February 1997
NEW YORK — Evaluation and management of pediatric urinary tract infections (UTI) have changed dramatically over the past 30 years — and keeps on changing, according to Stanley Hellerstein, MD, pediatric nephrologist, The Children's Mercy Hospital, Kansas City, Mo., who recently spoke here at the Ninth Annual Infectious Diseases in Children Symposium.
Hellerstein attributed these changes to new knowledge and technology and evolving perceptions of the pathogenesis of kidney damage in patients with UTI.
![[bar]](/art/gradred.gif){kind=small}
When evaluating the child with UTI, the physician should separate those with fever from those with afebrile infection. A 2- to 5-year-old child with a first UTI presenting as clinical cystitis who responds promptly to antibiotics may be evaluated with a urinary tract ultrasound. If the ultrasound is normal, a cystogram is not needed, he said.
A child, up to age 5, with a febrile UTI should have an ultrasound and a voiding cystourethrogram (VCUG), but treatment for older children differs. A child older than 5 with an afebrile UTI who responds to antibiotics should have an ultrasound. If the ultrasound is abnormal, Hellerstein suggested performing a cystogram. This applies to children up to 13.
For teen-age girls with a first infection who have a single afebrile UTI, he suggested omitting imaging studies if there is prompt response to treatment. If infection reoccurs within three months, he suggested an ultrasound. If positive, he recommended a cystogram.
If a teen-age girl has a febrile UTI as her first infection, she should have an ultrasound and a VCUG because vesicoureteral reflux (VUR) is heredity. She has probably had reflux since she was born, but may not have had a UTI until she was a teen-ager and began having sexual intercourse.
A difference exists in the incidence of reflux in infant boys and girls. Hellerstein recommended using a contrast cystogram, instead of a nuclear cystogram, in all children younger than 2 and all boys.
"We're always more concerned with the younger boys because obstructive lesions must be ruled out," he said.
Hellerstein said recent unconfirmed studies have shown male neonates and infants with VUR in the first few months of life show better resolution than girls; girls 2 and older have more frequent VUR.
Hellerstein also warned against always using bagged urine as a diagnostic aid. He said urine collected by bag is acceptable for ruling out presence of an infection, but should not be used to diagnose a UTI because of the high risk of contamination.
"You can be wrong too often," he said. Bagged urine can be used in follow-up studies, if one recognizes that only a negative urine by bag is reliable.
As a rule of thumb, Hellerstein said he keeps infants up to age 2 who have acute pyelonephritis on up to six months of suppressive antibacterial therapy whether or not they have VUR. If reflux is present, the treatment is continued for a year after which follow-up imaging studies are obtained. He suggested an oral cephalosporin for antibacterial suppression in infants younger than 6 weeks of age. Until age 1, he uses trimethoprim-sulfamethoxazole, and then switches to nitrofurantoin. Nitrofurantoin is indicated for the treatment of clinical cystitis but not for acute pyelonephritis.
Although follow-up urine cultures are currently suggested for the first three months when managing a patient on suppressive antibiotics with VUR with UTI, that concept is being studied. No unexpected UTIs have occurred so far in this study.
"I think in the not-too-distant future, the recommendation may well be made that children who are being followed on suppressive antibiotics who have reliable caregivers who do not have any kind of illness ... don't need routine follow-up cultures," Hellerstein said.
During the 1950s, ruling out obstruction was the main objective, Hellerstein said. He quoted a famous study saying, "When we encounter children with UTI, we must search for stasis, an obstructive lesion in the urinary tract to explain the presence and recurrence of UTI."
By the mid 1950s, two major developments occurred. The first was use of quantitative urinary cultures to separate contamination from actual bacteruria from true bladder infection. The second was the discovery of the relationship between VUR and dilatation of the upper urinary tract in patients who had neurogenic bladders. Prior to that, VUR was thought to be caused by a congenital anomaly or an anatomic obstruction of the outlet.
