- Email: [email protected]
A 4-year prospective study of septicemia in pediatric surgical patients at a tertiary care teaching hospital in Kuwait
Journal of Pediatric Surgery (2011) 46, 679–684
www.elsevier.com/locate/jpedsurg
A 4-year prospective study of septicemia in pediatric surgical patients at a tertiary care teaching hospital in Kuwait Eiman M. Mokaddas a,b,⁎, Shama A. Shetty b , Aneesa A. Abdullah b , V.O. Rotimi a a
Department of Microbiology, Faculty of Medicine, Kuwait University, PO Box 24923, Safat 13110, Kuwait Department of Laboratory Medicine, Ibn Sina hospital, Kuwait
b
Received 29 July 2010; revised 28 November 2010; accepted 1 December 2010
Key words: Microbiology; Septicemia; Pediatric surgery; Tertiary care hospital; Kuwait
Abstract Background: Critically ill children are at high risk for developing nosocomial infections that contributes to death in 4% of all pediatric intensive care unit admissions. This prospective study was undertaken to determine the prevalence of septicemia in the pediatric surgery department of a large tertiary care teaching hospital in Kuwait and to evaluate the risk factors, the microbial etiology, and the antimicrobial susceptibility pattern of the isolated microorganisms. Methods: All patients admitted to the pediatric surgery department from January 2001 until December 2004 with the diagnosis of septicemia were included in the study, and the microbiologically proven cases were then analyzed. The patients' demographics and risk factors for sepsis were recorded. All positive blood cultures were subjected to identification and antimicrobial susceptibility testing by VITEK 2 (bioMerieux, Marcy l'Etoile, France). Results: Of 3408 patients suspected to have septicemia, 78 (2.3%) patients developed microbiologically documented septicemias, 26% of those were low–birth weight patients, and 82% were patients with congenital anomalies; 87% of those needed surgical intervention. More than 50% were admitted to the intensive care unit, and 80.5% needed ventilatory support. Fifty-seven percent had early onset septicemia. Gram-positive and gram-negative bacteria accounted for 54% and 39% of the septicemia cases, respectively, whereas Candida spp was responsible for 7%. More than 50% of the staphylococci were resistant to cloxacillin, and all gram-positives were uniformly susceptible to glycopeptides and linezolid. Gram-negative bacteria showed variable resistance to cephalosporins (65%), piperacillin/tazobactam (29%), and carbapenems (11%). The attributable mortality rate for these septic episodes was 19% mainly because of gram-negative bacteria and Candida. Conclusion: The main etiologic agents of neonatal septicemia were coagulase-negative Staphylococcus, Pseudomonas aeruginosa, and members of the family Enterobacteriaceae. Empirical therapy with piperacillin/tazobactam or carbapenems for gram-negative septicemia and glycopeptides for gram-positive septicemia was effective. © 2011 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Department of Microbiology, Faculty of Medicine, Kuwait University, PO Box 24923, Safat 13110, Kuwait. Tel.: +965 498 6512. E-mail address: [email protected] (E.M. Mokaddas). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2010.12.001
Bloodstream infections represent a major cause of hospital-acquired infections in neonatal and pediatric intensive care units (ICUs). According to the National
680 Nosocomial Infection Surveillance Report published in 1999, bloodstream infections accounted for 28% of the total infections, followed by ventilator-associated pneumonia, 21% [1]. They are more likely to occur in the ICU than in non-ICU patients [2]. Neonatal sepsis is one of the major causes of morbidity and mortality in the newborn in general, particularly in the premature neonates [3]. Immature host defense mechanisms, underlying diseases, low birth weight, and immaturity together with invasive life support systems make the premature neonates particularly vulnerable to overwhelming infections [4]. Other risk factors include maternal colonization with Streptococcus agalactiae during labor and premature rupture of membranes [5].Various bacterial pathogens can cause septicemia in this age group. Grampositive bacteria, mainly staphylococci and streptococci, together with gram-negative bacteria, namely, the Enterobacteriaceae and Pseudomonas aeruginosa, are the major groups of pathogens [6]. Fungi, in particular, Candida spp, are reported to cause septicemia in around 1% of infants admitted to the neonatal ICU and 2% to 4.5% of low–birth weight neonates with an overall mortality of 50% [7]. Isolation of an organism from the blood culture of a neonate with clinical symptoms of infection constitutes the criterion standard for diagnosing neonatal sepsis [8]. In such a lifethreatening condition, knowledge of the likely causative organism and appropriate empirical antibiotic treatment is associated with a better chance for survival, regardless of other risk factors for fatality [9]. Early treatment with appropriate antibiotics would minimize both the risk of mortality and the emergence of multidrug resistant organisms in the ICU [3,10]. However, one of the confounding problems associated with empirical treatment in the ICU is the emergence of antibiotic resistance. Extended-spectrum β-lactamases (ESBLs), most commonly found in Klebsiella pneumoniae and Escherichia coli, have increased markedly in the past decade, particularly in the ICU setting [11]. Unfortunately, the detection of ESBL in any microbiology laboratory is only about 40% to 50% of the ESBL-producing gram-negative bacteria. Because these strains may be fairly susceptible to the third-generation cephalosporins in vitro, clinical failure may be as high as 75% with such antibiotics [11]. This study was undertaken to determine the prevalence of septicemia and evaluate the demographic data, underlying risk factors, and the microbial etiology in the pediatric surgery department of a large tertiary-care teaching hospital in Kuwait.
E.M. Mokaddas et al. pediatric surgery patients. In this institute, there is no medical pediatric service, and all the patients were referred from other general hospitals for surgical opinion and interventions. All patients with provisional diagnosis of septicemia during a period of 4 years spanning January 2001 to December 2004 were included in the study. The patients were admitted either to the ICU or the general ward. Patients suspected of having sepsis were identified based on the presence of one or more clinical signs, that are, lethargy, refusal of feeds, irritability, vomiting, respiratory distress, fever more than 38°C or hypothermia less than 36°C, leucocytosis (total peripheral leukocyte count of more than 15,000 cells/L), rise in C-reative protein (CRP), and then were subjected to septic workup. Two sets of blood (2 mL) sample were collected from each patient and inoculated into pediatric bottles and incubated in the BACTEC 9240 (Becton Dickinson, Sparks, MD) automated blood culture machine. Incubation was continued until the bottle signaled or up to a maximum of 7 days when the culture was declared negative. For each microbiologically proven septicemia case, the details of patient's demographics are recorded on a protocol data form. These included sex, age, gestational age at birth, and birth weight. All risk factors for developing sepsis such as underlying medical conditions, surgical procedures, intravascular catheterization, endotracheal intubations, and mechanical ventilation as well as the length of hospital stay were also documented. All positive cultures were subcultured onto appropriate media, and gram-stained smears were prepared. Identification and antimicrobial susceptibility testing of all isolates were done using VITEK 2 automated systems (bioMerieux, Marcy l'Etoile, France). Supplementary antimicrobial susceptibility tests were performed by the disc diffusion technique on Mueller Hinton agar (Oxoid, Basingstoke, UK). Any discrepancy in the above 2 methods were resolved by the E-test (AB Biodisk, Solana, Sweden). Extended-spectrum β-lactamase–producing gramnegative bacteria were detected by the ESBL E-test method that used both ceftazidime-ceftazidime/clavulanate and cefotaxime-cefotaxime/clavulanate strips [12]. After blood culture collection, all patients who developed clinical signs of septicemia were started on empirical antimicrobial therapy according to the antibiotic policy and guidelines of the pediatric surgery department, pending results of microbiology investigations.
