Urinary Tract Infections

Medically reviewed by David Shusterman, MD · Updated September 16, 2026

Most urinary tract infections are straightforward. A woman develops burning, frequency and urgency, takes a short course of an appropriate antibiotic, and is better within days. That pathway works, and for a first or occasional infection it does not need a specialist.

The reason this page is longer than that is the other group: women having three, four or six infections a year, patients whose cultures keep growing organisms resistant to what they were last given, men with infections at all, and people whose symptoms persist after treatment. Those situations are handled badly by repeating the original approach, and that is what most of them have had.

When a culture matters

A dipstick suggests; a culture proves and tells us what the organism is sensitive to. For an isolated, typical infection in an otherwise healthy woman, treating empirically without a culture is reasonable. For everything else — recurrence, treatment failure, men, pregnancy, recent antibiotics, recent hospitalisation, catheters, diabetes, immunosuppression — the culture should be sent before the first dose, not after the second failure. The most common reason a patient ends up in our office having taken four courses of antibiotics is that nobody obtained a culture at the start.

Recurrent infection is its own diagnosis

Recurrence is usually defined as two infections in six months or three in a year, and it splits into two mechanisms that need distinguishing. Reinfection is a fresh organism arriving from outside; relapse is the same organism returning because something is harbouring it — a stone, incomplete emptying, a foreign body, or a prostate in men. Comparing the organisms across cultures is how the two are told apart, which is another reason culture matters.

The evaluation for recurrence includes a post-void residual to confirm the bladder empties, imaging of the kidneys and bladder where relapse is suspected, and cystoscopy where there is blood, a suspicion of a stone or foreign body, or a persistent single organism. In postmenopausal women, vaginal atrophy is a major and very treatable contributor. In men, a urinary infection is never routine and deserves evaluation of the prostate and the upper tracts.

Prevention that has evidence behind it

Vaginal estrogen in postmenopausal women is the intervention with the strongest effect and the one most often not offered. It restores the vaginal environment that resists colonisation, it acts locally with minimal systemic absorption, and it reduces recurrence substantially. Many women have been told to avoid all estrogen after a breast cancer diagnosis; that is a conversation worth having with the oncologist rather than an automatic no.

Adequate fluid intake genuinely reduces recurrence in women who drink little — this was tested properly in a randomised trial rather than assumed. Methenamine hippurate, an old urinary antiseptic, was found non-inferior to daily antibiotic prophylaxis in a randomised comparison and avoids driving resistance; it is a good option for many patients. D-mannose and cranberry have modest and inconsistent evidence; they are low-risk and we do not discourage them, but we describe them accurately rather than as established therapy.

Behavioural measures — voiding after intercourse, avoiding spermicide and diaphragm use, treating constipation — are worth doing and are not the whole answer. Continuous or post-coital antibiotic prophylaxis works, and we reserve it for patients who have failed the non-antibiotic measures, because the resistance cost is real and lands on the same patient later.

Resistance, and why the fourth course fails

Resistance to trimethoprim-sulfamethoxazole and to fluoroquinolones is now high enough in many communities that empiric use is unreliable, and extended-spectrum beta-lactamase producing organisms are no longer confined to hospital patients. When a patient has had several recent courses, the organism is likely to be resistant to what worked before. This is precisely the situation where a culture, taken before treatment, converts guesswork into a decision. Occasionally the right answer is a short course of an intravenous or newer oral agent rather than another round of the same.

Asymptomatic bacteriuria is the other half of this. Bacteria in the urine without symptoms do not usually need treating, and treating them drives resistance without benefit. The exceptions — pregnancy and before certain urologic procedures — are specific. A positive culture in a patient who feels well is not automatically an infection.

How we see patients for this

Uncomplicated infections in patients who fit clear criteria can be handled through our virtual UTI and STI clinic, which is faster than waiting for an appointment and includes arranging a culture. Recurrent infection, any infection in a man, suspected kidney involvement and treatment failure are seen in the office, because they need examination, imaging and a plan rather than a prescription.