serological diagnosis of pertussis. Clin Infect Dis. 2011;53(9):958–61. PMID 21940419
Kowalzik F, et al. Laboratory methods for diagnosing pertussis. Expert Rev Anti Infect Ther. 2013;11(9):979–85. PMID 24024918
Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical manifestations. Clin Microbiol Rev. 2005;18(2):326–82. PMID 15831828
Loeffelholz MJ, Thompson CJ. Diagnosis of pertussis: moving from culture to molecular methods. Clin Lab Med. 2014;34(2):281–95. PMID 24856530
van der Zee A, et al. A clinical validation of Bordetella pertussis and Bordetella parapertussis polymerase chain reaction. J Clin Microbiol. 1993;31(10):2745–50. PMID 8253979
Patel M, et al. Epidemiology of pertussis in the United States. Pediatrics. 2021;147(3):e2020008946. PMID 33608490

Back to Table of Contents

Connections