PRISM Score Research Articles

Incidence, risk factors, and outcome of transfusion-related acute lung injury in critically ill children: A retrospective study

PurposeAcute lung injury (ALI) that develops within 6hours after transfusion (TRALI) is the leading cause of transfusion-related morbidity and mortality. Both incidence and patient and transfusion-related risk factors are well studied in the adult critically ill patient population. Clinical data on TRALI in the pediatric population are sparse and are mainly limited to case reports and hemovigilance reporting systems. The objective of this study was to determine incidence, risk factors, and outcome of TRALI in critically ill children. Materials and MethodsIn a retrospective cohort study, all first-time admissions to the pediatric intensive care unit from January 1, 2009, until December 31, 2012, were screened for onset of TRALI using the consensus criteria. ResultsOf 2294 admitted patients, 304 were transfused, of whom 21 (6.9%) developed TRALI. Compared with transfused control subjects, risk factors for TRALI were mechanical ventilation (odds ratio, 18.94 [2.38-2452.56]), sepsis (odds ratio, 7.20 [2.69-19.69]), and high Pediatric Risk of Mortality III score (odds ratio, 1.05 [1.01-1.10]). Patients with TRALI had a higher mortality and a longer duration of mechanical ventilation when compared with transfused control subjects. ConclusionsTransfusion-related ALI is relatively common in critically ill children. The incidence in the pediatric intensive care unit population is similar to that in adult intensive care unit patients. High PRISM score on admission, mechanical ventilation and sepsis were identified as independent risk factors, which may help to assess the risks and benefits of transfusion in critically ill patients.

PS-134 Prism Score And Noninvasive Ventilation (niv) For Acute Respiratory Failure

<h3>Introduction</h3> PRISM score (Paediatric risk mortality) is widely used to determine the risk of mortality in children in PICU. Recent studies had found correlation between the low values of PRISM with the success of noninvasive ventilation (NIV) and the high values with failure, but without clearly defined who will be called higher or lower value of PRISM score. <h3>Objective</h3> To evaluate the predictive value of PRISM score for NIV success in acute respiratory failure (ARF). <h3>Methods</h3> This is a prospective study. Are included all children admitted at PICU during January–December 2011. NIV was used as the primary support for ARF. We analysed the predictive value of the PRISM score using ROC curves and the trend of success change by Chi-square trend. <h3>Results</h3> A total of 42 patients were included. NIV success rate was 73.8%. Prism score in the success group was 9.5 ± 3.9 vs. 14.5 ± 6.6 points in the failure group (p = 0.0184). Max value was 27 points, min value 3 points. By ROC curves, PRISM &lt; 10 points before NIV results significant predictive factor for NIV success with predictive positive value 87.5%. By Chi-square trend it was found a significant trend of success reduction with increasing value of PRISM. For PRISM score = 10 up to 15 points, OR = 0.3 (95% CI 0.05–2.0) p = 0.2. For PRISM score &gt;15 points, OR = 0.08 (95% CI 0.01–0.5) p = 0.01. (χ²_{for linear trend} = 7.6, p &lt; 0.01). <h3>Conclusion</h3> PRISM score &lt;10 points is a significant predictive factor for NIV success. For PRISM score &gt;15 points the likelihood to have success is decreased significantly.

PO-0302 Prognostic Factors And Outcome Of Traumatic Brain Injury In Children Treated In Picu

