The prevalence of urinary tract infection (UTI) in children under five years old presenting with an acute illness in UK general practice

Urinary tract infections (UTI) in young children have been associated with serious long-term complications such as renal scarring, hypertension and renal failure. The presenting symptoms of UTI in children are non-specific. If UTI is not suspected, a urine sample is not obtained, and without this, UTI cannot be diagnosed. There is evidence that the diagnosis is often missed. Most published studies have not systematically sampled urine, and those that have are largely based in US emergency departments and only include highly selected groups of children. The true prevalence of UTI in acutely ill children presenting in UK general practice is therefore unknown. My thesis consists of a literature review discussing the association of childhood UTI with long-term complications, the challenges of diagnosis and the evidence that UTIs are being missed; a systematic review of papers reporting UTI prevalence in children which highlights the need for a study in UK general practice; a pilot study to determine the feasibility of recruiting children and obtaining urine samples in UK general practice; and a prospective cohort study to determine the point prevalence of UTI in 597 presenting children, determine the predictive value of presenting symptoms, signs and risk factors, and describe the clinical outcomes for children with UTI. I found that the prevalence of UTI was 5.9% (95% confidence interval: 4.3-8.0%). This may be sufficiently high to justify increased urine sampling in general practice. A multi-variable logistic regression model identified younger age range, pain on passing urine (dysuria) and urinary frequency as being associated with UTI. I propose a urine sampling strategy for GPs assessing acutely ill children and compare this to suspicion-led sampling and current guidelines. In my discussion I discuss the limitations, generalisability and implications of these findings