Recurrent UTIs in women: why they return and what gets checked

A urinary tract infection that keeps coming back is a different problem from a single episode, and it is handled as one. Most people reading this have already had several courses of treatment, feel that nobody has looked for a reason, and want to know whether what is happening to them has a name and a work-up. It does. Recurrent urinary tract infection is a recognised pattern with its own definitions, its own list of explanations, and a sequence of investigations that is broadly similar wherever you are seen.

This article sets out how recurrent infection is defined and why the definition is not identical between guidelines, what makes the bladder vulnerable in the first place, the difference between reinfection and relapse and why that distinction changes the tests, the risk factors that shift across the reproductive years and after the menopause, what investigation actually involves, why bacteria found without symptoms are usually left alone outside pregnancy, and which preventive measures have something behind them. The findings that need emergency help, and those that need assessment the same day, are set out in their own section below, placed early rather than left to the end.

What counts as a recurrent urinary tract infection

Recurrent urinary tract infection is about a confirmed pattern of episodes rather than a tally of prescriptions, though the definitions themselves are looser than that sounds. The usual shape of the definition is a set number of episodes within a set period. NICE describes recurrent urinary tract infection as two or more episodes in six months, or three or more in twelve months. The wording differs between bodies and it matters: NICE's own definition sets no requirement that each episode be culture-confirmed, while some other definitions count only episodes confirmed on a urine culture. If your episodes were treated without samples, that is a reason to start sending them, not a reason to think you fall outside the definition.

The threshold is a practical device rather than a biological boundary. It exists to mark the point at which a doctor stops treating each episode as an isolated event and starts looking for a reason behind the pattern. Falling just short of the definition does not mean nothing is wrong, and meeting it does not mean something serious will be found. Because guidance genuinely differs here, it is reasonable to ask which definition your own doctor is working to, rather than assuming there is one agreed number.

Counting is harder than it sounds. Many episodes are treated remotely without a sample ever being sent, so the record in your notes is often incomplete, and symptoms that were never infection sit in the count alongside those that were. Writing down the dates of each episode, the symptoms, whether a sample was sent and what it grew gives a doctor far more to work with than an impression of frequent infections, and it is the single most useful piece of preparation for the appointment.

Why infections take hold, and why they return in some people

The female urethra is short, and its opening sits close to both the vagina and the anus, so the distance bacteria have to travel to reach the bladder is small. That piece of anatomy is the main reason urinary infection is far commoner in women than in men, and it is also why the pattern tends to begin around the time sexual activity begins and then again after the menopause. Nothing about it reflects hygiene, and nothing about it is a fault you can correct.

Most episodes are caused by organisms that normally live in the bowel and colonise the skin around the urethra before moving upwards. The important consequence is that the bowel is a reservoir that is never cleared. Treatment empties the bladder of bacteria; it does not change the reservoir. That is why a course can work perfectly well and be followed by another episode some weeks later. The second episode is usually a fresh colonisation from the same source rather than a failure of the first treatment.

Anything that leaves urine sitting in the bladder works in the same direction. Urine passed regularly carries bacteria out with it, so emptying is one of the body's own defences. Incomplete emptying undermines that: a prolapse pressing on the bladder, pelvic floor muscles that do not relax properly, a neurological condition affecting bladder function, or simply a long habit of rushing and never finishing. Holding urine for hours through a shift or a journey has a similar effect, and it is a common thread in people whose episodes cluster around work.

Finally, susceptibility differs between people in ways that are not yet fully explained. Some bladder linings appear to be colonised more readily than others, and laboratory and animal work has shown that bacteria can survive inside the cells lining the bladder rather than being cleared entirely. How much that explains in people is still being worked out, and it is not something that can be tested for in ordinary practice. This is one of the reasons two women with identical habits can have entirely different experiences, and it is worth knowing, because it undercuts the assumption that repeated infection must mean something was done wrong.

Symptoms that need emergency help, and symptoms that need assessment the same day

Most bladder infections stay in the bladder and settle. A smaller number travel upwards to the kidney, and a few go on to make someone seriously unwell, so the warning signs divide into two tiers rather than one.