Background Prediction of outcome based on certain prognostic factors is still challenging in paediatric traumatic brain injuries (TBI). Purpose of our study is to evaluate prognostic factors and outcome in children with TBI admitted in PICU. Methods The sample was 252 children with TBI who were admitted to PICU of Hippokratio General Hospital of Thessaloniki from 1999 to 2012. The statistical analysis was performed with the statistical package SPSS 17 and the application of statistical tests x 2 -test, anova, Mann Whitney test, t-test and Pearson Correlation. Age, sex, Glasgow Coma Scale (GCS), Paediatric Risk Index of Mortality (PRISM score), days of mechanical ventilation, hypotension and need of inotropic support as well as late hyperglycemia were correlated to outcome. Results Age groups, sex and mechanism of injury showed no correlation to outcome. Patients who died had statistically significant higher mean PRISM score (25,7 ± 8,4),compared with patients with good outcome (7,5 ± 4,7). The application of Pearson Correlation showed that the increased rating of PRISM score, as well as the low grading GCS were positively correlated to death, late hyperglycemia, haemodynamic instability, use of inotrops, poor outcome and disabilities (p Conclusion A high PRISM score, a low GCS, hypotension and hyperglycemia after the first 48 h, are directly associated with poor outcome and death. Knowledge of these prognostic factors may help neurocritical care specialists to improve outcome in paediatric TBI.

PO-0138 Stress Induced Gastrointestinal Bleeding In A Paediatric Intensive Care Unit : Which Risk Factors Should Necessitate Prophilaxis ?

<h3>Objective</h3> To determine the frequency and the risk factors of stress induced gastrointestinal bleeding (GIB) in critically ill children and to investigate the effect of prophilaxis. <h3>Setting</h3> 14-beded, tertiary care PICU <h3>Methods</h3> Records of 182 children admitted consecutively from December 2012 to May 2013 were retrospectively reviewed. 136 patients were eligible. The age ranged from 40 days to 18 years. Diagnosis, demographic data, risk factors, administration of prophilaxis, drugs used in medication, prescence and degree of GIB and complications were recorded. <h3>Results</h3> The male-female ratio was 1.3. Mean age was 5.9. Mean PRISM III score was 12.2 and 49.3% had PRISM score ≥ 10. Most frequent diagnosis was infectious diseases. Sixtyone (44.9%) children received prophylaxis in which antacids was used in 28(45.9%), sucralfate in 18(29.5%), proton pomp inhibitors (PPIs) in 51(83.6%) and 5 (8.2%) received H2 reseptor antagonist. The incidence of GIB was 15.4% (n = 21), in which 66.7% (n = 14) were mild, 23.8% (n = 5) were moderate, 4.8% (n = 1) was significant and 4.8% (n = 1) was massive. In children who received prophylaxis 17 (27.9%) cases developed GIB. Mechanical ventilation were found to be the only risk factor significantly associated with stress induced GIB. Also; mechanical ventilation and trauma was strongly significant (p &lt; 0.001) and coagulopathy/ thrombocytopenia, PRISM III ≥ 10, renal and hepatic failure, hypotension, and heart failure/ arrhythmia was found to be associated with the development of GIB in critically ill children (p &lt; 0.05). <h3>Conclusions</h3> GIB is a serious concern for PICU clinicians and intensivists are confused about the conflicting evidence supporting prophilaxis. We believe that prophylaxis could be beneficial for mechanically ventilated children. Also trauma, coagulopathy/thrombocytopenia, PRISM III ≥ 10, renal and hepatic failure, hypotension, and heart failure/arrhythmia must be kept in mind as risk factors requiring attention in PICU setting.

