Urinary tract infections in girls: the cost-effectiveness of currently recommended investigative routines.

Current recommendations for the universal investigation of urinary tract infection (UTI) in children by ultrasonography, voiding cystourethrography, and dimercaptosuccinic acid renal scan (and sometimes intravenous pyelography as well) are not based on any convincing evidence as to the necessity or effectiveness of such a routine. Over 8% of all girls will have a UTI during childhood. About 87 individuals in a million will develop end-stage renal disease (ESRD) by the age of 60 years, caused in about 9% by pyelonephritis (PN) or reflux nephropathy. From these statistics, the maximal risk of a first diagnosed UTI progressing to ESRD is approximately 1:10,000. The risk of developing hypertension following a first UTI in childhood, without eventual evolution to ESRD, appears to be very small. The cost of the widely recommended routine imaging procedures ranges from U.S. $355 in Britain to U.S. $1,090 in the United States. The minimal cost of preventing a single progression to ESRD by early diagnosis of underlying pathology-if this were possible in all cases-would range between U.S. $5 million in Britain and U.S. $15 million in the United States. Since in many instances progressive renal damage can not be prevented, the true cost is considerably higher. Lower UTI in girls is a very common and, in most cases, benign finding in primary-care practice. It is suggested that girls with afebrile UTI, presenting with lower urinary tract symptoms alone, need not undergo any imaging procedures, but should be followed with urine examinations and cultures at the time of febrile illness. The recommended investigative routines should be reserved for UTI in infants and in girls with fever or other symptoms suggesting PN, and for proven recurrent UTI. Such a regimen will allow a marked saving in terms of costs and in terms of unnecessary radiation, psychological stress to children, and stress, inconvenience, and time loss to parents. There is no evidence that this approach will compromise the course or final outcome of this very common condition.