Get emergency help now, by calling an ambulance or going straight to an emergency department, if any of the following is present:

• New confusion, drowsiness or difficulty speaking
• Severe shivering or shaking with a fever
• Fast breathing or breathlessness
• Skin that is mottled, blue-tinged or unusually pale
• A temperature below 36C
• No urine passed all day
• Someone who simply looks very unwell to you

Those are the signs that the whole body is reacting to infection, which can happen from a bladder infection as well as a kidney one, and they are not something to wait on. Short of that, if any of the following apply, arrange to be seen the same day rather than waiting for the next appointment, and do not start a course left over from a previous episode, because doing so can make the sample uninterpretable just when it matters most.

• Fever, shivering, or shaking chills
• A temperature that is unusually low is as significant a sign as a fever, and a temperature below 36C is covered by the emergency advice above rather than by this list
• Pain in the back or the side, below the ribs, rather than low down over the bladder
• Feeling sick, vomiting, or being unable to keep fluids down
• Visible blood in the urine
• Any urinary symptom at all during pregnancy, however mild, including burning, frequency, low abdominal pain or backache
• An unexplained change in an older person who cannot describe their symptoms, such as new agitation, incontinence that is suddenly worse, or being off their feet; new confusion or drowsiness is covered by the emergency advice above
• Symptoms that are getting worse quickly at any point, or that have not improved within 48 hours of starting treatment, or that return as soon as treatment ends
• Age 65 or over, where urinary infection is more likely to become serious and less likely to present in the usual way
• A weakened immune system, whether from a condition, chemotherapy, or medication that suppresses immunity, and diabetes of any kind, whether or not it is well controlled
• Urinary symptoms in anyone with a single kidney, a transplanted kidney, a known stone, reduced kidney function, or a catheter

Pregnancy deserves its own line because the rules change. Urine drains more slowly in pregnancy and infection reaches the kidney more readily, so a urinary symptom in pregnancy is assessed urgently through your maternity team, and bacteria found without symptoms are treated in pregnancy when they would be left alone otherwise. Visible blood in the urine also needs following up once the infection has settled: a small number of cases turn out to have another cause, including a bladder tumour, and that is the reason examination and imaging are advised rather than a reason to be alarmed.

Reinfection or relapse, and why the difference changes the work-up

Doctors separate repeat episodes into reinfection and relapse, and the distinction is not academic. Reinfection is a new episode from a fresh colonisation, usually after a symptom free interval of some weeks, and often with a different organism or a different sensitivity pattern on the culture. Relapse is the same organism returning quickly after treatment finishes, which suggests that somewhere in the urinary tract sheltered it from treatment. The first points towards triggers and defences. The second points towards a structural cause, and that changes which tests are worth doing.

Telling the two patterns apart is only possible if samples were sent. Without organism names and sensitivities across several separate episodes, nobody can say whether this is one unresolved infection or a series of new ones, and the work-up becomes guesswork. This is the practical reason a culture taken before treatment is worth so much more than a culture taken afterwards, and why it is worth asking for a sample to be sent even when you are confident you know what the problem is.

Risk factors in the reproductive years

Sexual activity is the commonest trigger in younger women, and the mechanism is mechanical rather than hygienic: intercourse moves bacteria from around the opening of the urethra into the bladder. It is not a sign of poor hygiene, it is not an infection passed between partners, and it is not something to feel responsible for. A new partner, or a change in frequency, often marks the beginning of a run of episodes, and that timing is useful history rather than something to leave out of the consultation.

Contraceptive choice matters more than most people expect. Spermicide, used alone or on a condom, disturbs the normal vaginal bacteria that make the area resistant to colonisation, and is associated with repeated urinary infection. A diaphragm can act in two ways at once, since it is usually used with spermicide and its rim can press on the urethra and leave the bladder emptying incompletely. If a run of infections began when contraception changed, say so. What to use instead is a discussion for your own doctor rather than a change to make on your own.

Other factors cluster in the same years. Long gaps between passing urine during shifts or journeys, drinking very little, recovery after childbirth or pelvic surgery, and anything that makes complete emptying difficult all contribute. A family pattern is also real: having a mother or a sister with repeated infections is associated with the same tendency, which points towards inherited differences in how readily the urinary tract is colonised rather than towards anything either of you did or failed to do.

The menopause, oestrogen and the tissue changes that follow

The second peak in repeated urinary infection comes after the menopause, and the explanation is oestrogen. Oestrogen maintains the thickness of the vaginal and urethral lining and supports the lactobacilli that keep the vaginal environment acidic. When oestrogen falls, the lining thins and becomes drier, acidity falls, lactobacilli decline, and bowel organisms colonise the area far more readily. The change is in the tissue and its resident bacteria, not in behaviour. The wider transition is set out in this article on the stages and symptoms of the menopause.