English

BACKGROUND: One of the scoring systems in Pediatric Intensive Care is the PRISM III (PAEDITRIC RISK OF MORTALITY III) which is an updated version of PRISM. Study was done at time point 72 hrs of initiation of therapy as well in a view that it will better prognosticate the outcome. AIMS: 1. To find out PRISM III within 1st 12 hours of admission and at 72-84 hours of admission, 2. To evaluate any correlation between PRISM III and mortality, 4.To find out correlation between individual components of PRISM III and mortality DESIGN: A prospective cross-sectional study from 1 st June 2011 —31 st December 2013. Data was collected for 81 patients. METHODS AND MATERIAL: Variables used in the study were: Systolic BP, Heart rate, Temperature, Mental status, Pupillary reaction, Acidosis, pH, Total CO2, PaCO2, PaO2, Glucose, Potassium, BUN, Creatinine, WBC count, Platelet count, PT/aPTT. A detailed entry of various parameters included in the scoring system was made in the first 12 hours and thereafter 72-84 hours. The most abnormal value recorded. STATISTICAL ANALYSIS: All the data collected analyzed using STATA 8 after entering in excel sheet. Based on the clinical and laboratory data , PRISM III score was calculated. Categorical variables were compared with Chi-Square test. Calibration was assessed using Hosmer Lemeshow's goodness of fit test. CONCLUSION: Overall mortality rate was 27%. Among all the variables of PRISM III Systolic blood pressure, Heart rate, pupils, platelets, PT/APTT had statistical significance with mortality at both 0-12 and 72-84 hours. Temperature, GCS, BUN and creatinine showed statistical significance at 72-84 hours. Mortality was high with the as the scores increased in Systolic blood pressure, Heart rate, Temperature, GCS, BUN, Creatinine, WBC, PT/APTT at both 12 hours and 72 hours. Study showed significant relationship between PRISM III score and outcome. p value at 12 hours was 0.00961 which was significant and p value at 72-84 hours was 0.0000 which was extremely significant which suggested that mortality increased as the PRISM score increased and PRISM III score at 72-84 hours has depicted it better.

Predictors of Mortality in Mechanically Ventilated Critical Pertussis in a low Income Country.

BackgroundCritical pertussis is characterized by severe respiratory failure, important leukocytosis, pulmonary hypertension, septic shock and encephalopathy.AimTo describe the clinical course of critical pertussis, and identify predictors of death at the time of presentation for medical care.MethodologyRetrospective study conducted in children’s hospital Tunisian PICU between 01 January and 31 October 2013. Patients with critical pertussis confirmed by RT-PCR and requiring mechanical ventilation were included. Predictors of death were studied.ResultsA total of 17 patients was studied. Median age was 50 days. Mortality was 23%. Predictors risk of mortality were : high PRISM score (Pediatric Risk of Mortality Score) (p=0,007), shock (p=0,002), tachycardia (p=0,005), seizures (p=0,006), altered mental status (p=0,006), elevated WBC count (p=0,003) and hemodynamic support (p=0022). However, the difference did not reach statistical significance in comorbidity, pneumoniae, high pulmonary hypertension or exchange transfusion. Concomitant viral or bacterial co-infection was not related to poor outcome.ConclusionYoung infants are at high risk to have critical pertussis. Despite advances in life support and the treatment of organ failure in childhood critical illness, critical pertussis remains difficult to treat.

Lactate clearance as a marker of mortality in pediatric intensive care unit.

To correlate lactate clearance with Pediatric Intensive Care Unit (PICU) mortality. 45 (mean age 40.15 mo, 60% males) consecutive admissions in the PICU were enrolled between May 2012 to June 2013. Lactate clearance (Lactate level at admission - level 6 hr later x 100 / lactate level at admission) in first 6 hours of hospitalization was correlated to in-hospital mortality and PRISM score. Twelve out of 45 patients died. 90% died among those with delayed/poor clearance (clearance <30%) compared to 8.5% in those with good clearance (clearance >30%) (P<0.001). Lactate clearance <30% predicted mortality with sensitivity of 75%, specificity of 97%, positive predictive value of 90%, and negative predictive value of 91.42%. Predictability was comparable to PRISM score >30. Lactate clearance at six hours correlates with mortality in the PICU.

ABSTRACT 753

Background and aims: The PRISM score (Pediatric Risk of Mortality Score) is an objective measurement of 14 physiological variables which reflect the severity of illness. Aims: Aiming to assess significance of the PRISM score in the prediction of the outcome in newly established PICU at Tripoli-Libya. Methods: Prospectively and randomly selective study. This study involved 139 patients selected randomly from the period of May 2006 till May 2007, 122 of them survived and 17 died (12.2% mortality). Results: The PRISM score showed strong correlation with the outcome in terms of survival (p<0.01), and there was no significant difference between predicted and observed mortality rate at different PRISM scores especially at lower scores. The ROC curve plotted showed good results with an area of 70% under the curve. The age showed no predictive power for the outcome (p>0.6), and the predicted/observed mortality rates were nearly the same of neonates and older patients.The other study variables name, sex and diagnosis showed no significance to the outcome and were early excluded from the study. Conclusions: The study concluded that the PRISM score is suitable scoring system for the prediction of the mortality and survival rates in EL-KHADRA PICU provided the currant circumstances, services and experience.

