Renal cortical scintigraphy: effect on medical decision making in childhood urinary tract infection

Retrospective analysis of inflammatory parameters in acute pyelonephritis

OBJECTIVE

Different parameters were compared in children with acute pyelonephritis in order to facilitate the early detection of vesicoureteral reflux (VUR) in primary healthcare settings.

MATERIAL AND METHODS

This was a cross-sectional, hospital-based study. A total of 149 children with confirmed acute pyelonephritis were retrospectively analyzed between January 1999 and December 2000. The primary factors studied were body temperature, C-reactive protein level, white blood cell count and neutrophil ratio.

RESULTS

The study population comprised 82 males (55.0%) and 67 females (45.0%). The mean age of the patients was 2.7 +/- 3.9 years (age range 1 day to 18 years; 75% <3 years old). Of 123 patients who underwent voiding cystourethrography, 34 (27.6%) had VUR. When raised C-reactive protein, leukocytosis and raised neutrophil ratio occurred together, the specificity and positive predictive value for predicting VUR were obviously increased, but sensitivity was radically decreased. After controlling for the other covariates, multivariate logistic regression analysis showed that factors significantly related to VUR were age (odds ratio = 1.3, 95% confidence interval 1.02-1.67; p < 0.05) and raised neutrophil ratio (odds ratio = 4.2, 95% confidence interval 1.1-16.5; p < 0.05).

CONCLUSIONS

Our findings emphasize that the prevalence of VUR in children with acute pyelonephritis is extremely high. If a raised neutrophil ratio is observed, the potential risk of VUR is significantly increased in patients with clinically suspected acute pyelonephritis. Hence, the neutrophil ratio is recommended as an excellent parameter for predicting VUR.

UTI: diagnosis and evaluation in symptomatic pediatric patients

We retrospectively reviewed data on 260 hospitalized pediatric patients with symptomatic urinary tract infection (UTI). To ascertain the colony-forming units (CFU)/mL compatible with the diagnosis of UTI, a culture from a catheterized urine specimen containing >1,000 CFU/mL was considered diagnostic of UTI and resulted in imaging by renal ultrasound, voiding cystourethrography, and renal nuclear scan with Tc99m dimercaptosuccinic acid (DMSA). A positive DMSA renal scan is indicative of pyelonephritis. We used logistic regression analysis to determine which patient characteristics were predictive of pyelonephritis. We determined that, in hospitalized pediatric patients, the colony count of the positive urine culture, the type of organism grown in culture, and the voiding cystourethrography (VCUG) result (positive or negative for vesicoureteral reflux) did not predict which patients had pyelonephritis. In females, advancing age of the patient and positive renal ultrasound results were predictive of which patients had pyelonephritis when we controlled for other factors. We feel this emphasizes the importance of a thorough evaluation of hospitalized symptomatic patients, including patients with colony counts of 1,000 to 50,000 CFU/mL, to locate the level of infection and plan appropriate therapy.

Should renal ultrasonography be done routinely in children with first urinary tract infection?

To reassess the impact of renal ultrasonography on the care of children with first febrile urinary tract infection (UTI) we conducted a computer search and review of medical records of (1) all children who were admitted to our hospital with first febrile urinary tract infection and underwent renal ultrasonography during a 25-month period beginning February 1, 1995, (2) all children diagnosed by ultrasound to have hydronephrosis during the same time period. Of a total of 124 patients with UTI, renal ultrasound appeared normal or showed evidence of acute pyelonephritis in 105 (84.7%), and in another nine (7.2%) it showed only minor findings. In 10 children (8.1%) ultrasound showed hydronephrosis and/or hydroureter. In eight of the latter 10, voiding cystourethrography showed vesicoureteral reflux; in one, posterior urethral valves; and in one, who had a unilateral nonobstructed dilatated system, cystography appeared normal. Except for the last patient, who was given prophylactic antibiotics and continued to have urinary tract infections, in no other case did ultrasound alone have any impact on the patient's management. Four children with both abnormal-appearing renal ultrasound and voiding cystourethrography required surgical intervention. One hundred of the 124 children had a voiding cystourethrogram. In 38 children it detected vesicoureteral reflux and, in another two, bladder abnormalities. Thirty-five of those with abnormal-appearing cystogram but without an indication for surgery were given prophylactic antibiotics. During the same 25-month period, 63 children without urinary tract infection were diagnosed by ultrasound with hydronephrosis. In 45 of them (71.4%) the urologic abnormality had already been detected by prenatal ultrasound. Fourteen of these 45 children (31.1%) required surgery, all for congenital anomalies related to obstructive uropathy. We conclude that routine renal ultrasonography in children with first urinary tract infection has negligible influence on their clinical management. This seems to be due to the recent widespread use, in industrialized countries, of maternal-fetal ultrasonography, which already detects a significant number of children with congenital obstructive uropathy prenatally. On the other hand imaging of the lower urinary tract is of high yield and contributes significantly to patient care. Therefore, whereas imaging of the lower urinary tract should continue to be done routinely in children with first urinary tract infection, renal ultrasound may be reserved for more select cases.