2. Results 1. Materials and methods 1.1. Patients This study was performed at a tertiary care center in Kuwait, which is the only health care facility catering to
A total of 8303 patients were admitted to the pediatric and neonatal ward during the study period. The average admissions per year in the ICU and general wards for the 4-year period were 135.5 and 2365.3 patients, respectively. Of the 3408 patients suspected to have septicemia, according to the previously mentioned criteria, 78 (2.3%)
Prospective study of septicemia in pediatric surgical patients Table 1
The demographic profiles of patients with septicemia
Demographics Sex Male Female Age group b1 mo 1-6 mo 6-12 mo N1 year Gestational age Preterm Full term Birth weight b2 kg 2-3 kg N3 kg
No. of patients with sepsis (%) 54 (69) 24 (31) 31 (40) 28 (36) 10 (13) 9 (11) 29 (37) 49 (63) 20 (26) 37 (47) 21 (27)
developed 89 microbiologically documented septic episodes. Various patient's demographics are demonstrated in Table 1. The underlying diseases of these patients were analyzed and showed that 64 (82%) were born with surgical congenital anomalies, as shown in Table 2. Of the 78 patients ,68 developed sepsis after surgical intervention, Table 2 Distribution of different underlying clinical conditions of the septic patients Clinical condition
No. of patients
Necrotizing enterocolitis Oesophageal atresia Intestinal atresia Volvulus + malrotation Congenital diaphragmatic hernia Hirschprung disease Hydronephrosis Vesico-uretral reflux Anorectal malformation Exomphalos Gastroschisis Idiopathic hypertropic pyloric stenosis Undescended testis Inguinal hernia Appendicitis Vaginal atresia Tumor Cellulitis Osteomyelitis Choanal atresia Duodenal atresia Intusseception Cloacal extrophy Gastroenteritis Foreign body gut Constipation Total
16 9 7 6 3 3 3 2 2 2 1 2 2 1 2 2 2 3 2 1 1 1 1 1 1 2 78
681 Table 3 The risk factors associated with septic episodes in the 78 patients with septicemia Risk factors
No. of patients (%)
ICU stay Surgical intervention Ventillator support Lines Central line Arterial line Tubes and drains Nasogastric tube Endotracheal tube Peritonial drain Urethral catheter
41 (52) 68 (87) 33 (42) 24 (31) 5 (6) 61 32 6 60
(78) (41) (8) (77)
and only 10 developed septicemia without any surgical intervention but had some other underlying risk factors, as shown in Table 3. Furthermore, of the 78 septicemic cases, 41 (52%) were admitted to the ICU, and 33 (42%) of those needed ventillatory support. For those admitted to the ICU, the average hospital stay was 15.6 days, with an average of 9.8 days on ventillatory support. The prevalence of septicemia among the ICU patients was 6.3%, whereas on the ward, it was 0.5%. The correlation between the onset of septicemia in relation to the timing of hospital admission is shown in Table 4. Of the 78 patients, 45 (57.7%) acquired early onset septicemia within the first 7 days of admission. Of these 45, 35 (77.8%) acquired the infection within 48 hours. Thirtyfive percent and 75% of patients in the ICU and ward, respectively, developed early onset episodes compared to 64.7% of ICU patients and 25% of ward patients who had late onset septicemia. The distribution of different etiologic microbial agents of the septic episodes is shown in Table 5. Gram-positive bacteria predominated, accounting for 48 (54%) of 89 isolates. Gram-negative bacteria, on the other hand, accounted for 35 of the isolates (39%). Candida spp accounted for the rest (7%). Of the 89 episodes, 4 had polymicrobial etiology. The antimicrobial susceptibility of the gram-positive isolates showed uniform susceptibility to glycopeptides and linezolid, whereas 60% of all staphylococci were resistant to cloxacillin. A very high proportion (50%) of the
Table 4 Correlation between onset of septic episodes and timing of hospital admission Onset of septic episode
Within 48 h of admission 3-7 days of admission N7 days after admission Total
No. of patients ICU
Ward
Total no. (%)
9 3 22 34
26 7 11 44
35 (45) 10 (13) 33 (42) 78 (100)
682
E.M. Mokaddas et al.
Table 5 Distribution of etiologic microbial agents in various septicaemic episodes Microorganisms
No. of isolates (%)
Gram-positive Coagulase-negative staphylococci S aureus E faecalis V streptococci Gram-negative E coli K pneumoniae Citrobacter spp Enterobacter spp P aeruginosa Acinetobacter spp Stenotrophomonas maltophilia Sphingomonas paucimobilis Fungi C albicans C parapsilosis
48 37 2 5 4 35 4 9 1 1 10 7 1 2 6 4 2
(54) (77) (4) (10) (8) (39) (11) (28) (3) (3) (28) (19) (3) (6) (7) (67) (33)
Staphylococcus aureus isolates were resistant to cloxacillin (methicillin-resistant Staphylococcus aureus [MRSA]). Gram-negatives, on the other hand, showed variable susceptibilities to third generation cephalosporins, especially P aeruginosa and Acinetobacter spp, with an average resistance of 35%. An average resistance to aminoglycosides was 29% and 14% each to gentamicin and amikacin, respectively. Piperacillin-tazobactam and carbapenems showed an average resistance of 29% and 11%, respectively. Extended-spectrum β-lactamases were detected in 5 of the total 15 Enterobacteriaceae. Of the 78 patients who developed septicemia, 15 (19%) died. Thirteen of these deaths occurred after single or multiple surgical interventions, and 2 patients died before any surgery. The birth weight in these mortality cases ranged from 1.09 to 3.01 kg. Ten of these infants were preterm. Seven (47%) were necrotizing enterocolitis (NEC) with perforation, 3 (20%) were cases of intestinal atresia, 2 (13%) were gangrenous volvulus, and 1 case each of Hirchprung disease, anorectal malformation, and gastroschisis. The 13 mortality cases, all except the patient with gastroschisis, were reoperated .Various surgical interventions were carried out such as laparotomy, ileostomy, and resection of the gangrenous part of intestine. All of the 13 patients who underwent surgical intervention developed sepsis and died between the first and third week postoperatively. Nine (60%) of those cases had gram-negative septicemia owing to Acinetobacter spp (3), P aeruginosa (3), K pneumoniae (2), and Citrobacter spp (1). Four gram-positive septicemias (27%) were caused by coagulase-negative Staphylococcus (2) and one each of Enterococcus and Viridans streptococcus. Candida albicans accounted for 2 cases (13%).
Parenteral amikacin and piperacillin/tazobactam were empirically started, which had to be replaced with meropenem either because the organism was an ESBL-producing Enterobacteriaceae or because of multidrug-resistant Acinetobacter or P aeruginosa. Candidemia was treated with parenteral fluconazole. In 11 patients (73%), death was because of multiorgan failure and disseminated intravascular coagulation, and the other 2 patients (13%) went into renal failure after septicemia.
