Why is PCR accurate for STI testing?
Because it looks for the organism's own genetic material and copies it until it cannot be missed, rather than inferring an infection from a symptom or waiting for it to grow in a dish.
The short answer. PCR, or polymerase chain reaction, detects a sexually transmitted infection by finding a fragment of the organism's DNA and copying that fragment many times over until there is enough to detect. It does not need the organism to be alive, it does not need much of it, and it does not need you to have symptoms. The United States CDC states that nucleic acid amplification tests, the family PCR belongs to, perform better than any other available test for chlamydia and gonorrhoea, and that laboratories should use them. They are roughly 20% to 35% more sensitive than the non-culture tests that came before them.
In short
- PCR finds the organism's genetic material rather than inferring an infection from a symptom, so it detects infections that produce nothing to see.
- It is sensitive enough to work on samples you collect yourself: a swab, vaginal with no speculum or penile, given together with a urine sample.
- It cannot see an infection acquired too recently, so a recent exposure needs a retest.
- It identifies the organism but does not test which treatments will work.
How PCR finds an infection
Every organism carries genetic material unique to it. PCR takes whatever is in your sample, locks onto the specific sequence belonging to the organism it is hunting, and copies that sequence. Then copies the copies. After enough cycles, a trace too small to see at the start has become a signal that cannot be missed.
Three consequences follow from that, and they are the whole reason PCR replaced what came before. It does not need the organism to be alive, so the sample does not have to be rushed to a laboratory in perfect condition. It does not need much material, so a small sample is enough. And it does not care whether you feel anything, because DNA is present whether or not it is causing symptoms.
Compared with what, exactly
Worth being clear about the alternatives, because two of them are commonly confused.
| Treating by symptom | Culture | PCR | |
|---|---|---|---|
| What it looks at | What you can feel or see | Whether the organism grows in a dish | The organism's genetic material |
| Finds silent infection | No | Poorly. Sensitivity for gonorrhoea drops to around 50% without symptoms | Yes. This is what it is for |
| Names the organism | No, it infers a group | Yes | Yes, and several at once from one sample |
| Needs a live organism | Not applicable | Yes, so collection and transport must protect it | No |
| Sample | None | Usually a clinician-taken swab | Self-collected swab plus urine, no clinician needed |
| Guides treatment choice | By protocol | Yes, including resistance | No, it identifies but does not test susceptibility |
Nobody in South Africa is offered a choice between PCR and culture. Culture is the laboratory method PCR replaced, and it appears in that table because of what it shows: its sensitivity for gonorrhoea runs 85% to 95% in acute infection and falls to around 50% in people without symptoms. That is the same pattern as symptom-based care, in a laboratory rather than a clinic. The older method fails hardest in exactly the group that has nothing to report.
The choice you actually face is between being treated on a symptom, a panel covering the four curable infections, and a wider PCR panel. We have written separately about what symptom-based care misses and about what a four-infection panel leaves out.
This is not a fringe position. The CDC's laboratory recommendations state that nucleic acid amplification tests perform better than any other available test for chlamydia and gonorrhoea on sensitivity, specificity and ease of transport, and that laboratories should use them.
Why you can collect the sample yourself
This is the part that changes the experience rather than the statistics, and it follows directly from the sensitivity. Because PCR needs so little material, the CDC notes that a vaginal swab can stand in for an endocervical specimen, and that first-void urine is equivalent to a urethral specimen.
In practice that means no speculum, no examination and no stirrups. At Epicentre both versions are self-collected and both take two samples: a swab, vaginal with no speculum or penile, plus a urine sample. Taken at a branch or from a home kit posted to you in unbranded packaging.
What PCR cannot do
Three limits, and a laboratory that does not tell you about them is not being straight with you.
It cannot see an infection you acquired last week. Every test has a window period between exposure and detectability, and testing inside it can return a negative in someone who is infected. After a recent exposure, test now for what can be seen now and retest 2 to 4 weeks later. Our quiz routes this automatically.
It cannot tell a doctor which treatment will work. PCR identifies the organism; it does not test how that organism responds. That still needs culture, which is why culture has not disappeared and should not.
Throat samples need their own panel, and their own caution. A genital swab and urine sample do not cover oral exposure, so oral testing is a separate screen. One study of pharyngeal gonorrhoea also found PCR sensitivity of 91.9% but specificity of only 71.8% at that site, because harmless bacteria living in the throat are close relatives of the one being looked for, and other amplification chemistries performed better there. If oral exposure is your question, ask for the oral screen by name and ask which method it uses.
