|Year : 2016 | Volume
| Issue : 4 | Page : 257-262
A study of bacteriological and antibiotic susceptibility profile of pediatric urinary tract infection with special emphasis on extended spectrum beta-lactamase production in a tertiary care hospital of Eastern India
Nupur Pal1, Kalidas Rit1, Somnath Naskar2, Simit Kumar3, Rajyasri Guhathakurata4
1 Department of Microbiology, Institute of Post Graduate Medical Education and Research, Kolkata, West Bengal, India
2 Department of Community Medicine, Burdwan Medical College, Kolkata, West Bengal, India
3 Department of Microbiology, R. G. Kar Medical College, Kolkata, West Bengal, India
4 Department of Microbiology, Calcutta National Medical College, Kolkata, West Bengal, India
|Date of Web Publication||15-Nov-2016|
Dr. Kalidas Rit
70B, T.C. Mukherjee Street, P. O. Rishra, Hooghly - 712 248, West Bengal
Source of Support: None, Conflict of Interest: None
Introduction: Urinary tract infection (UTI) is a common bacterial infection in children and associated with significant morbidity. Recently, UTI has become more difficult to treat because of the appearance of multidrug resistance pathogens. Objectives: The main aim of this study was to evaluate the changing trends of etiology, and antibiotic susceptibility pattern of urinary isolates in children of 0-12 year age group attending our hospital. Materials and Methods: A total of 930 urine samples were analyzed from children aged 0-12 years complaining of fever with or without urinary symptoms attending pediatric out- or in-patient of a tertiary care hospital from March 2011 to September 2012. In children under 2 years, urine sample was collected in a sterile bag and in above 2 years freshly passed clean-catch mid-stream urine was collected. All specimens were subjected to urine culture and sensitivity tests. Results: Significant bacteriuria was detected in 24.95% of patients. The most common pathogens isolated were Escherichia coli (60.34%), Klebsiella pneumoniae (20.63%), and Enterococcus spp. (5.17%). Extended spectrum beta-lactamase (ESBL) detection rate was low (3.77%). Most of the pathogens were susceptible to amikacin, piperacillin-tazobactam, nitrofurantoin, and imipenem. E. coli which was the main isolate was found to be most susceptible to imipenem (94.28%), amikacin (88.57%), nitrofurantoin (81.43%) and piperacillin-tazobactam (74.29%). Conclusion: This study reveals that most of the urinary isolates in pediatric age group are multidrug resistant, although ESBL detection rate is low. This suggests regular monitoring and modification of empirical therapy, and it should be validated by culture report to prevent morbidity associated with this disease.
Keywords: Antibiotics susceptibility, extended spectrum beta-lactamase, pediatric urinary tract infection, significant bacteriuria
|How to cite this article:|
Pal N, Rit K, Naskar S, Kumar S, Guhathakurata R. A study of bacteriological and antibiotic susceptibility profile of pediatric urinary tract infection with special emphasis on extended spectrum beta-lactamase production in a tertiary care hospital of Eastern India. Int J Health Allied Sci 2016;5:257-62
|How to cite this URL:|
Pal N, Rit K, Naskar S, Kumar S, Guhathakurata R. A study of bacteriological and antibiotic susceptibility profile of pediatric urinary tract infection with special emphasis on extended spectrum beta-lactamase production in a tertiary care hospital of Eastern India. Int J Health Allied Sci [serial online] 2016 [cited 2021 Sep 27];5:257-62. Available from: https://www.ijhas.in/text.asp?2016/5/4/257/194129
| Introduction|| |
Urinary tract infection (UTI) is the most common nosocomial infection worldwide. It is one of the most common infections encountered by clinicians in developing countries and is of major public health importance. UTI is the third most common bacterial infection in children in developing countries after those of the gastrointestinal and respiratory tract. Unlike occult bacteremia or severe bacterial illness (in infants and children) little attention has been focused on the identification of UTI in children, despite recent information that suggests a high prevalence of UTIs and significant associated morbidity in these patients. Quite often, child receives antibiotics empirically, without adequate evaluation for UTI. UTIs are often difficult to diagnose in young children because of nonspecific signs and vague symptoms. It is very important to diagnose the condition timely and accurately as it could be the first presentation of an underlying urological anomaly or it may in itself lead to significant morbidity from renal scarring, hypertension and eventually end-stage renal disease. Approximately 13%-15% of end-stage renal disease is thought to be related to UTI in childhood that was often unrecognized and therefore undertreated.