ABSTRACT 669

Background and aims: The pediatric critical care patient has one of the highest risks of malnutrition and when this is present, the prognosis is worst. Aims: The aim of this study was to determine de relation between malnutrition and physiological stability with the mortality in a pediatric critical care unit. Methods: Descriptive, transversal, observational, analytic study. All the patients that were admitted in the pediatric critical care unit between august 2011 and august 2012 were included. The study variables were morbidity and mortality rates, nutritional assessment, age, gender, and physiological status. There was used descriptive statistics for the analysis of qualitative data and the discontinuous variables; and central and dispersion measurements for the continuous variables (SPPS v.19). Results: There were analyzed 219 patients, 33.8% of the patients presented a cardiovascular disease; the physiological stability of the patients was measure with the PRISM score, according with this, 80.4% of the patients were at low risk, 19.2% moderate risk and only 0.5% at high risk. The patients with a moderate or high risk PRISM score have 29 more chances of dying tan those with low risk PRISM score (Ji2; OR= 29.25, p<0.05). There were no significant correlation between the physiological stability and the nutritional assessment. Conclusions: We can conclude that the early identification of the PRISM score in the pediatric patient that is admitted in the pediatric critical care unit it’s very important because for further evaluation, treatment measures and prognosis in an effort for decreasing the mortality.

ABSTRACT 451

Background and aims: The mechanical ventilator related pneumonia is a high index morbidity and mortality complication with very particular features different from nosocomial pneumonia in non intubated patients. The aim of this study was to determine the length of stay in the pediatric patients with mechanical ventilator related pneumonia in a pediatric critical care unit. Aims: The aim of this study was to determine the length of stay in the pediatric patients with mechanical ventilator related pneumonia in a pediatric critical care unit. Methods: Descriptive, retrospective, observational study. All the patients that were admitted in the pediatric critical care unit between august 2010 and january 2011 with nosocomial pneumonia diagnose. The study variables were length of stay in days, diagnosis, temperature, positive culture, age, white blood cells count, chest X-ray, PaFi, OI, pneumothorax, cardiac arrest, antibiotic treatment and outcome. There was used descriptive statistics for the analysis with the statistical package EPI INFO 3.5.3. Results: There were analyzed the clinical records of 91 patients with nosocomial pneumonia in the pediatric critical care unit, difference in PRISM score (36 vs. 25), length of stay > 21 days, previous PICU admission, ward admittance are related with higher mortality. (p<0.01). The main pathogens were Klebsiella pneumoniae (26%) and Pseudomonas Aeruginosa (26%). Conclusions: We can conclude that the variables related with physiological stability at the admission and the level of damage in the gases interchange is positive correlated with more days of treatment and with a higher mortality rate; the risk of mortality increases as the age decreases.

ABSTRACT 466

Background and aims: The red cell distribution width (RDW) is a biomarker routinely reported in automated complete blood counts. Elevated RDW values are linked to worsened pulmonary function tests in the adult population. Aims: We examined whether RDW on admission was associated with pulmonary dysfunction in the PICU. Methods: We retrospectively examined data from PICU patients admitted between 2009 and 2013. All patients ventilated for over 24 hours were eligible for inclusion. RDW values were divided in quintiles; calculated differences were corrected for anemia, degree of inflammation and PRISM score. Results: 882 patients were screened, 271 were included. Mean age was 4.6 years, mean PRISM score 7.2, mean RDW on admission was 14.5, mean length of ventilation 8.9 days. Patients who had higher baseline RDW values had significantly more ventilation days (p < 0.001) and lower nadir PaO2/FiO2 values (p = 0.023). Hematocrit on admission and CRP values were not linked to RDW. Figures 1 and 2 display the differences between the quintiles.FigureConclusions: Higher RDW values on admission are associated with pulmonary dysfunction, reflected in more ventilation days and lower nadir PaO2/FiO2 values in our cohort of PICU patients. RDW should be evaluated as a possible parameter of organ failure in a PICU.