3. Discussion The data generated in this study highlighted the prevalence of pediatric septicemia among general admissions in a tertiary care hospital in Kuwait, the major predisposing factors, admission rates to the ICU and wards in relation to onset of infections, and the type of causative agents and their susceptibility pattern. In our study, the prevalence of septicemia was very low at about 1%. Different records on the prevalence of septicemia in neonates in general and pediatric surgical patients in particular have been reported. In one retrospective study by Bhattacharya et al [13] on a group of 608 patients from the pediatric surgical department for a 14-month period, 14 patients (2.3%) developed septicemia. Neonates in this study were at an increased risk for septicemia (4.2%) compared to older children and infants (3.1%) [13], which could be attributed to impaired host defenses of neonates [14-16]. In general, the data presented in this report were comparable to other reports [13-16]. In our study, neonates and infants 6 months or younger accounted for most of the 40% and 36% of patients, respectively. The main risk factors of septicemia were low birth weight and preterm deliveries. These figures of 26% and 57%, respectively, in our study contrast the study done in Norway where early onset sepsis among very low–birth weight infants are infrequent, and the incidence ranged between 1.5% and 2.7% [17,18]. However, in another study from India, both very low birth weight and prematurity were very important predisposing factors for the development of septicemia [6]. According to a report by the National Institute of Child Health and Human Development in 1996, culture-proven septicemia was reported in only 1.9% of very low–birth weight infants [19,20]. Thus, in our center, both low birth weight and prematurity are significant contributory risk factors for developing septicemia. Other very important factors are underlying medical and congenital problems, insertion of central venous catheters [20], and especially those who needed surgical interventions. Intensive care unit stay in our center predisposed to late onset septicemia while ward admission were the opposite with predominantly early-onset type. This may be partly due to the fact that ICU patients tended to acquire nosocomial infections more than ward patients who may have carried
Prospective study of septicemia in pediatric surgical patients community-acquired infections into the ward. The observation of health care–related septicemia in the ICU is also confirmed by the report of Singh-Naz et al [21] that indicated that pediatric ICU patients who undergo surgical operations are 2.5 times more likely to develop health care–related infections than nonoperated patients. Instrumentation and other manipulations such as insertion of central venous lines and use of endotracheal tubes and urinary catheters leading to device-associated infections [22] are known to be associated with lowering of host defense system. This is supported by the high percentage of patients with these devices in our study. The prevalence of 6.3% of septicemia among the ICU patients is similar to the report of a 9-month study done at St Louis Childrens' Hospital, Missouri, in which 57 (6.25%) of 911 patients admitted to the pediatric ICU developed septicemia. In this study, 50% had arterial catheters, 12% had multiple central venous catheters, and 70% were mechanically ventilated [1], most of which are discordant with our present findings where the major risk factors were surgery 67 (87%), insertion of nasogastric tubes 61 (81%), and urethral catheters 60 (77%). The leading cause of microbiologically proven septicemia in our patients was gram-positive cocci in which coagulase-negative staphylococci, predominantly S epidermidis, were on top of the list (77%). Enterococcus faecalis, Viridans streptococci, and S aureus were encountered less frequently. S epidermidis, in recent times, has emerged as an important pathogen in both pediatric and surgical patients [23,24]. Formerly considered as a contaminant and probably of low pathogenecity in healthy patients, S epidermidis is now a recognized pathogen in premature infants [25] including infants with NEC [26]. The slime layer secreted by this organism helps in adherence to prosthesis-like indwelling venous catheters, inhibiting penetration of antibiotics, thus allowing the microorganism to persist. Enterococci are infrequent blood stream pathogens [27] but were reported more frequently than S aureus in our study. The antimicrobial susceptibility pattern of gram-positive isolates showed uniform susceptibility to glycopeptides and linezolid. Cloxacillin resistance was seen in 60% of staphylococci in general and 50% of S aureus in particular. This high prevalence of cloxacillin resistance in the staphylococci is probably a reflection of the high level of circulating methicillinresistant S aureus and methicillin-resistant S epidermidis in our hospital. (E.M. Mokaddas, personal observation). Gram-negative bacterial isolates recovered from blood cultures were mainly P aeruginosa, K pneumoniae, and Acinetobacter spp. E coli and Enterobacter spp were not as common. In a similar prospective study in children by Levy et al [28], the predominant isolates causing gram-negative bacteremia were K pneumoniae, P aeruginosa, and E coli. Our study also compares well with a study carried out by Tallur et al [29], where P aeruginosa and K pneumoniae were the most common isolates in pediatric septicemia. Many other studies of Enterobacter and Klebsiella bacter-
683 emia in children showed that gastrointestinal and genitourinary lesions were the main underlying diseases [30-32]. In our study, antimicrobial susceptibility pattern of gramnegative organisms showed a high degree of resistance to third-generation cephalosporins as well as piperacillin/ tazobactam; the affected bacteria were mainly Acinetobacter spp and P aeruginosa. On the other hand, the gramnegative organisms were fairly susceptible to the aminoglycosides, quinolones, and carbapenems. This observation is in accord with a similar study in Israel some time ago [28]. According to the antibiotic guidelines followed in our department for all neonatal septicemic cases, empirical treatment with amikacin and piperacillin/tazobactam is the usual choice. However, our isolates of Acinetobacter spp and P aeruginosa were highly resistant to piperacillin/tazobactam. The use of carbapenems in infections caused by these resistant bacteria would be justified. Extended-spectrum β-lactamases were detected in 5 (33.3%) of 15 Enterobacteriaceae, 4 of which belong to Klebsiella spp and 1 to Enterobacter spp. This is in sharp contrast to a study carried out in 2003 by Jain et al [33] in which ESBLs were detected in 86.6% of Klebsiella spp, 73.4% of Enterobacter spp, and 63.6% of E coli. In these situations, the use of carbapenems as the drug of first choice is justified. The predominant cause of candidemia in our study was Candida albicans. This is comparable to a former 5-year study by Mokaddas et al [34] in pediatric surgical patients in Kuwait in which C albicans was isolated from 56% of the cases, whereas the remaining 44% yielded other Candida spp dominated by C parapsilosis. Several risk factors have been shown to predispose neonates and young infants to invasive candidiasis [35]. Such risk factors include major congenital malformations and damaged gastrointestinal mucosa because of NEC or after bowel surgery [36-38]. The importance of every single factor is difficult to evaluate because, in most cases, they may be all present in the patient at the same time, as demonstrated in the current study. Septicemia in children has a high mortality rate varying between 30% to 70% and depends on several factors including virulence of the pathogen and host factors [39,40]. In our study, the mortality rate because of septicemic episodes was 19%, in which gram-negative septicemia and candidemia accounted for the majority. Death because of gram-positive organisms was owing to staphylococcal septicemia in only 5% of the cases. All mortality cases followed gastrointestinal surgery, and in 13 of 15 mortalities, septicemia was acquired in the hospital. The mortality rate in our study is similar to a study done by Betty and Inderpreet [3], where in the mortality rate was about 19.4%. In conclusion, documentation of risk factors is the first step in establishing strategies to lower infection rates in pediatric surgery patients. Instead of all the money and effort spent in managing septicemia in such patients with expensive broad-spectrum antibiotics, real efforts should be devoted to early and accurate diagnosis of sepsis. In this context, the use
684 of newer markers of sepsis should be encouraged for both early diagnosis as well as follow up of therapy.