One thing that is a strength here rather than a caveat: how much you can trust a positive depends partly on how common the infection is in the population being tested. In a country with the prevalence South Africa has, a positive is more likely to be a true positive than it would be in a low-prevalence setting.
Getting a PCR panel in South Africa
You do not need a doctor's referral if you are paying cash. Epicentre is a walk-in laboratory with branches in Observatory in Cape Town, Parktown North in Johannesburg and Hillcrest in Durban, and you book your own panel.
Testing can be anonymous: a pseudonym is accepted, no identity document is required, and cash is accepted. We do not bill medical schemes for these tests, so nothing goes to your scheme by default; what we give you instead is the paperwork carrying our practice number and the test codes, so you can claim back yourself if you want to. Home kits are available for every PCR panel, couriered free within 10km of a branch or pickup point and R300 anywhere in the country.
Large panel, 17 pathogens
R1,925The fullest PCR panel and the one most people choose. Self-collected swab and urine sample, with no speculum.
Sample: self-collected swab and urine, at a branch or from a home kit.
Results: 5 to 7 working days. Oral exposure needs a separate oral screen.
Smaller PCR panels, blood-based screens, standalone HIV testing and a UTI panel are all on the sexual health page. Prices are 2026 cash rates including VAT and may change; in-store pricing applies.
Find out what you actually have
Four questions and the quiz will tell you which panel fits, including whether you should be testing now, later, or both.
Frequently Asked Questions
Why is PCR accurate for STI testing?
PCR detects an infection by finding a fragment of the organism's DNA and copying it many times over until there is enough to detect. It does not need the organism to be alive, it does not need much of it, and it does not depend on symptoms. The CDC states that nucleic acid amplification tests, the family PCR belongs to, perform better than any other available test for chlamydia and gonorrhoea, and that laboratories should use them. They are roughly 20% to 35% more sensitive than the non-culture tests that preceded them.
Do I need an internal examination for a PCR test?
No. PCR is sensitive enough to work on samples you collect yourself: a swab, vaginal with no speculum or penile, given together with a urine sample. The CDC notes that vaginal swabs can substitute for endocervical specimens, and that first-void urine is equivalent to a urethral specimen. Home kits are available for every PCR panel.
What can a PCR test not tell you?
Three things. It cannot detect an infection acquired too recently, which is why a retest is advised 2 to 4 weeks after a recent exposure. It cannot tell you which treatments an organism will respond to, which still requires culture. And a genital swab and urine sample do not cover oral exposure, which needs its own oral screen; one study also found PCR specificity of only 71.8% for gonorrhoea at the throat because of cross-reaction with harmless throat bacteria.
How long do PCR results take?
Five to seven working days for sexual health PCR panels at Epicentre. Blood-based tests are faster. Results are sent to you directly, and nothing is reported to a medical aid unless you ask for it to be.
Do I need a doctor's referral in South Africa?
Not if you are paying cash. Epicentre is a walk-in laboratory and you can book your own panel. Testing can be anonymous: pseudonyms are accepted, no identity document is required and cash is accepted. Epicentre does not bill medical schemes for these tests, but provides the paperwork you need to claim back yourself, which generally requires a referring doctor's details.
What these terms mean
- PCR
- Polymerase chain reaction. A laboratory method that copies a tiny amount of an organism's genetic material many times over until it can be detected.
- NAAT
- Nucleic acid amplification test. The family PCR belongs to, recommended by the CDC as the method laboratories should use for chlamydia and gonorrhoea.
- Culture
- Growing an organism from a sample in the laboratory. Less sensitive than PCR, particularly without symptoms, but still needed to check which treatments an organism will respond to.
- Sensitivity
- How good a test is at finding an infection that is present. A test with low sensitivity misses real infections.
- Specificity
- How good a test is at correctly clearing people who are not infected. A test with low specificity produces false alarms.
- Window period
- The time between exposure and the point at which a test can reliably detect an infection. Testing inside it can produce a negative result in someone who is infected.
References
- Recommendations for the laboratory-based detection of Chlamydia trachomatis and Neisseria gonorrhoeae, 2014. CDC MMWR Recommendations and Reports 63(RR-02). Read it
- Comparison between conventional culture and NAATs for the microbiological diagnosis in gonococcal infection. Diagnostic Microbiology and Infectious Disease. Read it
- Nucleic acid amplification tests for diagnosis of Neisseria gonorrhoeae oropharyngeal infections. Journal of Clinical Microbiology. Read it
This article is general health information and not medical advice. Epicentre is a diagnostic laboratory: we provide testing and reporting, and we do not diagnose, prescribe or treat. If you have symptoms and cannot wait, a public clinic will see and treat you today at no cost. Please discuss any result that needs attention with a doctor.
