Contribution to mortality and resource usage of nosocomial and community infections in an intensive care setting.
Severe sepsis is a major health problem in children, with more than 42,000 cases and 4,400 associated deaths per year in the United States. (1) Intensive care units (ICU)-based studies report mortality rates of 20% to 50% in severe community-acquired infections (CI), depending on admission criteria. (2) Also, severe sepsis in ICUs is estimated to be 19.0% with hospital mortality varying between 38.1% and 50%. (4)
Evidenced-based recommendations regarding the acute management of sepsis and septic shock are the first step toward improved outcomes for this important group of critically ill patients, (3) indicating potential targets for quality improvement initiatives that could decrease mortality and morbidity.3 There are no comparative studies, however, examining the relative contribution of NI or CI to the PICUs hard endpoints of outcome (MEDLINE search). We therefore sought to: 1) explore differences and similarities related to the frequency, outcome, demographic, clinical, therapeutic, and bacteriology characteristics of critically ill children with CI, NI, and mixed (MI) infections; 2) to compare their responded contribution to outcome endpoints with that of patients without infection (WI).
MATERIALS AND METHODS
This prospective observational cohort study was conducted at the PICU of the University Hospital, Crete. Subjects included patients admitted to the PICU (6 beds) between January 2004 and December 2008, if they fulfilled the inclusion criteria of the PICU stay for at least 24-h. As the study did not affect patient care, the institutional review board waived the need for informed parental consent.
Reasons for admission included postoperative care after pediatric surgery, severe trauma, and medical conditions requiring critical care. Patients without sepsis (or any other severe infection necessitating admission to the PICU), were classified to the WI group. Patients with community only infections (diagnosed upon admission and up to 48 hours) were classified as CI, whereas those with nosocomial infections (developed > 48 hours after admission), but without CI, classified to the NI group. Patients with CI who later developed NI were classified as patients with MI.
Data collections and clinical definitions
Clinical data were collected upon admission and on each consecutive PICU day. Patients with infection were categorised to sepsis, severe sepsis, or septic shock groups based on generally accepted consensus criteria published and modified for paediatric patients by the American College of Chest Physicians and the Society of Critical Care Medicine. Chronic co morbidity was classified as 1) none, independent of care of others for activities of daily life, 2) chronic disease or handicapped partially or fully dependent on care of others, and 3) cancer patient. The severity of illness was assessed by the Pediatric Risk of Mortality (PRISM) Score, the Therapeutic Intervention Scoring System (TISS), and indices of multiple organ system failure (MOSF).
Microbiological pharmacological data and definitions
Microbiological and pharmacological data were collected over the study period in all patients. Bacteraemias identified from blood cultures collected within 48 hours of admission were deemed to be community-acquired. Nosocomial infections were required to be established after 48 h of hospitalization. Nosocomial bloodstream infections, as well as other NI (urinary tract, pneumonia, wound infection) were defined according to criteria established by the Centers for Disease Control and Prevention. (7) Patients in whom there was a suspicion of clinical or radiologic ventilator-associated pneumonia (VAP) underwent bronchoscopy, and pneumonia was confirmed by the presence of bronchoalveolar lavage fluid culture. Antibiotic prescription in PICU was imposed by definite clinical indications, guided by microbial studies as it was described in detail previously. (8)
Normally distributed data are expressed as mean[+ or -]SEM, while non-normally distributed data are given as median and range. Statistical analysis was performed with a two-tailed t-test for normally distributed paired data after Levene's correction for equality of variances or by Mann-Whitney U-Wilcoxon rank sum W test for non-normally distributed data. Probability values of <.05 with two-tailed tests were considered significant. The ANOVA with Bonferroni post hoc tests was used for multiple comparisons. When a linear regression was calculated Pearson's correlation coefficient was employed. Chi-square test was used for the category data. Logistic Regression Forward Stepwise (Wald) analysis was used to examine the independent contribution of infections to mortality, adjusting for severity of illness and various clinical and demographic characteristics. All analyses were done using the Statistical Package for the Social Sciences (SPSS) for Windows (release 17, SPSS, Chicago, IL) software package.