This also creates a diagnostic trap. Thinning of the tissue causes burning, urgency, frequency and discomfort passing urine in its own right, with no bacteria involved at all. Those symptoms are easily read as another infection, treated as another infection, and then counted as another infection, which is one of the ways the tally of episodes becomes inflated. A urine culture separates the two. That is why a culture matters most in exactly the group most likely to be treated without one.

Treatment for the tissue change itself exists and is prescribed. Replacing oestrogen locally in the vaginal tissue is the approach usually considered first where repeated infection follows the menopause, and guidance supports considering it once everyday measures have not been enough. It is framed as a shared decision rather than an automatic step, it is reviewed within about a year rather than continued indefinitely, there are specific considerations for anyone with a history of breast cancer, and in some countries its use for this particular purpose sits outside the licensed indications for these products, which is a point to raise with your own doctor. Whether it is suitable for you, in what form, and alongside what else, is a decision for your own doctor, taking your full history into account. This article deliberately does not say what to take or how much, because that is a prescribing decision and not an article's job.

Pregnancy, diabetes, stones, catheters and the bowel

Pregnancy alters the urinary tract both mechanically and hormonally. The ureters relax and widen, the enlarging uterus presses on the bladder, and urine drains more slowly, so bacteria that reach the bladder have longer to multiply and a clearer route upwards. This is why urinary infection in pregnancy is taken more seriously than the same infection outside pregnancy, why a urine culture is sent early in pregnancy to look for bacteria even without symptoms, and why bacteria found that way are treated. Any urinary symptom in pregnancy is a same day matter, as set out above.

Diabetes raises the risk by several routes at once: glucose appearing in the urine, changes in the immune response, and, where nerves are affected, a bladder that does not empty fully. Blood glucose that has drifted upwards unnoticed is worth excluding in anyone with a new run of infections, which is why some doctors add a blood glucose or HbA1c test when they are looking for an explanation, although no guideline makes blood tests a routine part of the work-up. The broader relationship between glucose handling and reproductive health is covered in this article on insulin resistance.

Anything that obstructs flow or offers bacteria a surface to live on behaves in the same way. A stone in the kidney or the bladder is both an obstruction and a shelter, and a relapsing pattern with the same organism is one of the classic reasons to arrange imaging. A catheter, short term or long term, bypasses the body's defences entirely and carries its own risk for as long as it stays in place, which is why catheters are reviewed and removed as early as is safe.

Constipation deserves a line of its own because it is so often overlooked. A loaded rectum presses on the bladder, interferes with complete emptying, and keeps a large reservoir of bowel organisms sitting close to the urethra. In children and in older adults, dealing with constipation alone sometimes changes the pattern of infections more than anything else attempted, and it is a reasonable thing to raise even when it feels unrelated to the urinary symptoms that brought you in.

What gets investigated, and when

The single most important investigation is a urine culture on a properly collected mid-stream sample, taken before treatment starts. A dipstick is a screening test, nothing more. It can be positive where there is inflammation without infection and negative where there is infection, and its reliability falls with age, which is why most bodies that have reviewed it advise against diagnosing urinary infection in older adults on a dipstick alone. A culture does something a dipstick cannot: it names the organism and reports which treatments it is sensitive to.

• Urine cultures taken before treatment in two or three separate episodes, so the organisms can be compared
• A written record of dates, symptoms, and whether each episode was sampled
• Blood tests where there is a specific reason for them, such as kidney function or a check of blood glucose, rather than as an automatic step
• A measurement of how much urine is left in the bladder after passing it, generally with a small ultrasound scanner
• An ultrasound of the kidneys, ureters and bladder where the pattern suggests a stone, an obstruction or incomplete emptying
• An examination of the vulva and vagina, which in practice identifies tissue changes and other explanations that a remote consultation cannot
• Referral to a urologist or urogynaecologist where the pattern relapses, where imaging is abnormal, where there is visible blood in the urine, or where the diagnosis stays unclear

Not everyone with repeated infections needs a scan, and guidance differs on exactly where to draw the line. There is broad agreement that a straightforward reinfection pattern in an otherwise well younger woman does not require imaging, and that relapse with the same organism, organisms associated with stones, visible blood in the urine, impaired kidney function or a history of urinary problems in childhood does. Between those two positions the decision is clinical, which is why two careful doctors can reasonably differ, and why it is fair to ask why a scan is or is not being arranged.