In community and hospital settings the etiology of UTIs and the antimicrobial susceptibility of uropathogens have been changing over the years., Factors such as the changing patient population, extensive use, and misuse of antimicrobial agents could all contribute to changes in the bacterial profile of UTI. Knowledge of the antimicrobial resistance pattern of common uropathogens according to local epidemiology is essential for providing clinically appropriate and cost-effective therapy for UTI., This study was undertaken to evaluate the changing trends of etiology, and antibiotic susceptibility pattern of urinary isolates recovered from UTIs in children of 0-12 years age group to assist the clinician for accurate diagnosis and prompt treatment of childhood UTI.
| Materials and Methods|| |
This study was conducted in the Department of Microbiology of a tertiary care hospital in Eastern India from March 2011 to September 2012. Necessary clearance from the Institutional Ethical Committee was obtained before the study. Informed consent form of all the patients was taken before process their clinical samples. Urine samples were collected from all 930 children complaining of fever with or without urinary symptoms. In children under 2 years of age urine was collected in a sterile bag and in above 2 years freshly passed clean-catch mid-stream urine was collected aseptically. Semiquantitative culture was performed by inoculating urine on blood agar, MacConkey agar, and cystine lactose electrolyte deficient agar media. Standard loop which carries 0.04 ml of urine was used and colony count was calculated by multiplying number of colonies with 250., The presence of 105 bacteria per ml of urine significant bacteriuria was indicated by the presence of 105 bacteria or more per ml of urine., Urine microscopy was done to see pus cells. All plates were incubated at 37° C for overnight under aerobic condition to obtain accurate colony count. Colony count was performed on blood agar. The number of colony forming units (CFUs) was multiplied by 100 to determine the number of microorganisms per milliliter in the original specimen. On culture of urine, a colony count of ≥105 /ml organisms of a single species or for each of two potential pathogens was considered significant. In addition, a pure culture of Staphylococcus aureus was considered significant regardless of the number of CFUs and the presence of yeast in any number was considered significant and identified to the species level. Cultures showing insignificant growth (except in some cases) were considered no growth and in cases of mixed growth of three or more pathogens or growth of nonpathogens were asked to send repeat sample after proper collection of urine. Identification of growth was done using standard biochemical methods. The antibiotic susceptibility testing was carried out using Kirby-Bauer disc diffusion method following guidelines of the Clinical Laboratory Standard Institute (CLSI). The antimicrobials tested for the Gram-negative bacteria were amikacin (30 µg), amoxyclav (30 µg), cephalexin (30 µg), ceftriaxone (30 µg), ofloxacin (5 µg), nitrofurantoin (300 µg), norfloxacin (10 µg), cotrimoxazole (25 µg), piperacillin-tazobactam (100:10 µg), and imipenem (10 µg). Moreover for Gram-positive enterococci isolates tested antibiotics were amoxyclav (30 µg), nitrofurantoin (300 µg), ofloxacin (5 µg), doxycyclin (30 µg), vancomyc in (30 µg), and linezolid (30 µg). Extended spectrum beta-lactamase (ESBL) production in Gram-negative bacilli and methicillin resistance in S. aureus (MRSA) was detected using CLSI guidelines. ESBL production in Gram-negative bacilli was detected by CLSI confirmatory test on the isolates which were resistant to ceftriaxone (30 µg), zone of inhibition ≤25 mm by above Kirby-Bauer disc diffusion method. ESBL detection was done by combined disc method on Mueller-Hinton agar using ceftazidime (30 µg) versus ceftazidime plus clavulanate (30 + 10 µg) (oxoid). The test was considered positive when an increase in the growth-inhibitory zone around the ceftazidime plus clavulanate disc was 5 mm or greater of the diameter around the disc containing ceftazidime alone. MRSA was detected by the cefoxitin disc diffusion method. The test was carried out on Mueller-Hinton agar using a 30 µg cefoxitin disc and incubated at 35°C for 18-24 h. An inhibition zone diameter of ≤21 mm was reported as methicillin-resistant (MRSA) and a diameter of ≥22 mm was considered methicillin sensitive.,,
The control strains used were Escherichia More Details coli American Type Culture Collection (ATCC) 25922, Pseudomonas aeruginosa ATCC 27853, and S. aureus ATCC25923. For ESBL test control strain used was Klebsiella ATCC 700603.
