Abstract
Urinary tract infection (UTI) is common in infants and children, and Escherichia coli is the leading pathogen. The aims of this study were to compare first episode of UTI with recurrent infection, reveal organisms that cause UTI, uropathogen resistance, and presence of bacteria producing extended-spectrum β-lactamase (ESBL). The first-UTI group included 456 children. E coli was the leading pathogen (80.5%), and Pseudomonas aeruginosa was found in 1.5%. The uropathogens were resistant to gentamicin (3.41%) and cefuroxime (5.71%), and highly resistant to cefamezin (37.39%). The recurrent-infection group included 106 children. E coli was also the leading pathogen, but 7.5% of the isolates were P aeruginosa (P = .002 compared with first-episode group); 6.6% were ESBL-producing bacteria compared with 1.1% in the first-episode group (P = .002). E coli is the leading pathogen in both groups. P aeruginosa and ESBL-producing bacteria were more common in the recurrent infection group.
Introduction
Urinary tract infection (UTI) is a common disease in pediatric patients in the first years of life and it is associated with acute or late complications. Escherichia coli is the leading causative pathogen, followed by Klebsiella pneumoniae, Proteus mirabilis, Pseudomonas aeruginosa, and Enterococcus species.1,2 Antimicrobial treatment of UTI is divided into 2 groups: (a) oral treatment, which includes first-, second-, and third-generation cephalosporins, amoxicillin/clavulanate, and trimethoprim/sulfamet-hoxazole and (b) parenteral treatment, such as gentamicin, tobramycin, second- and third-generation cephalosporins and piperacillin in infants and children 2 to 24 months old who are hospitalized as a result of UTI. 3 Early antimicrobial treatment for children with acute infection can reduce renal damage, 4 whereas recurrent infection can cause an increase in renal scars. 5 The indications for hospitalization of infants and children with suspected UTI are the following: children suffering of vomiting and dehydration, those who are ill appearing or are unable to drink and take medication, and also in cases where urosepsis is suspected. Therefore, for children who are admitted with a suspected UTI, a proper urine culture should be obtained and intravenous antibiotic treatment and fluids can be initiated while awaiting the results of urine culture. Knowledge of the pathogens that cause UTI in pediatric patients and their resistance patterns can ensure efficient and adequate empiric treatment in children with first UTI or recurrent infection. The policy in our pediatric department is to give gentamicin once daily as empiric therapy for infants and children who are hospitalized with community-acquired pyelonephritis.
In the past decade, several reports have shown a worrisome trend of increasing resistance of uropathogens to antibiotics, including the production of extended-spectrum β-lactamase (ESBL) by E coli, K pneumonia, and other Enterobacteriaceae.6,7 Risk factors for infections due to ESBL produced by E coli or K pneumoniae include underlying disease, hospitalization, recurrent infections, use of antibiotics within the prior 3 months, and children receiving prophylaxis therapy with cephalosporins.8,9 Carbapenem remains the best treatment option for infection with ESBL-producing bacteria. Other options, such as piperacillin/tazobactam or aminoglycosides, can also be suggested to avoid overuse of carbapenems. 10
The aims of this study were to compare first-episode UTI with recurrent infection and reveal organisms that cause UTI, uropathogen resistance patterns, presence of bacteria producing ESBL, and to check whether the empirical use of gentamicin is still suitable as an initial treatment for first episode of UTI and also for recurrent infections in hospitalized infants and children.
Methods
A retrospective study included children younger than 18 years with UTI who were hospitalized at our medical center in the north of Israel from January 2003 to December 2010. The study was divided into 2 major groups: the first group included children with first episode of UTI; in this group we excluded children with recurrent infection, known abnormal malformations in the urinary tract, or vesicoureteral reflux. Children being treated with antibiotics before admission were also excluded. The second group included children with recurrent UTI who were also treated as inpatients. In a comparison of these 2 groups, the common variables were fever, age, gender, white blood cell values, urine analysis, urine culture, and resistance patterns of the detected uropathogens. The mean variable was resistance of organisms to the commonly used antibiotics. Urine samples were collected by suprapubic aspiration in neonates, urethral catheterization till the age of 3 years, and clean-catch specimen collection by the midstream method in older children. Diagnostic criteria for UTI included the following: any growth of Gram-negative bacilli, more than a few thousand colony forming units (CFUs) per milliliter (CFU/mL) of Gram-positive cocci from urine taken by suprapubic aspiration, more than 104 CFU/mL of a single pathogen for urine taken by urethral catheterization and more than 105 CFU/mL for clean-catch midstream samples. Antimicrobial susceptibility testing was performed using a Vitek 1 (Biomerieux) instrument in accordance with the Clinical Laboratory Standard Institute susceptibility criteria. Categorical variables were presented as frequencies and percentages, and continuous variables were presented as mean ± SD. Differences between patients with first episode of UTI and patients with recurrent episodes were determined by T test or Wilcoxon 2-sample test. Association between categorical variables and the study groups was performed by χ2 test or Fisher’s exact test. Significance was considered at P < .05.