References [1] Yogaraj JS, Elward AM, Fraser VJ. Rate, risk factors, and outcomes of nosocomial primary bloodstream infection in pediatric intensive care unit patients. Pediatrics 2002;110:481-5. [2] Brun-Buisson C, Doyon F, Carlet J. Bacteremia and severe sepsis in adults: a multicenter prospective survey in ICUs and wards of 24 hospitals. French Bacteremia-Sepsis Study Group. Am J Respir Crit Care Med 1996;154:617-24. [3] Betty C, Inderpreet S. Early onset neonatal sepsis. Indian J Pediatr 2005;72:23-6. [4] Kaufman D, Fairchild KD. Clinical microbiology of bacterial and fungal sepsis in very low birth weight infants. Clin Microbiol Rev 2004;17:638-80. [5] Tokars JI, Cookson ST, McArthur MA, et al. Prospective evaluation of risk factors for bloodstream infection in patients receiving home infusion therapy. Ann Intern Med 1999;131:340-7. [6] Weinstein MP, Merrett S, Reimer LG, et al. Controlled evaluation of BacT/Alert standard aerobic and FAN aerobic blood culture bottles for detection of bacteremia and fungemia. J Clin Microbiol 1995;33: 978-81. [7] Khoory BJ, Vino L, Dall'Agnola A, et al. Candida infections in newborns: a review. J Chemother 1999;11:367-78. [8] Yadav AK, Wilson CG, Prasad PL, et al. Polymerase chain reaction in rapid diagnosis of neonatal sepsis. Indian Pediatr 2005;42:681-5. [9] Leibovici L, Shraga I, Drucker M, et al. Benefit of appropriate empirical antibiotic treatment. J Intern Med 1998;244:379-86. [10] Phillips I. Resistance as a cause of treatment failure. J Antimicrob Chemother 1986;18(SupplC):255-60. [11] Patterson JE. Extended-spectrum beta-lactamases. Semin Respir Crit Care Med 2003;24:79-88. [12] Jamal W, Rotimi VO, Khodakhast F. Prevalence of extended spectrum beta-lactamases in Enterobacteriaceae, Pseudomonas and Stenotrophomonas as determined by the VITEK 2 and E test systems in a Kuwait teaching hospital. Med Princ Pract 2005;14:325-31. [13] Bhattacharyya N, Kosloske AM, Macarthur C. Nosocomial infection in pediatric surgical patients: a study of 608 infants and children. J Pediatr Surg 1993;28:338-44. [14] Chiswick ML. Infection and defense in neonates. BMJ 1983;286: 1377-8. [15] Christensen RD, Anstall HB, Rothstein G. Deficiencies in the neutrophil system of newborn infants and the use of leukocyte transfusion in the treatment of neonatal sepsis. J Clin Aphersis 1982;1: 3-41. [16] Chandra RK. Serum thymic hormone activity and cell-mediated immunity in healthy neonates, preterm infants, and small for gestational age infants. Pediatrics 1981;67:407-11. [17] Stoll BJ, Hansen N, Fanaroff AA, et al. Changes in pathogens causing early-onset sepsis in very low–birth weight infants. N Eng J Med 2002;347:240-7. [18] Lopez Sastre JB, Coto Cotallo GD, Fernandez CB, et al. Neonatal sepsis of vertical transmission: an epidemiological study from the “Grupo de Hospitales Castrillo.” J Perinat Med 2000;28:309-15.
E.M. Mokaddas et al. [19] Stoll BJ, Gordon T, Korones SB, et al. Early-onset sepsis in very low birth weight neonates: a report from the National Institute of Child Health and Human Development Neonatal Research Network. J Pediatr 1996;129:72-80. [20] Raad II, Hohn DC, Gilbreath BJ, et al. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. Infect Control Hosp Epidemiol 1994;15:231-8. [21] Singh-Naz N, Sprague BM, Patel KM, et al. Risk factors for nosocomial infection in critically ill children: a prospective cohort study. Crit Care Med 1996;24:875-8. [22] Ford-Jones EL, Mindorff CM, Langley JM, et al. Epidemiologic study of 4684 hospital-acquired infections in pediatric patients. Pediatr Infect Dis J 1989;8:668-75. [23] Scherer LR, West KW, Weber TR, et al. Staphylococcus epidermidis sepsis in pediatric patients: clinical and therapeutic considerations. J Pediatr Surg 1984;19:358-61. [24] Freeman J, Platt R, Sidebottom DG, et al. Coagulase-negative staphylococcal bacteremia in the changing neonatal intensive care unit population. Is there an epidemic? JAMA 1987;258:2548-52. [25] Patrick CC, Kaplan SL, Baker CJ, et al. Persistent bacteremia due to coagulase-negative staphylococci in low birth weight neonates. Pediatrics 1989;84:977-85. [26] Mollitt DA, Tepas JJ, Talbert JL. The role of coagulase-negative Staphylococcus in neonatal necrotizing enterocolitis. J Pediatr Surg 1988;23:60-3. [27] Toltzis P, Blumer JL. Problems with resistance in pediatric intensive care units. New Horizons 1996;4:353-60. [28] Levy I, Leibovici L, Drucker M, et al. A prospective study of gramnegative bacteremia in children. Pediatr Infect Dis J 1996;15:117-22. [29] Tallur SS, Kasturi AV, Nadgir SD, et al. Clinico-bacteriological study of neonatal septicemia in Hubli. Indian J Pediatr 2000;67:169-74. [30] Gallagher PG. Enterobacter bacteremia in pediatric patients. Rev Infect Dis 1990;12:808-12. [31] Andersen J, Asmar B, Dajani A. Increasing Enterobacter bacteremia in pediatric patients. Pediatr Infect Dis J 1994;13:787-92. [32] Bonadio WA. Polymicrobial bacteremia in children. Am J Dis Child 1988;142:1158-60. [33] Jain A, Roy I, Gupta MK, et al. Prevalence of extended-spectrum β-lactamase-producing Gram-negative bacteria in septicemic neonates in a tertiary care hospital. J Med Microbiol 2003;52:421-5. [34] Mokaddas EM, Ramadan SA, Abo el Maaty SH, et al. Candidemia in pediatric surgery patients. J Chemother 2000;12:332-8. [35] Rabalais GP, Samiec TD, Bryant KK, et al. Invasive candidiasis in infants weighing more than 2500 grams at birth admitted to a neonatal intensive care unit. Pediatr Infect Dis J 1996;15:348-52. [36] Weese-Mayer DE, Fondriest DW, Brouillette RT, et al. Risk factors associated with Candidemia in the neonatal intensive care unit: a casecontrol study. Pediatr Infect Dis J 1987;6:190-6. [37] Johnson DE, Thompson TR, Green TP, et al. Systemic Candidiasis in very low–birth-weight infants (b1500 grams). Pediatrics 1984;73: 138-43. [38] Smith SD, Tagge EP, Miller J, et al. The hidden mortality in surgically treated necrotizing enterocolitis: fungal sepsis. J Pediatr Surg 1990;10: 1030-3. [39] Wenzel RP, Pinsky MR, Ulevitch RJ, et al. Current understanding of sepsis. Clin Infect Dis 1996;22:407-12. [40] Wheeler AP, Bernard GR. Treating patients with severe sepsis. N Engl J Med 1999;340:207-14.