Timing is part of the work-up too. A sample sent while you are already being treated is often uninformative, and a sample sent weeks after symptoms have gone answers a different question from the one you are asking. The two timings answer different questions rather than competing. A culture before treatment is what names the organism and its sensitivities, which is why it remains the single most valuable sample. A second culture shortly after treatment finishes answers a narrower question that only matters if your pattern looks like relapse: whether the organism was actually cleared, or was never cleared at all. Ask when to produce the sample rather than guessing, since the answer depends on what your doctor is trying to establish.

Why a urine culture matters more than another empirical course

Treating episode after episode without ever sending a sample is the commonest avoidable problem in this whole area, and it causes two distinct harms. Neither is theoretical, and both tend to arrive quietly, over years, in people who were trying to do the sensible thing by getting symptoms treated quickly. Understanding them is the clearest argument for insisting that a sample is sent. Sending it does not mean waiting to be treated: the sample is taken before the first dose and treatment is started at the same visit if your doctor judges it necessary, so the result is there to guide the next decision rather than to delay this one.

The first harm is resistance. Every course selects for organisms able to survive it, and the organisms that survive are the ones left in the reservoir to cause the next episode. Resistance is why a treatment that worked reliably for years stops working, and why the organism found at a fourth episode can be harder to treat than the one found at the first. Antimicrobial resistance is treated as a major threat to health care by the bodies that monitor it, and treating urinary symptoms repeatedly without confirming infection is one of the ordinary everyday ways it is driven.

The second harm is misdiagnosis. Burning, urgency and frequency are not specific to infection. Bladder pain syndrome, an overactive bladder, oestrogen deficient tissue after the menopause, vaginal thrush, bacterial vaginosis, a sexually transmitted infection such as chlamydia, and sometimes a stone all produce a similar complaint. Someone treated repeatedly for infections they never had can go years without the diagnosis they actually have, and each unnecessary course adds side effects and resistance without touching the real cause.

Bacteria without symptoms: asymptomatic bacteriuria

A urine sample can grow bacteria in someone who has no urinary symptoms whatever. This is called asymptomatic bacteriuria, and it is common, particularly with increasing age and in people with a long term catheter, in whom it is close to universal. It also occurs in pregnancy at much the same rate as outside it, and is listed separately there not because it is commoner but because of what it can lead to. It is a laboratory finding rather than an illness, and it is one of the reasons people are told they have an infection when they feel entirely well. If this describes you, the question worth asking is which symptoms are being treated.

Outside a small number of defined situations it is deliberately not treated, and that is a considered position held by multiple bodies rather than a shortcut or an oversight. Treating it has not been shown to prevent later symptomatic infection, and it does not protect kidney function, while it does expose people to side effects and to the selection of resistant organisms. In other words, treatment here is known to carry harm without a matching benefit, which is why guidance says to leave the finding alone and watch for symptoms instead.

The accepted exceptions are pregnancy, where bacteriuria is screened for and treated because of what it can lead to for the kidney and for the pregnancy, and the period before certain urological procedures likely to breach the lining of the urinary tract. Guidance on other groups, including people with diabetes, older people in residential care, and people with a catheter, has converged on not treating, although individual bodies word their exceptions slightly differently and local hospital policies are not identical. Where you are told something different from what you have read, ask which situation you fall into.

Two practical points follow. Urine that is cloudy or smells strong is not in itself a reason to send a sample or to start treatment, because concentrated urine, dehydration and diet all do the same thing. And a positive culture in an older person with new confusion but no urinary symptoms is a well recognised trap: confusion has many causes, bacteriuria is common at that age, and treating the urine while the real cause goes unexamined is a known error rather than a cautious approach.

Urinary symptoms with a negative culture

A negative culture in someone with genuine symptoms is not a dead end, and it does not mean the symptoms are imagined. Collection explains some of these results: a sample taken after treatment has started, after drinking a great deal, or one that sat warm for hours before reaching the laboratory can all come back negative. Laboratories also report a colony count, and the threshold used to call a sample positive is not identical in every laboratory or every guideline, so a low count alongside clear symptoms is sometimes meaningful and worth discussing rather than dismissing.