| Results|| |
A total of 930 children samples were collected of which 232 samples showed significant growth of single species or predominant growth among two species, giving rise the total incidence of 24.95%. Insignificant growth and polymicrobial growth were not included in the study group. Out of 930 samples, 636 urine samples from hospitalized patients and 294 from outpatients were collected. Among hospitalized patients, 178 (27.99%) showed significant growth and 458 (72.01%) samples were sterile. Among outpatients 54 (18.37%) showed significant growth, 240 (81.63%) samples were sterile. From 232 children (24.95%) were culture positive, of which 104 (44.83%) were males and 128 (55.17%) were females. Overall male and female ratio was 1:1.23 [Table 1]. In this study, Gram-negative bacilli were the predominant causative group of childhood UTI accounting 91.38%. Whereas Gram-positive cocci were 6.90%, and Candida spp. were 1.72%. E. coli 140 (60.34%), was the predominant etiological agent isolated. Other Gram-negative bacilli isolated were Klebsiella species 48 (20.63%), Proteus mirabilis 8 (3.45%), P. aeruginosa 6 (2.59%), Enterobacter species 4 (1.72%), Citrobacter species 4 (1.72%), and Acinetobacter species 2 (0.86%). Among Gram-positive isolates Enterococcus species 12 (5.17%) was predominant followed by Staphylococcus saprophyticus 4 (1.72%) [Graph 1 and [Table 2]].
|Table 1: Age - and gender - wise distribution of culture positive cases |
Click here to view
|Table 2: Distribution of culture positive cases according to microbial species in urine and extended spectrum beta lactamase producing Enterobacteriaceae |
Click here to view
In the present study, E. coli and Klebsiella isolates were almost equally resistant to amoxy-clav (91.43%, 93.75%, respectively), and 3rd generation cephalosporins - ceftriaxone (81.49%, 81.25%, respectively) and cephalexin (76.2% and 79.17%, respectively). Among other Gram-negative bacilli, the rates of resistance to cephalosporins including 3rd generation were also high >60%. Resistance to quinolones was increased among Gram-negative bacilli isolated from hospitalized patients and outpatients also. 91.31% of E. coli, 87.5% of Klebsiella species, 59% of P. aeruginosa, and 72% of Acinetobacter species isolated from hospitalized patients showed resistance to norfloxacin. Amikacin resistance rate among E. coli, Klebsiella, Proteus were 11.43%, 8.33%, and 0% respectively. In this study E. coli isolates were highly sensitive against nitrofurantoin as compared to Klebsiella (81.39% and 62.5%, respectively).
In this study, overall Imipenem resistance were 5.72% for E. coli, and 4.17% for Klebsiella pneumoniae, whereas, other Gram-negative uropathogens were found to be100% sensitive to Imipenem. Compared to other Enterobacteriaceae E. coli, Klebsiella showed higher resistance against antimicrobials [Table 3]. In the case of Enterococci spp., vancomycin and linezolid were 100% sensitive, nitrofurantoin sensitivity was 83.33%, for the other antimicrobials (amoxy-clav, doxycycline, and ofloxacin) sensitivity was <50% [Table 4]. ESBL detection rate was low (3.77%) of which Citrobacter prevalence was high 50% as compared to 4.17% for Klebsiella, and 2.86% E. coli isolates [Table 2].