During the 5-year period, 510 patients (mean age 5[+ or -].2 (.6-14) years, 59.4% boys - 40.6% girls) were admitted to the PICU. Free medical history had less than 50% (42.2%) of the patients (previously healthy children), cancer 22.7%, and chronic disease 35.1%. The place of origin was 13.9% directly from the emergency department, 42.9% from the wards, 12.5% were transferred from other hospitals, 24.1% from the operative room (elective major surgeries), 1.4% from emergent operations, 3.7% were long distance air-transported, and 1.4% were chronically ill patients who were directly transferred from their homes. On the total, 269 patients (52.7%) were supported with mechanical ventilation (MV); 165 (32.4%) were surgical and 345 (67.6%) medical patients. The mortality rate was 10.2%, the mean length stay (LOS) 7.9[+ or -].9 days, length of MV (LOMV) 4.9[+ or -].8 days, PRISM 8.8[+ or -].34, and TISS 24.6[+ or -].6.
More than 50% (278/510, 54.5%) of the studied children did not have any infection (WI group), followed by those who were admitted for severe CI (157/510, 30.8%), developed NI (65/510, 12.7%) or had MI (10/510, 2%). Patients' clinical characteristics are shown in Table I. Children admitted with CI were more often healthy children, compared to patients with NI who had more often co morbidity and to the WI group who had almost equal percentages of co morbidity and cancer. Severity of illness and resource utilization were significantly higher in the NI and/or MI groups (Table I).
The three "infection groups" differed (p<.0001) regarding severe sepsis, septic shock (Table I), and the kind of infection within group and within the whole studied population (Table II). More pathogenic microorganisms were isolated in NI (32.3%: blood 9.7%, BAL 16.2%, urine 3.1%, exudates 3.3%) or MI patients (47.7%: blood 20%, BAL 80 %) than in those with community infections (9.6%: blood 1.3%, BAL 22.5%, CSF 2.7%, urine 1.9%, exudates 1.2%), (p<0.0001). The most frequently isolated microorganisms in NI were staphylococci epidermidis (26%), pseudomonas species (19%), and A. Coli (12%) followed by klebsiella (10%), candida (9%), enterobacter cloaca (8.6%), and staphylococci aureus (8%). In community infections the most frequent isolates were meningococci, (50%), followed by streptococci pneumonia (22%), staphylococci aureus (16%), and A. Coli (6%).
Compared to patients WI or with CI (Table 1), those with either NI or MI were more often using resources in PICU, p<.0001). Acute renal failure supported by CVVH did not differ between CI and NI groups (2.5% vs. 2.3%, p<.0001). While in PICU, and in contrast to most WI patients who were either not (36%) or for preventive reasons (42.4%) receiving antibiotics, CI, NI and MI patients were primarily receiving antibiotics for either clinically evident (51%, 55.4%, and 50%) or for cultures proven infection (11.5%, 12.3%, and 20%, respectively, p<.0001).
Patients in the NI group developed significantly more complications compared to CI (ARDS 8.6% vs. 3.2% and MOSF 16% vs. 7.4%, p<.0001). Among patients receiving inotropes or vasopressors, those in the WI group more frequently received 1 drug (4%), in the CI 2 drugs (8.3%), and in the NI and MI groups 4 drugs (18.5% and 20%, respectively, p<.0001). More patients in the NI and MI groups were receiving more than 3 antibiotics (13.7% and 20%) as compared to the CI (1.9%) or WI group (0%, p<.0001).
[FIGURE 1.a OMITTED]
[FIGURE 1.b OMITTED]
Fifty two patients died (10.2%). Patients with NI had higher mortality rate (16/65, 24.6% within NI group, 3.1% of the total study population) compared to patients with CI (7/157, 4.5%, 1.4%) and those without any infection (23/278, 8.3%, 4.5%). The fourth group of patients with CI who subsequently developed NI infections (MI group) exhibited the highest mortality rate (6/10, 60%, 1.2%), all deaths been a terminal event among patients with cancer (60%) or severe co morbidity (40%). Although mortality differed significantly between groups (p<.0001), each infection group's relative contribution to the total PICU mortality was lower than that of the larger group of critically ill patients (organ failures, trauma, congenital diseases, cancer) without CI or NI. In a Logistic Regression Forward Stepwise (Wald) model, only PRISM (p<.0001) and TISS (p<.001), but not infection, were independently associated with mortality.