www.elsevier.com/locate/jpedsurg
A 4-year prospective study of septicemia in pediatric surgical patients at a tertiary care teaching hospital in Kuwait Eiman M. Mokaddas a,b,⁎, Shama A. Shetty b , Aneesa A. Abdullah b , V.O. Rotimi a a
Department of Microbiology, Faculty of Medicine, Kuwait University, PO Box 24923, Safat 13110, Kuwait Department of Laboratory Medicine, Ibn Sina hospital, Kuwait
b
Received 29 July 2010; revised 28 November 2010; accepted 1 December 2010
Key words: Microbiology; Septicemia; Pediatric surgery; Tertiary care hospital; Kuwait
Abstract Background: Critically ill children are at high risk for developing nosocomial infections that contributes to death in 4% of all pediatric intensive care unit admissions. This prospective study was undertaken to determine the prevalence of septicemia in the pediatric surgery department of a large tertiary care teaching hospital in Kuwait and to evaluate the risk factors, the microbial etiology, and the antimicrobial susceptibility pattern of the isolated microorganisms. Methods: All patients admitted to the pediatric surgery department from January 2001 until December 2004 with the diagnosis of septicemia were included in the study, and the microbiologically proven cases were then analyzed. The patients' demographics and risk factors for sepsis were recorded. All positive blood cultures were subjected to identification and antimicrobial susceptibility testing by VITEK 2 (bioMerieux, Marcy l'Etoile, France). Results: Of 3408 patients suspected to have septicemia, 78 (2.3%) patients developed microbiologically documented septicemias, 26% of those were low–birth weight patients, and 82% were patients with congenital anomalies; 87% of those needed surgical intervention. More than 50% were admitted to the intensive care unit, and 80.5% needed ventilatory support. Fifty-seven percent had early onset septicemia. Gram-positive and gram-negative bacteria accounted for 54% and 39% of the septicemia cases, respectively, whereas Candida spp was responsible for 7%. More than 50% of the staphylococci were resistant to cloxacillin, and all gram-positives were uniformly susceptible to glycopeptides and linezolid. Gram-negative bacteria showed variable resistance to cephalosporins (65%), piperacillin/tazobactam (29%), and carbapenems (11%). The attributable mortality rate for these septic episodes was 19% mainly because of gram-negative bacteria and Candida. Conclusion: The main etiologic agents of neonatal septicemia were coagulase-negative Staphylococcus, Pseudomonas aeruginosa, and members of the family Enterobacteriaceae. Empirical therapy with piperacillin/tazobactam or carbapenems for gram-negative septicemia and glycopeptides for gram-positive septicemia was effective. © 2011 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Department of Microbiology, Faculty of Medicine, Kuwait University, PO Box 24923, Safat 13110, Kuwait. Tel.: +965 498 6512. E-mail address: [email protected] (E.M. Mokaddas). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2010.12.001
Bloodstream infections represent a major cause of hospital-acquired infections in neonatal and pediatric intensive care units (ICUs). According to the National
680 Nosocomial Infection Surveillance Report published in 1999, bloodstream infections accounted for 28% of the total infections, followed by ventilator-associated pneumonia, 21% [1]. They are more likely to occur in the ICU than in non-ICU patients [2]. Neonatal sepsis is one of the major causes of morbidity and mortality in the newborn in general, particularly in the premature neonates [3]. Immature host defense mechanisms, underlying diseases, low birth weight, and immaturity together with invasive life support systems make the premature neonates particularly vulnerable to overwhelming infections [4]. Other risk factors include maternal colonization with Streptococcus agalactiae during labor and premature rupture of membranes [5].Various bacterial pathogens can cause septicemia in this age group. Grampositive bacteria, mainly staphylococci and streptococci, together with gram-negative bacteria, namely, the Enterobacteriaceae and Pseudomonas aeruginosa, are the major groups of pathogens [6]. Fungi, in particular, Candida spp, are reported to cause septicemia in around 1% of infants admitted to the neonatal ICU and 2% to 4.5% of low–birth weight neonates with an overall mortality of 50% [7]. Isolation of an organism from the blood culture of a neonate with clinical symptoms of infection constitutes the criterion standard for diagnosing neonatal sepsis [8]. In such a lifethreatening condition, knowledge of the likely causative organism and appropriate empirical antibiotic treatment is associated with a better chance for survival, regardless of other risk factors for fatality [9]. Early treatment with appropriate antibiotics would minimize both the risk of mortality and the emergence of multidrug resistant organisms in the ICU [3,10]. However, one of the confounding problems associated with empirical treatment in the ICU is the emergence of antibiotic resistance. Extended-spectrum β-lactamases (ESBLs), most commonly found in Klebsiella pneumoniae and Escherichia coli, have increased markedly in the past decade, particularly in the ICU setting [11]. Unfortunately, the detection of ESBL in any microbiology laboratory is only about 40% to 50% of the ESBL-producing gram-negative bacteria. Because these strains may be fairly susceptible to the third-generation cephalosporins in vitro, clinical failure may be as high as 75% with such antibiotics [11]. This study was undertaken to determine the prevalence of septicemia and evaluate the demographic data, underlying risk factors, and the microbial etiology in the pediatric surgery department of a large tertiary-care teaching hospital in Kuwait.
E.M. Mokaddas et al. pediatric surgery patients. In this institute, there is no medical pediatric service, and all the patients were referred from other general hospitals for surgical opinion and interventions. All patients with provisional diagnosis of septicemia during a period of 4 years spanning January 2001 to December 2004 were included in the study. The patients were admitted either to the ICU or the general ward. Patients suspected of having sepsis were identified based on the presence of one or more clinical signs, that are, lethargy, refusal of feeds, irritability, vomiting, respiratory distress, fever more than 38°C or hypothermia less than 36°C, leucocytosis (total peripheral leukocyte count of more than 15,000 cells/L), rise in C-reative protein (CRP), and then were subjected to septic workup. Two sets of blood (2 mL) sample were collected from each patient and inoculated into pediatric bottles and incubated in the BACTEC 9240 (Becton Dickinson, Sparks, MD) automated blood culture machine. Incubation was continued until the bottle signaled or up to a maximum of 7 days when the culture was declared negative. For each microbiologically proven septicemia case, the details of patient's demographics are recorded on a protocol data form. These included sex, age, gestational age at birth, and birth weight. All risk factors for developing sepsis such as underlying medical conditions, surgical procedures, intravascular catheterization, endotracheal intubations, and mechanical ventilation as well as the length of hospital stay were also documented. All positive cultures were subcultured onto appropriate media, and gram-stained smears were prepared. Identification and antimicrobial susceptibility testing of all isolates were done using VITEK 2 automated systems (bioMerieux, Marcy l'Etoile, France). Supplementary antimicrobial susceptibility tests were performed by the disc diffusion technique on Mueller Hinton agar (Oxoid, Basingstoke, UK). Any discrepancy in the above 2 methods were resolved by the E-test (AB Biodisk, Solana, Sweden). Extended-spectrum β-lactamase–producing gramnegative bacteria were detected by the ESBL E-test method that used both ceftazidime-ceftazidime/clavulanate and cefotaxime-cefotaxime/clavulanate strips [12]. After blood culture collection, all patients who developed clinical signs of septicemia were started on empirical antimicrobial therapy according to the antibiotic policy and guidelines of the pediatric surgery department, pending results of microbiology investigations.