|Table 3: Susceptibility pattern of organisms in different antimicrobials |
Click here to view
|Table 4: Sensitivity pattern of Enterococci spp. against different antimicrobials |
Click here to view
| Discussion|| |
The prevalence of culture-positive isolates observed in our study was 24.95% which was lower to the finding of 34.5% by Dash et al. and 36.6% by Mehta et al., However, this was higher as compared to another study conducted by Mohanty et al. (14.7%) at the All India Institute of Medical Sciences, New Delhi, Delhi, India. The similarities and differences in the type and distribution of uropathogens may result due to different age groups and environmental conditions of different geographical regions. The present study showed that females were more commonly infected with UTIs compared to males both in hospitalized (79.50%) and outpatients (71%). Male to female ratio was 1:1.23. Males outnumbered females during the 1st year of life with a ratio of 2.17:1. This is in full agreement with other studies.,,,, With increase in age, the positivity gap between male and female gradually reduced to give high culture positivity in female than male in the 5-12 years of age group (1:2.22). This found highly significant (P ≤ 0.02). Several studies in children above 1 year of age till 15 years reported female predominance with a variable ratio ranging from 6:1 to 1.33:1 depending on the different sample size and difference in age group being studied. During the first few months of life, uncircumcised male infants are at increased risk for UTIs than their female counterparts. The periurethral area was found to be more frequently and more heavily colonized with uropathogens, especially E. coli in uncircumcised infants than circumcised infants. In this study, more than 95% of the male infants were not circumcised. After infancy females are more prone to develop UTI because of shorter urethra.
In this study, E. coli was the predominant uropathogen, accounted for approximately 60.34% of all clinically significant urinary isolates and 66.04% of all Enterobacteriaceae. This is consistent with the findings of previous studies in which E. coli was the predominant pathogen isolated all around worldwide.,, Enterobacteriaceae have several factors responsible for their attachment to the uroepithelium. The Gram-negative aerobic bacteria colonize the uroepithelial mucosa with adhesins, pili, fimbriae, and P1 blood group phenotype. However, K. pneumoniae are rarely encountered in cases of community-acquired UTI., In this study, 20.69% of Klebsiella isolates were found among all uropathogens. Similar findings were observed in a study done by Akram et al. in 2007 where Klebsiella accounted 22%. This increased trend may be due to increased colonization of multidrug resistant Klebsiella pneumonia in hospital set-up.
On antimicrobial susceptibility testing, it was noted that both the Gram-negative bacilli and the Gram-positive cocci showed a significantly high resistance to the β-lactam group of antimicrobials similar to the findings of the study carried out in general population by Kothari and Sagar. This similarities observed may be due to transmission of genetic resistance factors such as plasmids between different age groups, facilitated through frequent interaction between children and adults. The overall resistance of E. coli to ofloxacin was 85.71%. Fluoroquinolones have a wide variety of indications, permeate most body compartments, and are mostly empirically prescribed, thereby accounting for the emergence of their resistance. Coming to the aminoglycoside susceptibility, amikacin resistance rate among E. coli, Klebsiella, Proteus were 11.43%, 8.33%, and 0% respectively, which was very similar to some studies., In this study, E. coli isolates were highly sensitive against nitrofurantoin as compared to Klebsiella (81.39% and 62.5%, respectively). This drug also exhibited low resistance rate in the major part of the world (0%-5.4%). This may be due to nitrofurantoin is reserved drug used almost exclusively for treating UTI.
Higher resistance rate to all the above antibiotics may be explained as uncontrolled empirical consumption of these antibiotics during the past decade in our region. In most developing countries, it is the standard practice to treat empirically with an antibiotic when a child presents to primary health care with a suspected UTI. Sometimes urine is tested for culture and sensitivity only if the illness does not respond to first-line antibiotics. This can lead to high reported resistance rate of these antibiotics. This problem could be removed if only incidence cases are included and systematic urine sampling is used. In this study, prevalence of ESBL producing organisms were Citrobacter freundii (50%), Klebsiella (4.17%), E. coli (2.86%) which was comparatively low prevalence of ESBL-producing E. coli and Klebsiella spp. when compared to adult studies, but similar to those reported by SENTRY investigators for US pediatric institutions in 2004.,,, Among Gram-negative organisms isolated from children in the SENTRY study, 1.9% of E. coli and 1.1%-3.2% of Klebsiella spp. were ESBL producers, and 0.87% and 1.3%, respectively in the USA study. A recent French study also reported relatively low rates of ESBL-production among Enterobacteriaceae (not including K. pneumoniae) at 1.4%. High rates of ESBL production have been reported for organisms isolated from high-risk settings in the US and internationally., The main risk factors for ESBL production in these children were previous exposure to antimicrobials (100%), prolonged hospital stay, severe illness (92.3%), and female gender (84.6%). Two recent studies reported the prevalence of ESBL-producing organisms in pediatric bloodstream infections in the US and Korea to be between 7-17.9% for E. coli and 18%-52.9% for Klebsiella spp., In this study out of total two ESBL producing E. coli, one was having congenital anomaly (posterior urethral valve with B/L Grade IV vesicoureteral reflux with severe hydronephrosis), another had past h/o recurrent UTI, admitted with impending renal failure and ultrasonography kidney/ureter/bladder suggested hyperechoic renal cortex in both kidneys. One ESBL positive C. freundii was also isolated from a premature infant with sepsis.