Higher was the mortality rate in the septic shock group (26/80, 32.5%) compared to the severe sepsis group (3/117, 2.6%) and toWI group (23/278, 8.3%) or the respiratory viral infection group (0%, p<.0001). Of those who died 23 (44.2% of all deaths) had no infection (WI group), 3 (5.8%) presented with severe sepsis (1.9% NI, 3.8% MI) and 26 (50%) with septic shock (13.5% CI, 28.8% NI, 7.7% MI) (p<.0001). Those patients admitted after emergency operation had the highest mortality rate (28.6%), followed by those transferred from various departments including oncology (15.5%), other hospitals (12.5%), ED (9.9%), and elective operations (.8%), whereas all air-transported or transferred from home patients survived. Mortality was higher in patients with two (35.7%), three (3.3%) or more inotropes (80%) than in those with one inotrope (12.1%) or without (2%, p<.0001). More patients with cancer died (17.2%) than those with various congenital diseases (9.5%) or those without co morbidity (7%, p<.012). Similarly, more patients immune-compromised died (29.2%) compared to immune-competent patients (6.3%, p<.0001), and more with out- or in-hospital cardiac arrest (25%) compared to those without (8%, p<.0001). The contribution of cancer and chronic or congenital disease to NI or MI bad outcome is depicted in Figure 1 (p<.0001).
In this study we showed that less than half of critically ill children are admitted with CI (30.8%) and/or subsequently develop NI (12.7%) or MI (2%). Although patients with NI have higher mortality rate compared to patients with either CI or WI, most of those with poor outcome are immune-compromised patients with cancer or co morbidity. We also showed that only severity of illness is directly related to mortality and that severity, along with septic shock, co morbidity, complications, pathogenic microorganisms, and resource utilization are significantly higher in the NI and/or MI compared to the CI group. We finally showed that each infection group's relative contribution to the total PICU mortality is lower than that of the larger group of critically ill patients (organ failures, trauma, accidents, congenital diseases, cancer) without infection.
Our findings are clearly supported by those of previous studies, in which case mortality rates were higher in adults with septic shock (52.5%), or among patients requiring multiple inotropic agents (42.9%), undergoing bone marrow transplantation (38.5%) or having MOSF (18.6%). (10) Similarly, the risk of mortality in children admitted to the PICU after haematopoietic stem cell transplantation was very high, with one of the main life-threatening complications being systemic infections. (11) In these sensitive sub cohorts or patients, dysactivation of the immune system during an acute insult, with the subsequent release of humoral mediators from activated immune cells, leads to tissue injury and may be involved in the pathogenesis of ARDS, DIC, capillary leak syndrome, and to the development of multiple organ system failure. However, it should not be forgotten that, regarding septic shock, various polymorphisms are also contributing to increased illness severity and mortality in both NI and CI. (12)
Our results are also in agreement to those of previous studies, which showed that the risk of NI increased with central line use, prolonged intubation, ventilation, paralysis, total parenteral nutrition, or with a PRISM score greater than or equal to 10. (13) In a matched case-control study, there was a trend toward increased hospital mortality among cases with nosocomial bloodstream infections (23.7%vs. 10.5%), (14) which were almost identical to the one recorded in our study for patients with NI (24.6%) and WI (8.3%). Interestingly, hospital mortality in adult ICUs (27.8% overall) was higher in patients admitted with infection than in those who acquired infection in the ICU (45.0% vs. 32.4%). In that study, although the presence of infection per se did not influence mortality, the conditions of severe sepsis and septic shock were strong prognostic factors. (15) In addition, the rate of NI has been shown to be significantly higher in patients with inadequate empirical therapy (16.1%) than in those treated empirically with adequate antibiotics (3.4%). What we showed in this study, however, is that children with CI (4.5%) have the lowest mortality among groups, including the one WI, and the smallest contribution to the overall PICU mortality (1.4%), a finding that has not been recorded in previous studies. In an epidemiologic study of severe sepsis hospital mortality was 10.3%, been higher in children with underlying disease, surgical procedures, or human immunodeficiency virus infection. (1)
In conclusion, we showed that mortality was mostly attributed to the associated severe co morbidity, especially cancer or immune-compromised and independently associated with the severity of illness. The major limitation of our study relates to the use of a single unit data. However, our homogenous population, not including NICU's patients or patients from different geographic areas, might better generate nonbiased results regarding CI, NI, MI, and WI incidence and epidemiology in a PICU.
Conflict of interest: None declared.
(1.) Watson RS, Carcillo JA, Linde-Zwirble WT, Clermont G, Lidicker J, Angus DC. The epidemiology of severe sepsis in children in the United States. Am J Respir Crit Care Med 2003; 167(5): 695-701.
(2.) Moine P, Vercken JB, Chevret S, Chastang C, Gajdos P. Severe community-acquired pneumonia. Etiology, epidemiology, and prognosis factors. French Study Group for Community-Acquired Pneumonia in the Intensive Care Unit. Chest 1994; 105: 1487-1495.