Where cultures are repeatedly negative, the useful next step is to look again at the diagnosis rather than to treat harder. The conditions listed earlier account for most of these cases, and several of them respond to something quite different from treatment aimed at bacteria. Urinary symptoms alongside pain with sex, for instance, point towards the pelvic floor and the vulval tissue rather than the bladder. This is usually the point at which a referral is the most valuable thing a general practitioner can arrange, and it is reasonable to ask for one.

Everyday measures, what can be prescribed, and the evidence on cranberry and probiotics

The measures with the most sensible basis are unglamorous. They work by reducing how long bacteria sit in the bladder and how readily they are introduced in the first place, which is all that can realistically be influenced from home. None of them replaces finding out why the episodes are happening, and none of them is a reason to delay a culture when symptoms appear, but together they are what most guidance suggests trying alongside a proper work-up.

• Drink enough that urine stays pale through the day; increasing fluid intake is the measure with the most consistent support, particularly in people who normally drink little
• Pass urine when you first notice the urge rather than holding on, and allow enough time to empty fully instead of rushing
• Pass urine after sex, which helps flush out bacteria moved towards the bladder
• Deal with constipation, since a full rectum interferes with complete bladder emptying
• Wipe from front to back, and leave the normal vaginal bacteria alone: no douching, and no perfumed washes or sprays on the vulva or inside the vagina, since disturbing those bacteria is the part with a mechanism behind it
• Raise spermicide or diaphragm use with your doctor if a run of infections began when contraception changed

Beyond what can be done at home, there is a prescribed pathway, and it is worth knowing it exists so you can ask about it. Guidance sets out a sequence rather than a single answer: local oestrogen where the pattern follows the menopause; then, where that is not effective or not appropriate, preventive antibiotic options; and a non-antibiotic bladder treatment used as an alternative to antibiotics. Every step on that pathway is reviewed rather than continued indefinitely. Which of these is right, in what order, and whether any of it suits you is entirely a decision for your own doctor, and this article names no drug, no dose and no schedule on purpose. What matters here is that running out of home measures is not the end of the list.

Cranberry and probiotics are the two things asked about most often, and the honest answer is that the evidence is limited and contested. Reviews of cranberry products have reached different conclusions depending on which trials are included and how people who dropped out are handled. Guidance is not unanimous. ACOG says cranberry products may lower the risk, while noting that the amount and the duration needed are still being studied; NICE describes cranberry products as something some women who are not pregnant may wish to try while stating plainly that the evidence of benefit is uncertain and that there is no evidence of benefit in older women, and the NHS lists cranberry products among things there is some evidence may help prevent repeated cystitis, with a caution about their sugar content and a warning that cranberry should be avoided by anyone taking warfarin. Supplements sold for this purpose are outside the scope of this article, and whether any of them is appropriate for you is a question for your own doctor. Probiotics intended to restore vaginal lactobacilli are biologically plausible and have been studied, but the trials are small and inconsistent. This article does not recommend any product.

It is worth separating the measures above from the ones that circulate without support. Studies looking for a link between recurrence and baths, underwear fabric or extra washing steps have not found one, so although advice to wear cotton underwear and avoid tight underwear is widely given, including by the NHS, it is general comfort advice rather than something shown to prevent recurrence, and the assumption that repeated infection reflects poor hygiene is both wrong and unkind. If you have already changed everything you have read about and the episodes continue, the problem is not your diligence. It is that the cause has not yet been identified, and that is what the work-up described above is for.

Related Reading

• UTI after sex: why it happens and what reduces the risk
• Vaginal odour (vaginal odor): the common causes and what each pattern means
• What is menopause? The stages, symptoms and age range explained
• How Insulin Resistance Affects Fertility
• Pain during sex and pain after sex: causes explained
• UTI vs yeast infection: how to tell the difference

Sources and references

This article is supported by the following independent, authoritative sources:

• NHS (United Kingdom), urinary tract infections
• ACOG (American College of Obstetricians and Gynecologists), urinary tract infections
• NICE (United Kingdom), urinary tract infection (recurrent): antimicrobial prescribing
• CDC (United States), urinary tract infection basics
• MedlinePlus (US National Library of Medicine), urinary tract infections
• Cleveland Clinic, urinary tract infection