Results
The first-episode UTI group consisted of 456 children, 83.3% females with a mean age 22.6 ± 36.2 months (median 11 months). The second group of patients, with recurrent UTI, included 106 children, 91.5% were female with mean age 45.5 ± 36.4 months (median 32 months; Table 1). The urinary analysis results were similar in both groups. The urine analysis was positive for leukocytes in 78.3% in the first group and in 78.3% of the cases in the second group. The nitrite analysis was positive in 41.7% and 48.1%, respectively. Hematological results and kidney function tests are shown in Table 1.
Data From First-Episode and Recurrent Urinary Tract Infection Groups.
In the first group, E coli was the most frequent pathogen found in the urine cultures, followed by K pneumoniae and Enterococcus faecalis; in the recurrent UTI group E coli was also the leading uropathogen followed by P aeruginosa, P mirabilis, and K pneumoniae (Table 2).
Incidence of Different Pathogens in First Versus Recurrent Urinary Tract Infection.
In the second group of patients, those with recurrent UTI, 51 (48%) presented with a second episode of UTI, 11(10.3%) presented in a third episode, and in 44 (41.5%) a fourth or higher episode.
We compared the bacteriological results of the organisms from each group; E coli was the leading pathogen in both groups, with similar frequency (80.5% vs 74.5%, respectively; Table 2). P aeruginosa was found at a higher rate in the recurrent group 7.5% versus 1.5% in the first-episode group, being the second most common pathogen in the recurrent UTI group (P = .002).
The resistance patterns of all the organisms in both groups are presented in Table 3. In both groups we have found a low resistance to gentamicin (P = .560). Of note, 64.95%, 38.17%, and 25.40% of the organisms were resistant to ampicillin, amoxicillin/clavulanate, and trimethoprim/sulfamethoxazole, respe-ctively in the first episode group, without significant statistical difference when compared with the second group.
Resistance Patterns of all Uropathogens in First Versus Recurrent Urinary Tract Infection to Different Antimicrobial Agents.
More uropathogens were resistant to cefuroxime in the recurrent-episode versus first-episode group (P = .002) and high resistance to cefamezin between the 2 groups (P = .048; Table 3). No correlation was found between recurrent episodes of UTI and the development of bacterial resistance to aminogly-cosides, but such a correlation was found with second-generation cephalosporin (Table 3). The number of ESBL-producing organisms in the 2 groups was 12: 6 E coli, 2 K pneumoniae, 2 Enterobacter cloacae, 1 Enterobacter aerogenes, and 1 Morganella morganii. Four of these 12 organisms (25%) were resistant to gentamicin and the other 8 organisms were sensitive; without resistance to amikacin, ofloxacin, or carbapenems. There were more ESBL-producing isolates in the recurrent-episode group (6.6%) than in the first-episode group (1.1%; P = .002; Table 2).
Discussion
The aims of this study were to compare first-episode UTI with recurrent infection and reveal organisms that cause UTI, uropathogen resistance patterns, and presence of ESBL-producing bacteria. The results show resistance rates of 37.39% and 5.71% to first- and second-generation cephalosopin, respectively, in patients with first-episode UTI, whereas in patients with recurrent UTI, the respective resistances were 48.9% and 17.5%, respectively. Marcus et al 11 found that 93.8% of the pathogens in UTI are Gram-negative bacteria, with E coli being the leading pathogen, and that 24%, 7%, 5% of the pathogens are resistant to first-, second-, and third-generation cephalosporins, respectively. Another report showed high resistance (36%) of E coli to first-generation cephalosporin. 12 Our results show an even higher rate of resistance to first-generation cephalosporins. This might be because this drug has been used as an empiric treatment in the community for the past 2 decades in our region, and it is still used as prophylactic therapy after UTI for patients with vesicoureteral reflux.