(3.) Martin CM, Priestap F, Fisher H, et al. A prospective, observational registry of patients with severe sepsis: the Canadian Sepsis Treatment and Response Registry. Crit Care Med 2009; 37(1): 81-88.
(4.) Rello J. Demographics, guidelines, and clinical experience in severe community-acquired pneumonia. Crit Care 2008;12 Suppl 6:S2.
(5.) Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Intensive Care Med 2008; 34(1): 17-60.
(6.) Goldstein B, Giroir B, Randolph A, and the Members of the International Consensus Conference on Pediatric Sepsis. International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med 2005; 6: 2-8.
(7.) Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections. Am J Infect Control 1988; 16: 128-140.
(8.) Briassoulis G, Natsi L, Tsorva A, Hatzis T. Prior antimicrobial therapy in the hospital and other predisposing factors influencing the usage of antibiotics in a pediatric critical care unit. Ann Clin Microbiol Antimicrob 2004; 17: (3:4):1-14.
(9.) Moreno RP, Metnitz B, Adler L, Hoechtl A, Bauer P, Metnitz PG; SAPS 3 Investigators. Sepsis mortality prediction based on predisposition, infection and response. Intensive Care Med 2008; 34(3): 496-504.
(10.) Kutko MC, Calarco MP, Flaherty MB, et al. Mortality rates in pediatric septic shock with and without multiple organ system failure. Pediatr Crit Care Med 2003; 4(3): 333-337.
(11.) Tomaske M, Bosk A, Eyrich M, Bader P, Niethammer D. Risks of mortality in children admitted to the paediatric intensive care unit after haematopoietic stem cell transplantation. Br J Haematol 2003; 121(6): 886-891.
(12.) Jessen KM, Lindboe SB, Petersen AL, Eugen-Olsen J, Benfield T. Common TNF-alpha, IL-1 beta, PAI-1, uPA, CD14 and TLR4 polymorphisms are not associated with disease severity or outcome from Gram negative sepsis. BMC Infect Dis 2007; 18(7): 108.
(13.) Pollock EM, Ford-Jones EL, Rebeyka I, et al. Early nosocomial infections in pediatric cardiovascular surgery patients. Crit Care Med 1990; 18(4): 378-384.
(14.) Slonim AD, Kurtines HC, Sprague BM, Singh N. The costs associated with nosocomial bloodstream infections in the pediatric intensive care unit. Pediatr Crit Care Me. 2001; 2(2): 170-174.
(15.) Malacarne P, Langer M, Nascimben E, et al. Building a continuous multicenter infection surveillance system in the intensive care unit: findings from the initial data set of 9,493 patients from 71 Italian intensive care units. Crit Care Med 2008; 36(4): 1105-1113.
(16.) Garnacho-Montero J, Ortiz-Leyba C, Herrera-Melero I, et al. Mortality and morbidity attributable to inadequate empirical antimicrobial therapy in patients admitted to the ICU with sepsis: a matched cohort study. J Antimicrob Chemother 2008; 61(2): 436-441.
Corresponding author: George Briassoulis, MD,
Associate Professor, University of Crete
Head, Paediatric Intensive Care Unit, University Hospital
P.O. Box 1352, 71110 Heraklion, Crete, Greece
Tel. 2810-375024, e-mail: email@example.com
S. Ilia , A.M. Spanaki , E Geromarkaki , O. Filippou , E. Briassouli , D.M. Fitrolaki , E. Vasilaki , G. Briassoulis 
 Pediatric Intensive Care Unit, University Hospital, School of Health Sciences, University of Crete, Heraklion, Greece
 Paediatric Department, "Asklipeion Voulas" General Hospital, Athens, Greece
 1st Department of Internal Medicine--Propaedeutic, University of Athens, Greece