| Conclusion|| |
An alarming observation of this study is that most of the isolates are multidrug resistant. Most of the Gram-negative isolates are sensitive to Imipenem, Amikacin and Nitrofurantoin compared to the other antibiotics tested. These may be the drugs of choice for the treatment of childhood UTIs in this region, and these drugs should be used with caution so that they could be used in further years to come. In the present scenario, the changing etiological agents of UTIs and the increasing resistance of organisms clearly dictate regular monitoring and modification of empirical therapy are required. Empirical therapy should be validated by culture report to prevent the morbidity associated with the disease.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Bagga A, Sharma J. Urinary tract infections clinical features, evaluation and treatment. Pediatr today 2000;3:395-401.
Conway JJ, Cohn RA. Evolving role of nuclear medicine for the diagnosis and management of urinary tract infection. J Pediatr 1994;124:87-90.
New HC. Urinary tract infections. Am J Med 1996;100:S63-70.
Jones RN. Impact of changing pathogens and antimicrobial susceptibility pattern in treatment of serious infections in hospitalized patients. Am J Med 1996;100:S3-12.
Ferry S, Burman LG, Holm SE. Clinical and bacteriological effects of therapy of urinary tract infection in primary health care: Relation to in vitro
sensitivity testing. Scand J Infect Dis 1988;20:535-44.
Kass EH. Bacteriuria and pyelonephritis of pregnancy. Arch Intern Med 1960;105:194-8.
Collee JG. Laboratory strategy in the diagnosis of infection syndromes: Mackie and McCartney: Practical Medical Microbiology. 14 th
ed. London, UK: Churchill Livingstone Elsevier; 2006. p. 87-8.
Forbes AB, Sahm FD, Weissfeld SA. Bailey and Scotts Diagnostic Microbiology. 12 th
ed. St Louis: Mosby; 2007. p. 850-1.
Stamm WE. Urinary tract infections and pyelonephritis. In: Kasper DL, Braunwald E, Fauci AS, Hauser SL, Longo DL, Jameson JL, editors. Harrison′s Principles of Internal Medicine. 16 th
ed. New York: McGraw-Hill; 2005. p. 1713.
Gatermann SG. Bacterial infections of the urinary tract. In: Borriello SP, Murray PR, Funke G, editors. Topley and Wilson′s Microbiology and Microbial Infections. 10 th
ed. New York: Oxford University Press; 2005. p. 672-80.
Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing; 20 th
Informational Suppl. CLSI/NCCLS M100-S20. Wayne, PA: Clinical and Laboratory Standards Institute; 2010.
Swenson JM, Tenover FC, Cefoxitin Disk Study Group. Results of disk diffusion testing with cefoxitin correlate with presence of mecA in Staphylococcus
spp. J Clin Microbiol 2005;43:3818-23.
Lunacek A, Koenig U, Mrstik C, Radmayr C, Horninger W, Plas E. Unexpected multidrug resistance of methicillin-resistant Staphylococcus aureus
in urine samples: A single-center study. Korean J Urol 2014;55:349-53.
Dash M, Padhi S, Mohanty I, Panda P, Parida B. Antimicrobial resistance in pathogens causing urinary tract infections in a rural community of Odisha, India. J Family Community Med 2013;20:20-6.
Mehta M, Bhardwaj S, Sharma J. Screening of urinary isolates for the prevalence and antimicrobial susceptibility of enterobacteria other than Escherichia coli
. Int J Life Sci Pharma Res 2013;3:100-4.
Mohanty S, Kapil A, Das BK, Dhawan B. Antimicrobial resistance profile of nosocomial uropathogens in a tertiary care hospital. Indian J Med Sci 2003;57:148-54.
Jakobsson B, Esbjorner E, Sverker H. Minimum incidence and diagnostic rate of first urinary tract infection - Statistical data included. Pediatrics 1999;104:222-6.