At that time, cystourethrography became popular and this was applied to those with UTIs. This process showed doctors that VUR is present in 20% to 25% of young girls with first infection and is even more pervasive for those with reinfection.
Once VUR was determined as the cause of hydronephrosis in the neurogenic bladder, cystourethrography was used to evaluate other patients with unexplained hydronphrosis and those with UTIs, Hellerstein said.
With the use of cystourethrography, bladder neck obstruction became the primary cause of VUR and upper tract infection, and surgery was generally recommended. Although thousands of young girls had surgery, many physicians doubted the technique's validity because these cases really didn't meet the criteria for obstruction, Hellerstein said.
"The view at the time was that this was basically a plumbing problem," Hellerstein said. "The concept was that there was obstruction at the outlet, which caused VUR, and that we must fix that plumbing problem."
In 1968, a "coincidental observation" by a British radiologist led to increased understanding of the pathogenesis of pyelonephritis in children with UTIs and VUR. During a cystogram on an 8-year-old girl, the doctor observed a focal pyelotubular back-flow of urine into the kidney, or intrarenal reflux. In addition, a relationship was shown between the site of intrarenal reflux and focal renal scarring.
The next significant development was called the "Big Bang" theory of nephrology. This theory suggested that children who had potential for VUR and intrarenal reflux would have normal kidneys until first infection. First infection was the "big bang." The area of intrarenal reflux would then become infected and ultimately scarred. But if they had papillae that did not permit reflux into the pelvicaliceal system, they didn't have further damage in those areas, Hellerstein said.
Soon after, studies revealed that kidney damage was not caused by VUR alone, but by a combination of reflux and infection. Children with UTIs, with or without VUR, were followed for as many as 20 years and renal scarring occurred almost exclusively in those with VUR with a new infection, a theory which is still used.
"These findings are the underlying basis of our current management of children with VUR in which suppressive antibiotics are used based on the observations that, in general, you don't get scarring if you prevent infection in child with no obstruction, as long as they don't get infected," Hellerstein said.
He mentioned a small study of dialysis and transplant patients from Children's Mercy Hospital which revealed that none of the patients who went on to end-stage kidney disease because of a simple UTI in the absence of severe VUR or outflow obstruction.
"I think most small scars are of little clinical significance," he said, adding that renal scars should not alter the management of a UTI.
By the 1980s, two major causes of kidney infections associated with a UTI were known: a UTI combined with VUR and intrarenal reflux, known as reflux nephropathy; and, patients without VUR who were infected with p-fimbriated pathogenic Escherichia coli which caused kidney infection in patients who did not VUR.
He recalled the most significant breakthrough during the 1980s as the development of cortical imaging.
Cortical imaging localizes the site of infection using either Tc-labeled glucoheptonate or dimercaptosuccinic acid (DMSA). These substances are removed from blood flowing through the renal pericapillary circulation by renal tubular cells. The tracers are concentrated in the proximal tubular cells and provide a functional image of the renal cortical tissue.
"This was a gigantic breakthrough as a tool for evaluating children with UTIs because, literally, for the first time we had a convenient method which enabled us to correlate clinical findings with actual findings in the kidney," Hellerstein said.
Studies in piglets with experimentally-induced UTIs showed that cortical imaging had a sensitivity rate of 89% and specificity of 100%. There are no similar data on humans, but human studies have shown :
"This was a bit of a revelation," said Hellerstein. "The suggestion was made that p-fimbriated E. coli may be important to the febrile response to a UTI whether it's upper tract or lower tract."
These results may lead to a change in the current concepts about UTI because the clinical diagnosis often differs from cortical imaging diagnosis. Hellerstein said there are no studies that demonstrate the sensitivity and specificity of cortical imaging in humans.
For more information:
- Hellerstein S. Evolving concepts in the care of children with urinary tract infections. Presented at the Ninth Annual Infectious Diseases in Children Symposium. Nov. 24, 1996. New York.
You can express your
views on this article, or other relevant themes, in the
Infectious Diseases in Children
Specialty Forums.