Table I. Demographic and clinical characteristics of studied patients related to infection classification. Infection WI CI groups Mean (SD) Mean (SD) Age (years) 5.9 (5) * 3.6 (4.4) * PRISM 7.6 (8) * 8.4 (6.5)# TISS 22.9 (12.2) * # 23.5 (12.4)AE LOS (days) 3.9 (10) * 5.5 (8.1) # LOMV (days) 2.4 (9.5) * 2.1 (5) # Frequency (%) Frequency (%) Boys / girls 60.8 / 39.2 61.8 / 38.2 Surgical patients 73.3 20 Major elective 78.9 15.4 surgeries Emergent operations 42.9 14.3 Severe sepsis 0 56 Septic shock 0 21.7 Free history 37.4 61.8 Co morbidity 32.7 33.1 Cancer 29.9 5.1 Immune-compromised 1.8 3.8 Mechanical 51.8 37.6 ventilation Inotropes/ 11.5 21.7 vasopressors Total parenteral 4 6.4 nutrition Central catheters 69.80 68.6 Invasive 11.6 22.9 Infection NI MI ANOVA groups Mean (SD) Mean (SD) p-Values Age (years) 5.1 (5.1) 3 (2.9) <.0001 PRISM 14.7 (7.3) * # 11.2 (7.4) <.0001 TISS 33.4 (13.7) * AE 34.1 (15.7) # <.0001 LOS (days) 30.3 (49.8) * #AE 10.6 (15.5)AE <.0001 LOMV (days) 23 (44) * #AE 3.2 (3.5)AE <.0001 Frequency (%) Frequency (%) Chi-square Boys / girls 53.8 / 46.2 20 / 80 <.05 Surgical patients 6.7 0 <.0001 Major elective 5.7 0 <.0001 surgeries Emergent operations 28.6 14.3 <.0001 Severe sepsis 36.9 50 <.0001 Septic shock 63.1 50 <.0001 Free history 21.5 0 <.0001 Co morbidity 49.2 40 <.0001 Cancer 29.2 60 <.0001 Immune-compromised 15.4 30 <.0001 Mechanical 90.8 70 <.0001 ventilation Inotropes/ 63.1 50 <.0001 vasopressors Total parenteral 10.8 30 <.0001 nutrition Central catheters 98.5 100 <.001 Invasive 49 85.7 <.0001 monitoring WI, without infection; CI, community infection; NI, nosocomial infection; MI, mixed infection; PRISM, Pediatric Risk of Mortality score; TISS, Therapeutic Intervention Scoring System; LOS, length of stay; LOMV, length of MV Table II. Absolute and relative frequencies of the commonest severe community (CI), nosocomial (NI) and mixed (MI) infections. INFECTION CI * Patients Within CI Within PICU (157) (510) n % % Septic Shock / (bacteremia , pneumonia , meningitis, endocarditis , osteomyelitis, toxic shock syndromes) 34 21.7 6.7 Pneumonia / severe sepsis 44 28 8.6 Meningitis / encephalitis / severe sepsis 26 16.6 5 Viral bronchiolitis / Upper respiratory tract infection a (cute respiratory failure) 35 22.3 6.9 Surgical infections / severe sepsis (empyema, abscesses) 10 6.4 2 Abdominal (peritonitis , pancreatitis) / severe sepsis 8 5 1.6 Urinary tract infection / severe sepsis 0 0 0 Total 157 100 30.8 INFECTION NI * Patients Within NI Within PICU (64) (510) n % % Septic Shock / (bacteremia , pneumonia , meningitis, endocarditis , osteomyelitis, toxic shock syndromes) 41 64.1 8 Pneumonia / severe sepsis 11 17.2 2.2 Meningitis / encephalitis / severe sepsis 0 0 0 Viral bronchiolitis / Upper respiratory tract infection a (cute respiratory failure) 0 0 0 Surgical infections / severe sepsis (empyema, abscesses) 4 6.2 0.8 Abdominal (peritonitis , pancreatitis) / severe sepsis 3 4.7 .6 Urinary tract infection / severe sepsis 5 7.8 1 Total 64 100 12.6 INFECTION MI * Patients Within MI Within PICU (10) (510) n % % Septic Shock / (bacteremia , pneumonia , meningitis, endocarditis , osteomyelitis, toxic shock syndromes) 5 50 1 Pneumonia / severe sepsis 3 30 .6 Meningitis / encephalitis / severe sepsis 1 10 .4 Viral bronchiolitis / Upper respiratory tract infection a (cute respiratory failure) 0 0 0 Surgical infections / severe sepsis (empyema, abscesses) 0 0 0 Abdominal (peritonitis , pancreatitis) / severe sepsis 1 10 .4 Urinary tract infection / severe sepsis 1 Total 10 100 2 WI, without infection; CI, community infection; NI, nosocomial infection; MI, mixed infection; PICU, Pediatric Intensive Care Unit
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ORIGINAL ARTICLE|
|Author:||Ilia, S.; Spanaki, A.M.; Geromarkaki, E.; Filippou, O.; Briassouli, E.; Fitrolaki, D.M.; Vasilaki, E|
|Publication:||Archives: The International Journal of Medicine|
|Date:||Oct 1, 2009|
|Previous Article:||Monaural performance intensity functions of average esophageal speech.|
|Next Article:||The effects of methylphenidate combined with exhaustive practice on the white cells of the male rat.|