Cold sores: why they keep coming back, what treats them and does L-lysine help

Herpes simplex virus type 1 (HSV-1) is usually passed on in childhood through saliva and close contact, often without any noticeable symptoms. WHO estimates based on a systematic review suggest that in 2016 about 3.75 billion people aged 0-49 had HSV-1 infection at any site, or 66.6% of that age group. That means most people in the world carry the virus, although only some of them ever have a visible cold sore.

After the primary infection the virus travels along nerve fibres to the trigeminal ganglion, where it enters a latent state. In that state it does not continuously produce infectious particles and is largely invisible to the immune system. A recurrence means the virus multiplies again, travels along the nerve to the skin around the mouth and causes the typical blisters. The 2023 review by Gopinath et al. describes this as continuous transitions between a chronic latent phase and an acute recurrent phase, which allows the virus to evade immunity while spreading to other people.

The practical conclusion is simple and slightly uncomfortable: no treatment currently removes HSV-1 from the body. Every available strategy aims to shorten an episode, ease symptoms or reduce how often recurrences happen, not to cure the infection. The authors of the 2015 Cochrane review say plainly that there is no cure and that the disease recurs periodically.

The most commonly described triggers are ultraviolet radiation, febrile illness (hence the common name 'fever blister', although a cold is only one of several possible backgrounds), intense psychological stress and local trauma to the tissues around the mouth, for example after dental or cosmetic procedures. Hormonal changes, exhaustion and nutritional deficiencies are also mentioned, but the strength of evidence varies widely: for UV we have an experimental study with a control group, while for most other factors we rely on patient reports and observational studies.

This study is valuable for one reason: it shows causation rather than mere association. Participants received a controlled UV dose, so one factor was changed at a time. The limitation is equally important. Four minimal erythema doses is a strong laboratory exposure, not an ordinary walk on a sunny day. The 2015 Cochrane review pooled two experimental-irradiation trials (pooled RR 0.07 for recurrence with sunscreen) and set them against a trial in which participants were exposed to natural sunlight and no difference was found (RR 1.13; 51 participants). That trial was small, so it does not rule out a benefit, but it shows that translating a laboratory result into everyday life is not straightforward.

For stress and fever the mechanism is less precisely studied. Stress hormones and inflammation are thought to affect how immune cells keep the dormant virus in check, but there is no randomised evidence that a specific stress-reduction technique reduces the number of recurrences in most patients. We describe the effect of chronic stress on immunity more broadly in our knowledge base entry on chronic stress; here we only note that it is a plausible but hard-to-measure factor in any individual case.

What do these numbers mean for someone who has blisters on the lip right now? First, the effect is real but modest: we are talking about shortening an episode by about a day, against a typical healing time counted in days. Second, aciclovir and penciclovir creams perform similarly in the review, and the single trial comparing them directly (HR 0.88; 95% CI 0.67-1.16) found no significant difference. Third, oral valaciclovir has clearer evidence than oral aciclovir, but its use needs a doctor's decision, and the benefit is of a similar order to the creams.

The review also points out an important gap: in the main comparisons no study reported participant-rated improvement of symptoms, lesion severity or pain in a way that could be pooled. In other words, we know the drugs shorten time to healing, but we know much less about how much they reduce pain and discomfort, which are the things patients feel most. It is therefore reasonable to expect a moderate effect and to treat medication as one element of management, not as the solution to the problem.

The review also covers complementary approaches such as honey, zinc or photodynamic preparations, which earlier reviews often left out. We do not carry conclusions for those forms over here without the detailed product-by-product comparisons, because the individual comparisons rest on small trials. If someone offers a 'natural' cold sore product, the control question is whether there is a randomised trial with a placebo group and a measured healing time, not just user testimonials.

Antivirals act on viral replication, so the most can be gained by starting them in the prodromal phase (tingling, burning or itching at the spot where blisters will appear a few hours later). This principle is common in guidance and makes mechanistic sense, but to be honest we are not basing it here on a single trial comparing early with late start. The conclusion comes mainly from understanding the biology of the infection: once the blisters are fully developed the virus has already multiplied, and the drug has less to do.

Why does this result appear in the article when the group is so small? Because it illustrates a direction that has been studied for years: a cold sore is to a large degree an inflammatory response to the virus, so combining an antiviral with an anti-inflammatory drug is theoretically justified. At the same time, 42 people who actually had a recurrence during follow-up are too few to use such a combination routinely. A corticosteroid on herpetic lesions without medical supervision is not recommended, because in some herpes infections (especially around the eye) it can make the course worse.

The result of the review can be summed up in three sentences. Chronic daily use of an antiviral (so-called suppressive treatment) reduces the number of recurrences, but the size of the benefit is small, and the decision to start it belongs to a doctor and is usually reserved for people with very frequent or severe recurrences. Smearing a cream on 'just in case' before an expected recurrence has no proven preventive effect. The other methods (thymopentin, low-level laser therapy, hypnotherapy) have at most one or two trials, and conclusions are hard to draw.

What remains, then, for someone who does not want to take a drug long term? Eliminating the triggers you can control. The best-supported measure is limiting strong sun exposure on the lips and using a lip balm with UV filter in situations that provoked recurrences before (skiing, sailing, long beach days). Where data on natural sunlight are mixed, the argument for sunscreen is its low cost and lack of harm, not certainty of effect. The rest is sensible management of general health: sleep, treating coexisting infections, avoiding injury to the lips. We describe the link between sleep and immunity in our entry on sleep and immunity.

Where does the popularity of lysine come from? From a simple, attractive mechanism: the HSV virus needs the amino acid arginine to multiply, and lysine is supposed to compete with it. Something like this has been seen in laboratory conditions, but the road from a cell culture to a real reduction in recurrences in people is long and, as controlled trials show, has not been travelled. Until well-designed trials with a large group and a clearly defined outcome appear, the honest statement is: the effectiveness of lysine in preventing herpes labialis is not proven.

The effectiveness of home measures for easing discomfort, such as a cold compress, a protective ointment or painkillers, is not the subject of the main studies we cite here; they are used symptomatically.

It is worth adding that not every lesion on the lip is a cold sore. Angular cheilitis (cracks at the corners of the mouth), impetigo, contact dermatitis or canker sores (which occur inside the mouth, not on the vermilion of the lip) can look similar to an untrained eye, and their treatment is different. If episodes seem to repeat in the same place and always with a tingling phase and blisters, that is the typical picture of a cold sore; if not, the diagnosis should be confirmed. When herpes of the eye is suspected, the decision to be seen urgently should not wait for a 'let's see whether it passes'.

Cold sores are one of those problems where the market for products promises more than the research shows. The realistic picture is this: the virus cannot be removed, drugs shorten an episode by roughly a day, a UV filter makes sense where sun has provoked recurrences, and lysine has no support in controlled trials. That is not a reason to give up treatment, only to set expectations: the most is gained by reacting quickly to the first symptoms, avoiding infecting others and recognising the situations that need a doctor.