Langley JM, Hanakowski M, Leblanc JC. Unique epidemiology of nosocomial urinary tract infection in children. Am J Infect Control 2001;29:94-8.
Chhetri PK, Rai SK, Pathak UN, Thapa JB, Devkota KC, Shrestha BO, et al
. Retrospective study on urinary tract infection at Nepal Medical College Teaching Hospital, Kathmandu. Nepal Med Coll J 2001;3:83-5.
Keah Sh, Wee E, Chng K, Keah K. Antimicrobial susceptibility of community-acquired uropathogens in general practice. Malays Fam Physician 2007;2:64-9.
Kiffer CR, Mendes C, Oplustil CP, Sampaio JL. Antibiotic resistance and trend of urinary pathogens in general outpatients from a major urban city. Int Braz J Urol 2007;33:42-8.
Das RN, Chandrashekhar TS, Joshi HS, Gurung M, Shrestha N, Shivananda PG. Frequency and susceptibility profile of pathogens causing urinary tract infections at a tertiary care hospital in Western Nepal. Singapore Med J 2006;47:281-5.
Manges AR, Natarajan P, Solberg OD, Dietrich PS, Riley LW. The changing prevalence of drug-resistant Escherichia coli
clonal groups in a community: Evidence for community outbreaks of urinary tract infections. Epidemiol Infect 2006;134:425-31.
Kumar MS, Lakshmi V, Rajagopalan R. Related articles, occurrence of extended spectrum beta-lactamases among Enterobacteriaceae
spp. isolated at a tertiary care institute. Indian J Med Microbiol 2006;24:208-11.
Akram M, Shahid M, Khan AU. Etiology and antibiotic resistance patterns of community-acquired urinary tract infections in J N M C Hospital Aligarh, India. Ann Clin Microbiol Antimicrob 2007;6:4.
Kothari A, Sagar V. Antibiotic resistance in pathogens causing community-acquired urinary tract infections in India: A multicenter study. J Infect Dev Ctries 2008;2:354-8.
Taneja N, Rao P, Arora J, Dogra A. Occurrence of ESBL and Amp-C beta-lactamases and susceptibility to newer antimicrobial agents in complicated UTI. Indian J Med Res 2008;127:85-8.
Aggarwal R, Chaudhary U, Sikka R. Detection of extended spectrum ß-lactamase production among uropathogens. J Lab Physicians 2009;1:7-10.
Fedler KA, Biedenbach DJ, Jones RN. Assessment of pathogen frequency and resistance patterns among pediatric patient isolates: Report from the 2004 SENTRY Antimicrobial Surveillance Program on 3 continents. Diagn Microbiol Infect Dis 2006;56:427-36.
Blaschke AJ, Korgenski EK, Daly JA, LaFleur B, Pavia AT, Byington CL. Extended-spectrum beta-lactamase-producing pathogens in a children′s hospital: A 5-year experience. Am J Infect Control 2009;37:435-41.
Raymond J, Nordmann P, Doit C, Vu Thien H, Guibert M, Ferroni A, et al.
Multidrug-resistant bacteria in hospitalized children: A 5-year multicenter study. Pediatrics 2007;119:e798-803.
National Nosocomial Infections Surveillance System. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2004, issued October 2004. Am J Infect Control 2004;32:470-85.
Al Muharrmi Z, Rafay AM, Balkhair A, Al-Tamemi S, Al Mawali A, Al Sadiri H. Extended-spectrum β-lactamase (ESBL) in Omani Children: Study of prevalence, risk factors and clinical outcomes at Sultan Qaboos University Hospital, Sultanate of Oman. Sultan Qaboos Univ Med J 2008;8:171-7.
Zaoutis TE, Goyal M, Chu JH, Coffin SE, Bell LM, Nachamkin I, et al.
Risk factors for and outcomes of bloodstream infection caused by extended-spectrum beta-lactamase-producing Escherichia coli
species in children. Pediatrics 2005;115:942-9.
Kim YK, Pai H, Lee HJ, Park SE, Choi EH, Kim J, et al.
Bloodstream infections by extended-spectrum beta-lactamase-producing Escherichia coli
and Klebsiella pneumoniae
in children: Epidemiology and clinical outcome. Antimicrob Agents Chemother 2002;46:1481-91.
[Table 1], [Table 2], [Table 4]