Aminoglycosides such as gentamicin or amikacin are among the antibiotics of choice for parenteral treatment and empiric therapy. In the current study, we found a low level of resistance to aminoglycosides, and the use of gentamicin was adequate for both first UTI and recurrent infection, the latter having more Pseudomonas and Gram-negative ESBL-producing bacteria. Several reports support the efficacy and safety of using a once-daily dose of gentamicin,13-15 and this antibiotic can be used with patients who have risk factors for antibiotic resistance. 16 Yüksel et al 6 reported a 25% resistance rate to gentamicin in children younger than 12 months, higher than the rate found in our report, and they suggested that amikacin is more suitable for children with UTI in the first year of life, whereas gentamicin is adequate for children who are older than 1 year. Prelog et al 17 suggested that infants and critically ill patients suffering from community-acquired first or recurrent UTI be treated parenterally with aminoglycosides or third-generation cephalosporin, each in combination with a β-lactam antibiotic. Aminoglycosides are required for urological abnormalities and for patients with a high rate of Pseudomonas infection. The active spectrum of aminoglycosides is more specific to Gram-negative bacteria, and they reach higher levels in kidney tissue than the third-generation cephalosporins. Based on our results and experience, empiric treatment with gentamicin can be suggested as an adequate treatment for first-episode UTI and for recurrent UTIs, in agreement with other reports.
Despite the low rate of resistance to nitrofurantoin, clinical cure with this antibiotic is low compared with the traditional treatments; it can be explained by the inadequate tissue concentration of nitrofurantoin in the renal parenchyma. Then nitrofurantoin cannot be useful for empiric treatment in pediatric patients who present with pyelonephritis.
A worrying problem that was found in our report is the high resistance rate to ampicillin, ampicillin/clavulanate, and trimethoprim/sulfamethoxazole (64.95%, 38.17%, and 25.40%, respectively). These data indicate that these antibiotics are inadequate and insufficient as empiric treatments for UTI.
Although ampicillin cannot be suggested as an empiric treatment for UTI, it is used against Enterococcus species. Several reports have shown results similar to our findings.1,6,17-19
In our region, ampicillin/clavulanate is considered an adequate empiric treatment for UTI; however, when we compared the resistance of UTIs caused by E coli from 1999 to 2000; data published in 2003 by our group with the results of the present study, we found an increase in the resistance rate to ampicillin/clavulanate from 10% to 38.17%. 20 A similar result was recently reported by Chakupurakal et al, 21 that is, 48% of the E coli UTI were resistant to ampicillin/clavulanate.
Reports from Austria, 17 the United States,1,19 and Belgium 22 have also revealed high rates of uropathogen resistance to trimethoprim/sulfamethoxazole, in partial accordance with our and others’ results.
An emerging cause of resistance in E coli, K pneumonia, and other Enterobacteriaceae is the production of ESBL, which causes resistance to β-lactam antibiotics, as found in the present study, mainly in patients with recurrent infections (6.1%). All cases of ESBL-producing bacteria in our report were treated with aminoglycosides. These patients suffered from recurrent infections and most of them received prophylaxis therapy.
In children, treatment with ertapenem has been reported to be a successful therapy for infection with ESBL-producing Gram-negative organisms in complicated UTIs. 23 In contrast, a report from Greece by Tratselas et al 7 compared the outcome of patients with UTI caused by ESBL-producing bacteria and those with UTI caused by non-ESBL-producing bacteria. More case patients received inappropriate therapy than controls. Clinical, microbiological outcome and formation of renal scars did not differ between the ESBL positive and non-ESBL groups. We suggest that more research is needed to determine the best treatment option for children with UTI caused by ESBL-producing bacteria.
Conclusions
Escherichia coli is the leading pathogen in first-episode and recurrent UTI. According to our results, aminoglycosides and second-generation cephalosporin can still be the first treatment of choice to treat first-episode UTI, while gentamicin or amikacin should be used to treat patients with recurrent UTI. The uropathogens found in this report show high resistance rates to amoxicillin, cefamezin, and amoxicillin/clavulanate. First-generation cephalosporin is not recommended for use as empiric therapy both in first or recurrent UTI. For oral empiric treatment, we recommend the use of cefuroxime. P aeruginosa and ESBL-producing bacteria are more common in recurrent episodes of UTI, and then aminoglycosides still can be appropriate treatment. Each center has to reevaluate periodically their common pathogens and the sensitivity to the common used antibiotics to pinpoint an efficient empirical therapy for children with first or recurrent UTI.
Footnotes
Declaration of Conflicting Interests
The authors(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.