2. Results 1. Materials and methods 1.1. Patients This study was performed at a tertiary care center in Kuwait, which is the only health care facility catering to
A total of 8303 patients were admitted to the pediatric and neonatal ward during the study period. The average admissions per year in the ICU and general wards for the 4-year period were 135.5 and 2365.3 patients, respectively. Of the 3408 patients suspected to have septicemia, according to the previously mentioned criteria, 78 (2.3%)
Prospective study of septicemia in pediatric surgical patients Table 1
The demographic profiles of patients with septicemia
Demographics Sex Male Female Age group b1 mo 1-6 mo 6-12 mo N1 year Gestational age Preterm Full term Birth weight b2 kg 2-3 kg N3 kg
No. of patients with sepsis (%) 54 (69) 24 (31) 31 (40) 28 (36) 10 (13) 9 (11) 29 (37) 49 (63) 20 (26) 37 (47) 21 (27)
developed 89 microbiologically documented septic episodes. Various patient's demographics are demonstrated in Table 1. The underlying diseases of these patients were analyzed and showed that 64 (82%) were born with surgical congenital anomalies, as shown in Table 2. Of the 78 patients ,68 developed sepsis after surgical intervention, Table 2 Distribution of different underlying clinical conditions of the septic patients Clinical condition
No. of patients
Necrotizing enterocolitis Oesophageal atresia Intestinal atresia Volvulus + malrotation Congenital diaphragmatic hernia Hirschprung disease Hydronephrosis Vesico-uretral reflux Anorectal malformation Exomphalos Gastroschisis Idiopathic hypertropic pyloric stenosis Undescended testis Inguinal hernia Appendicitis Vaginal atresia Tumor Cellulitis Osteomyelitis Choanal atresia Duodenal atresia Intusseception Cloacal extrophy Gastroenteritis Foreign body gut Constipation Total
16 9 7 6 3 3 3 2 2 2 1 2 2 1 2 2 2 3 2 1 1 1 1 1 1 2 78
681 Table 3 The risk factors associated with septic episodes in the 78 patients with septicemia Risk factors
No. of patients (%)
ICU stay Surgical intervention Ventillator support Lines Central line Arterial line Tubes and drains Nasogastric tube Endotracheal tube Peritonial drain Urethral catheter
41 (52) 68 (87) 33 (42) 24 (31) 5 (6) 61 32 6 60
(78) (41) (8) (77)
and only 10 developed septicemia without any surgical intervention but had some other underlying risk factors, as shown in Table 3. Furthermore, of the 78 septicemic cases, 41 (52%) were admitted to the ICU, and 33 (42%) of those needed ventillatory support. For those admitted to the ICU, the average hospital stay was 15.6 days, with an average of 9.8 days on ventillatory support. The prevalence of septicemia among the ICU patients was 6.3%, whereas on the ward, it was 0.5%. The correlation between the onset of septicemia in relation to the timing of hospital admission is shown in Table 4. Of the 78 patients, 45 (57.7%) acquired early onset septicemia within the first 7 days of admission. Of these 45, 35 (77.8%) acquired the infection within 48 hours. Thirtyfive percent and 75% of patients in the ICU and ward, respectively, developed early onset episodes compared to 64.7% of ICU patients and 25% of ward patients who had late onset septicemia. The distribution of different etiologic microbial agents of the septic episodes is shown in Table 5. Gram-positive bacteria predominated, accounting for 48 (54%) of 89 isolates. Gram-negative bacteria, on the other hand, accounted for 35 of the isolates (39%). Candida spp accounted for the rest (7%). Of the 89 episodes, 4 had polymicrobial etiology. The antimicrobial susceptibility of the gram-positive isolates showed uniform susceptibility to glycopeptides and linezolid, whereas 60% of all staphylococci were resistant to cloxacillin. A very high proportion (50%) of the
Table 4 Correlation between onset of septic episodes and timing of hospital admission Onset of septic episode
Within 48 h of admission 3-7 days of admission N7 days after admission Total
No. of patients ICU
Ward
Total no. (%)
9 3 22 34
26 7 11 44
35 (45) 10 (13) 33 (42) 78 (100)
682
E.M. Mokaddas et al.
Table 5 Distribution of etiologic microbial agents in various septicaemic episodes Microorganisms
No. of isolates (%)
Gram-positive Coagulase-negative staphylococci S aureus E faecalis V streptococci Gram-negative E coli K pneumoniae Citrobacter spp Enterobacter spp P aeruginosa Acinetobacter spp Stenotrophomonas maltophilia Sphingomonas paucimobilis Fungi C albicans C parapsilosis
48 37 2 5 4 35 4 9 1 1 10 7 1 2 6 4 2
(54) (77) (4) (10) (8) (39) (11) (28) (3) (3) (28) (19) (3) (6) (7) (67) (33)
Staphylococcus aureus isolates were resistant to cloxacillin (methicillin-resistant Staphylococcus aureus [MRSA]). Gram-negatives, on the other hand, showed variable susceptibilities to third generation cephalosporins, especially P aeruginosa and Acinetobacter spp, with an average resistance of 35%. An average resistance to aminoglycosides was 29% and 14% each to gentamicin and amikacin, respectively. Piperacillin-tazobactam and carbapenems showed an average resistance of 29% and 11%, respectively. Extended-spectrum β-lactamases were detected in 5 of the total 15 Enterobacteriaceae. Of the 78 patients who developed septicemia, 15 (19%) died. Thirteen of these deaths occurred after single or multiple surgical interventions, and 2 patients died before any surgery. The birth weight in these mortality cases ranged from 1.09 to 3.01 kg. Ten of these infants were preterm. Seven (47%) were necrotizing enterocolitis (NEC) with perforation, 3 (20%) were cases of intestinal atresia, 2 (13%) were gangrenous volvulus, and 1 case each of Hirchprung disease, anorectal malformation, and gastroschisis. The 13 mortality cases, all except the patient with gastroschisis, were reoperated .Various surgical interventions were carried out such as laparotomy, ileostomy, and resection of the gangrenous part of intestine. All of the 13 patients who underwent surgical intervention developed sepsis and died between the first and third week postoperatively. Nine (60%) of those cases had gram-negative septicemia owing to Acinetobacter spp (3), P aeruginosa (3), K pneumoniae (2), and Citrobacter spp (1). Four gram-positive septicemias (27%) were caused by coagulase-negative Staphylococcus (2) and one each of Enterococcus and Viridans streptococcus. Candida albicans accounted for 2 cases (13%).
Parenteral amikacin and piperacillin/tazobactam were empirically started, which had to be replaced with meropenem either because the organism was an ESBL-producing Enterobacteriaceae or because of multidrug-resistant Acinetobacter or P aeruginosa. Candidemia was treated with parenteral fluconazole. In 11 patients (73%), death was because of multiorgan failure and disseminated intravascular coagulation, and the other 2 patients (13%) went into renal failure after septicemia.
