Short or Long Antibiotic Course for Infant UTI?
Brady PW, Conway PH, Goudie A
Pediatrics. 2010;126:196-203
Many infants with urinary tract infections (UTIs) also have concurrent pyelonephritis, prompting guidelines to suggest that these children receive 7-14 days of antibiotic therapy. The mode of antibiotic delivery, intravenous (IV) vs oral; the duration of IV treatment; and the duration of total treatment are all subjects of debate, with recent data suggesting that early IV therapy or shorter courses of total treatment may provide benefits similar to longer total or longer IV courses.
Brady and colleagues conducted a retrospective cohort study, focusing on infants < 6 months old who were hospitalized for UTIs. The hypotheses were that the course of IV treatment varied greatly among the infants and that shorter treatment was not associated with adverse outcomes or treatment failure. The investigators used data from 24 freestanding children's hospitals in the United States, collected from 1999 to 2004. They excluded children with comorbid conditions that would affect outcomes (such as immunodeficiencies), but they did not exclude children with genitourinary anomalies. Children were also excluded if they had received IV antibiotics for > 14 days or who were hospitalized > 14 days because these likely represented children with complex medical problems. Length of IV therapy was evaluated both as a dichotomous (≤ 3 days or ≥ 4 days) and as a continuous variable in the analysis.
The median duration of IV therapy was 3 days. Treatment failure was defined as rehospitalization for a UTI-related reason within 30 days of the index discharge. The analyses accounted for the relatively infrequent occurrence of readmission. The investigators identified 12,333 study children, 1930 of whom (15.6%) had genitourinary anomalies and 1353 of those (70%) had genitourinary reflux. Boys were more likely to receive long-course therapy (56.5% of boys vs 43.5% of girls) as were black and Hispanic children as well as children with genitourinary anomalies. Treatment failure occurred in 240 (1.9%) of the children. The duration of IV antibiotic therapy was not associated with the frequency of treatment failure (1.6% of short-course children and 2.2% of long-course children experienced failure, but this difference was not significant after adjustment for other variables). In regression analyses, adjusting for all factors, the odds ratio (OR) for failure for long IV treatment vs short IV treatment was 1.02 (95% confidence interval [CI], 0.77-1.35). The findings were similar when the duration of antibiotics was tested as a continuous variable and when tested via Poisson regression. Known genitourinary anomalies were associated with increased risks for treatment failure (OR, 1.83; 95% CI, 1.20-2.79), as was increased severity of illness (measured by a composite score). Other interesting findings included large variation between hospitals in the percentage of children who were treated with long-course therapy (15%-87%). The investigators concluded that treatment failure of children initially treated intravenously in the hospital for UTI is uncommon and is not associated with the duration of initial IV antibiotic therapy.
Over the past 6 years, I have reviewed multiple studies for Medscape that have tested the assumptions of how we treat UTIs in children. In general, studies have shown that (1) prophylaxis of children with known genitourinary reflux may not prevent infections but only select for resistant organisms; (2) not all children with UTIs who are admitted to the hospital require IV therapy; and (3) relatively short IV therapy (eg, 3 days) may be as good as longer courses. The basic problem that still exists is that the prospective trials have generally had small enough enrollment that they don't serve as definitive studies, and those with larger samples -- such as this one -- tend to be retrospective. Our current approach has been established by expert opinion; therefore, we will likely have to rely on expert opinion after review of the newer data to see the guidelines revised again.
Abstract
Length of Intravenous Antibiotic Therapy and Treatment Failure in Infants With Urinary Tract Infections
Brady PW, Conway PH, Goudie A
Pediatrics. 2010;126:196-203
Study Summary
Many infants with urinary tract infections (UTIs) also have concurrent pyelonephritis, prompting guidelines to suggest that these children receive 7-14 days of antibiotic therapy. The mode of antibiotic delivery, intravenous (IV) vs oral; the duration of IV treatment; and the duration of total treatment are all subjects of debate, with recent data suggesting that early IV therapy or shorter courses of total treatment may provide benefits similar to longer total or longer IV courses.
Brady and colleagues conducted a retrospective cohort study, focusing on infants < 6 months old who were hospitalized for UTIs. The hypotheses were that the course of IV treatment varied greatly among the infants and that shorter treatment was not associated with adverse outcomes or treatment failure. The investigators used data from 24 freestanding children's hospitals in the United States, collected from 1999 to 2004. They excluded children with comorbid conditions that would affect outcomes (such as immunodeficiencies), but they did not exclude children with genitourinary anomalies. Children were also excluded if they had received IV antibiotics for > 14 days or who were hospitalized > 14 days because these likely represented children with complex medical problems. Length of IV therapy was evaluated both as a dichotomous (≤ 3 days or ≥ 4 days) and as a continuous variable in the analysis.
The median duration of IV therapy was 3 days. Treatment failure was defined as rehospitalization for a UTI-related reason within 30 days of the index discharge. The analyses accounted for the relatively infrequent occurrence of readmission. The investigators identified 12,333 study children, 1930 of whom (15.6%) had genitourinary anomalies and 1353 of those (70%) had genitourinary reflux. Boys were more likely to receive long-course therapy (56.5% of boys vs 43.5% of girls) as were black and Hispanic children as well as children with genitourinary anomalies. Treatment failure occurred in 240 (1.9%) of the children. The duration of IV antibiotic therapy was not associated with the frequency of treatment failure (1.6% of short-course children and 2.2% of long-course children experienced failure, but this difference was not significant after adjustment for other variables). In regression analyses, adjusting for all factors, the odds ratio (OR) for failure for long IV treatment vs short IV treatment was 1.02 (95% confidence interval [CI], 0.77-1.35). The findings were similar when the duration of antibiotics was tested as a continuous variable and when tested via Poisson regression. Known genitourinary anomalies were associated with increased risks for treatment failure (OR, 1.83; 95% CI, 1.20-2.79), as was increased severity of illness (measured by a composite score). Other interesting findings included large variation between hospitals in the percentage of children who were treated with long-course therapy (15%-87%). The investigators concluded that treatment failure of children initially treated intravenously in the hospital for UTI is uncommon and is not associated with the duration of initial IV antibiotic therapy.
Viewpoint
Over the past 6 years, I have reviewed multiple studies for Medscape that have tested the assumptions of how we treat UTIs in children. In general, studies have shown that (1) prophylaxis of children with known genitourinary reflux may not prevent infections but only select for resistant organisms; (2) not all children with UTIs who are admitted to the hospital require IV therapy; and (3) relatively short IV therapy (eg, 3 days) may be as good as longer courses. The basic problem that still exists is that the prospective trials have generally had small enough enrollment that they don't serve as definitive studies, and those with larger samples -- such as this one -- tend to be retrospective. Our current approach has been established by expert opinion; therefore, we will likely have to rely on expert opinion after review of the newer data to see the guidelines revised again.
Abstract
SHARE