3. Discussion The data generated in this study highlighted the prevalence of pediatric septicemia among general admissions in a tertiary care hospital in Kuwait, the major predisposing factors, admission rates to the ICU and wards in relation to onset of infections, and the type of causative agents and their susceptibility pattern. In our study, the prevalence of septicemia was very low at about 1%. Different records on the prevalence of septicemia in neonates in general and pediatric surgical patients in particular have been reported. In one retrospective study by Bhattacharya et al [13] on a group of 608 patients from the pediatric surgical department for a 14-month period, 14 patients (2.3%) developed septicemia. Neonates in this study were at an increased risk for septicemia (4.2%) compared to older children and infants (3.1%) [13], which could be attributed to impaired host defenses of neonates [14-16]. In general, the data presented in this report were comparable to other reports [13-16]. In our study, neonates and infants 6 months or younger accounted for most of the 40% and 36% of patients, respectively. The main risk factors of septicemia were low birth weight and preterm deliveries. These figures of 26% and 57%, respectively, in our study contrast the study done in Norway where early onset sepsis among very low–birth weight infants are infrequent, and the incidence ranged between 1.5% and 2.7% [17,18]. However, in another study from India, both very low birth weight and prematurity were very important predisposing factors for the development of septicemia [6]. According to a report by the National Institute of Child Health and Human Development in 1996, culture-proven septicemia was reported in only 1.9% of very low–birth weight infants [19,20]. Thus, in our center, both low birth weight and prematurity are significant contributory risk factors for developing septicemia. Other very important factors are underlying medical and congenital problems, insertion of central venous catheters [20], and especially those who needed surgical interventions. Intensive care unit stay in our center predisposed to late onset septicemia while ward admission were the opposite with predominantly early-onset type. This may be partly due to the fact that ICU patients tended to acquire nosocomial infections more than ward patients who may have carried
Prospective study of septicemia in pediatric surgical patients community-acquired infections into the ward. The observation of health care–related septicemia in the ICU is also confirmed by the report of Singh-Naz et al [21] that indicated that pediatric ICU patients who undergo surgical operations are 2.5 times more likely to develop health care–related infections than nonoperated patients. Instrumentation and other manipulations such as insertion of central venous lines and use of endotracheal tubes and urinary catheters leading to device-associated infections [22] are known to be associated with lowering of host defense system. This is supported by the high percentage of patients with these devices in our study. The prevalence of 6.3% of septicemia among the ICU patients is similar to the report of a 9-month study done at St Louis Childrens' Hospital, Missouri, in which 57 (6.25%) of 911 patients admitted to the pediatric ICU developed septicemia. In this study, 50% had arterial catheters, 12% had multiple central venous catheters, and 70% were mechanically ventilated [1], most of which are discordant with our present findings where the major risk factors were surgery 67 (87%), insertion of nasogastric tubes 61 (81%), and urethral catheters 60 (77%). The leading cause of microbiologically proven septicemia in our patients was gram-positive cocci in which coagulase-negative staphylococci, predominantly S epidermidis, were on top of the list (77%). Enterococcus faecalis, Viridans streptococci, and S aureus were encountered less frequently. S epidermidis, in recent times, has emerged as an important pathogen in both pediatric and surgical patients [23,24]. Formerly considered as a contaminant and probably of low pathogenecity in healthy patients, S epidermidis is now a recognized pathogen in premature infants [25] including infants with NEC [26]. The slime layer secreted by this organism helps in adherence to prosthesis-like indwelling venous catheters, inhibiting penetration of antibiotics, thus allowing the microorganism to persist. Enterococci are infrequent blood stream pathogens [27] but were reported more frequently than S aureus in our study. The antimicrobial susceptibility pattern of gram-positive isolates showed uniform susceptibility to glycopeptides and linezolid. Cloxacillin resistance was seen in 60% of staphylococci in general and 50% of S aureus in particular. This high prevalence of cloxacillin resistance in the staphylococci is probably a reflection of the high level of circulating methicillinresistant S aureus and methicillin-resistant S epidermidis in our hospital. (E.M. Mokaddas, personal observation). Gram-negative bacterial isolates recovered from blood cultures were mainly P aeruginosa, K pneumoniae, and Acinetobacter spp. E coli and Enterobacter spp were not as common. In a similar prospective study in children by Levy et al [28], the predominant isolates causing gram-negative bacteremia were K pneumoniae, P aeruginosa, and E coli. Our study also compares well with a study carried out by Tallur et al [29], where P aeruginosa and K pneumoniae were the most common isolates in pediatric septicemia. Many other studies of Enterobacter and Klebsiella bacter-
683 emia in children showed that gastrointestinal and genitourinary lesions were the main underlying diseases [30-32]. In our study, antimicrobial susceptibility pattern of gramnegative organisms showed a high degree of resistance to third-generation cephalosporins as well as piperacillin/ tazobactam; the affected bacteria were mainly Acinetobacter spp and P aeruginosa. On the other hand, the gramnegative organisms were fairly susceptible to the aminoglycosides, quinolones, and carbapenems. This observation is in accord with a similar study in Israel some time ago [28]. According to the antibiotic guidelines followed in our department for all neonatal septicemic cases, empirical treatment with amikacin and piperacillin/tazobactam is the usual choice. However, our isolates of Acinetobacter spp and P aeruginosa were highly resistant to piperacillin/tazobactam. The use of carbapenems in infections caused by these resistant bacteria would be justified. Extended-spectrum β-lactamases were detected in 5 (33.3%) of 15 Enterobacteriaceae, 4 of which belong to Klebsiella spp and 1 to Enterobacter spp. This is in sharp contrast to a study carried out in 2003 by Jain et al [33] in which ESBLs were detected in 86.6% of Klebsiella spp, 73.4% of Enterobacter spp, and 63.6% of E coli. In these situations, the use of carbapenems as the drug of first choice is justified. The predominant cause of candidemia in our study was Candida albicans. This is comparable to a former 5-year study by Mokaddas et al [34] in pediatric surgical patients in Kuwait in which C albicans was isolated from 56% of the cases, whereas the remaining 44% yielded other Candida spp dominated by C parapsilosis. Several risk factors have been shown to predispose neonates and young infants to invasive candidiasis [35]. Such risk factors include major congenital malformations and damaged gastrointestinal mucosa because of NEC or after bowel surgery [36-38]. The importance of every single factor is difficult to evaluate because, in most cases, they may be all present in the patient at the same time, as demonstrated in the current study. Septicemia in children has a high mortality rate varying between 30% to 70% and depends on several factors including virulence of the pathogen and host factors [39,40]. In our study, the mortality rate because of septicemic episodes was 19%, in which gram-negative septicemia and candidemia accounted for the majority. Death because of gram-positive organisms was owing to staphylococcal septicemia in only 5% of the cases. All mortality cases followed gastrointestinal surgery, and in 13 of 15 mortalities, septicemia was acquired in the hospital. The mortality rate in our study is similar to a study done by Betty and Inderpreet [3], where in the mortality rate was about 19.4%. In conclusion, documentation of risk factors is the first step in establishing strategies to lower infection rates in pediatric surgery patients. Instead of all the money and effort spent in managing septicemia in such patients with expensive broad-spectrum antibiotics, real efforts should be devoted to early and accurate diagnosis of sepsis. In this context, the use
684 of newer markers of sepsis should be encouraged for both early diagnosis as well as follow up of therapy.
References [1] Yogaraj JS, Elward AM, Fraser VJ. Rate, risk factors, and outcomes of nosocomial primary bloodstream infection in pediatric intensive care unit patients. Pediatrics 2002;110:481-5. [2] Brun-Buisson C, Doyon F, Carlet J. Bacteremia and severe sepsis in adults: a multicenter prospective survey in ICUs and wards of 24 hospitals. French Bacteremia-Sepsis Study Group. Am J Respir Crit Care Med 1996;154:617-24. [3] Betty C, Inderpreet S. Early onset neonatal sepsis. Indian J Pediatr 2005;72:23-6. [4] Kaufman D, Fairchild KD. Clinical microbiology of bacterial and fungal sepsis in very low birth weight infants. Clin Microbiol Rev 2004;17:638-80. [5] Tokars JI, Cookson ST, McArthur MA, et al. Prospective evaluation of risk factors for bloodstream infection in patients receiving home infusion therapy. Ann Intern Med 1999;131:340-7. [6] Weinstein MP, Merrett S, Reimer LG, et al. Controlled evaluation of BacT/Alert standard aerobic and FAN aerobic blood culture bottles for detection of bacteremia and fungemia. J Clin Microbiol 1995;33: 978-81. [7] Khoory BJ, Vino L, Dall'Agnola A, et al. Candida infections in newborns: a review. J Chemother 1999;11:367-78. [8] Yadav AK, Wilson CG, Prasad PL, et al. Polymerase chain reaction in rapid diagnosis of neonatal sepsis. Indian Pediatr 2005;42:681-5. [9] Leibovici L, Shraga I, Drucker M, et al. Benefit of appropriate empirical antibiotic treatment. J Intern Med 1998;244:379-86. [10] Phillips I. Resistance as a cause of treatment failure. J Antimicrob Chemother 1986;18(SupplC):255-60. [11] Patterson JE. Extended-spectrum beta-lactamases. Semin Respir Crit Care Med 2003;24:79-88. [12] Jamal W, Rotimi VO, Khodakhast F. Prevalence of extended spectrum beta-lactamases in Enterobacteriaceae, Pseudomonas and Stenotrophomonas as determined by the VITEK 2 and E test systems in a Kuwait teaching hospital. Med Princ Pract 2005;14:325-31. [13] Bhattacharyya N, Kosloske AM, Macarthur C. Nosocomial infection in pediatric surgical patients: a study of 608 infants and children. J Pediatr Surg 1993;28:338-44. [14] Chiswick ML. Infection and defense in neonates. BMJ 1983;286: 1377-8. [15] Christensen RD, Anstall HB, Rothstein G. Deficiencies in the neutrophil system of newborn infants and the use of leukocyte transfusion in the treatment of neonatal sepsis. J Clin Aphersis 1982;1: 3-41. [16] Chandra RK. Serum thymic hormone activity and cell-mediated immunity in healthy neonates, preterm infants, and small for gestational age infants. Pediatrics 1981;67:407-11. [17] Stoll BJ, Hansen N, Fanaroff AA, et al. Changes in pathogens causing early-onset sepsis in very low–birth weight infants. N Eng J Med 2002;347:240-7. [18] Lopez Sastre JB, Coto Cotallo GD, Fernandez CB, et al. Neonatal sepsis of vertical transmission: an epidemiological study from the “Grupo de Hospitales Castrillo.” J Perinat Med 2000;28:309-15.
E.M. Mokaddas et al. [19] Stoll BJ, Gordon T, Korones SB, et al. Early-onset sepsis in very low birth weight neonates: a report from the National Institute of Child Health and Human Development Neonatal Research Network. J Pediatr 1996;129:72-80. [20] Raad II, Hohn DC, Gilbreath BJ, et al. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. Infect Control Hosp Epidemiol 1994;15:231-8. [21] Singh-Naz N, Sprague BM, Patel KM, et al. Risk factors for nosocomial infection in critically ill children: a prospective cohort study. Crit Care Med 1996;24:875-8. [22] Ford-Jones EL, Mindorff CM, Langley JM, et al. Epidemiologic study of 4684 hospital-acquired infections in pediatric patients. Pediatr Infect Dis J 1989;8:668-75. [23] Scherer LR, West KW, Weber TR, et al. Staphylococcus epidermidis sepsis in pediatric patients: clinical and therapeutic considerations. J Pediatr Surg 1984;19:358-61. [24] Freeman J, Platt R, Sidebottom DG, et al. Coagulase-negative staphylococcal bacteremia in the changing neonatal intensive care unit population. Is there an epidemic? JAMA 1987;258:2548-52. [25] Patrick CC, Kaplan SL, Baker CJ, et al. Persistent bacteremia due to coagulase-negative staphylococci in low birth weight neonates. Pediatrics 1989;84:977-85. [26] Mollitt DA, Tepas JJ, Talbert JL. The role of coagulase-negative Staphylococcus in neonatal necrotizing enterocolitis. J Pediatr Surg 1988;23:60-3. [27] Toltzis P, Blumer JL. Problems with resistance in pediatric intensive care units. New Horizons 1996;4:353-60. [28] Levy I, Leibovici L, Drucker M, et al. A prospective study of gramnegative bacteremia in children. Pediatr Infect Dis J 1996;15:117-22. [29] Tallur SS, Kasturi AV, Nadgir SD, et al. Clinico-bacteriological study of neonatal septicemia in Hubli. Indian J Pediatr 2000;67:169-74. [30] Gallagher PG. Enterobacter bacteremia in pediatric patients. Rev Infect Dis 1990;12:808-12. [31] Andersen J, Asmar B, Dajani A. Increasing Enterobacter bacteremia in pediatric patients. Pediatr Infect Dis J 1994;13:787-92. [32] Bonadio WA. Polymicrobial bacteremia in children. Am J Dis Child 1988;142:1158-60. [33] Jain A, Roy I, Gupta MK, et al. Prevalence of extended-spectrum β-lactamase-producing Gram-negative bacteria in septicemic neonates in a tertiary care hospital. J Med Microbiol 2003;52:421-5. [34] Mokaddas EM, Ramadan SA, Abo el Maaty SH, et al. Candidemia in pediatric surgery patients. J Chemother 2000;12:332-8. [35] Rabalais GP, Samiec TD, Bryant KK, et al. Invasive candidiasis in infants weighing more than 2500 grams at birth admitted to a neonatal intensive care unit. Pediatr Infect Dis J 1996;15:348-52. [36] Weese-Mayer DE, Fondriest DW, Brouillette RT, et al. Risk factors associated with Candidemia in the neonatal intensive care unit: a casecontrol study. Pediatr Infect Dis J 1987;6:190-6. [37] Johnson DE, Thompson TR, Green TP, et al. Systemic Candidiasis in very low–birth-weight infants (b1500 grams). Pediatrics 1984;73: 138-43. [38] Smith SD, Tagge EP, Miller J, et al. The hidden mortality in surgically treated necrotizing enterocolitis: fungal sepsis. J Pediatr Surg 1990;10: 1030-3. [39] Wenzel RP, Pinsky MR, Ulevitch RJ, et al. Current understanding of sepsis. Clin Infect Dis 1996;22:407-12. [40] Wheeler AP, Bernard GR. Treating patients with severe sepsis. N Engl J Med 1999;340:207-14.