Mo1132 A Prospective Analysis of the Rates and Predictors of Readmissions Following Hospitalization for Gastrointestinal Bleeding

Background: Most patients with gastro-esophageal reflux disease (GERD) experience relief following treatment with proton pump inhibitors (PPI). As many as 17-44% of patients, however, exhibit partial or non-response to PPI therapy. Self-report of symptoms is fundamental to diagnosing and monitoring treatment response in these patients, yet most extant GERD PRO instruments fail to meet development best-practices as described by the FDA. We therefore sought to develop and validate a PRO instrument for clinical trials involving patients with GERD who are PPI partial-responders. Methods: We conducted a systematic literature review, held patient focus groups, convened an expert panel, and conducted cognitive interviews in order to establish content validity. Eligible participants took PPI therapy for at least 8 weeks, had undergone an upper endoscopy revealing no more than L.A. Classification Grade A esophagitis, and scored at least 8 points on the GerdQ. We used ATLAS.TI software to code focus group transcripts, generating frequencies at the domain, subdomain, and primary code levels. Results informed the construction of a conceptual framework, and guided development of 26 draft items. These items and subsequent revisions were reviewed by two expert panels and debriefed with cognitive interview participants. These efforts resulted in PRISM, a 21-item instrument that underwent psychometric evaluation during a Phase IIB drug trial (Protocol SPD557-206). Results: A diverse sample participated in focus groups (N=38; mean age=50.8; 55%M), saturating the codebook and supporting a preliminary conceptual framework. Following drafting of items, a new sample of patients participated in cognitive interviews for purposes of item revision (N=20; Age=52.6; 55%M). During the trial, an exploratory item assessment (n=104) revealed that all but one of the PRISM items met expectations for internal and external indices. A confirmatory factor analysis (n=220) resulted in a 4-factor model displaying the highest goodness-of-fit scores. All domains had a high inter-item correlation (Cronbach's α>0.8). Test-retest reliability was excellent with intraclass correlation also exceeding 0.8. Evidence for convergent validity was strong, with highly significant (p<.01) correlations between average weekly PRISM scores and severity anchors, and significant (p<.05) correlations with RDQ sub-scales. Three of 4 PRISM sub-scale scores were highly responsive to changes in both self-reported severity and symptom relief. Cumulative distribution functions revealed significant differences between responders and non-responders. Conclusion: PRISM demonstrates strong validity and psychometric properties in a clinical trial setting. Developed in line with FDA guidance on PROs, PRISM represents an new outcome measure for GERD patients with partial or nonresponse to PPI therapy.

Mo1131 Development and Validation of Prism: A Patient Reported Outcome Measure for GERD Patients Who Are Partial or Non-Responders to PPI Therapy

Background: Most patients with gastro-esophageal reflux disease (GERD) experience relief following treatment with proton pump inhibitors (PPI). As many as 17-44% of patients, however, exhibit partial or non-response to PPI therapy. Self-report of symptoms is fundamental to diagnosing and monitoring treatment response in these patients, yet most extant GERD PRO instruments fail to meet development best-practices as described by the FDA. We therefore sought to develop and validate a PRO instrument for clinical trials involving patients with GERD who are PPI partial-responders. Methods: We conducted a systematic literature review, held patient focus groups, convened an expert panel, and conducted cognitive interviews in order to establish content validity. Eligible participants took PPI therapy for at least 8 weeks, had undergone an upper endoscopy revealing no more than L.A. Classification Grade A esophagitis, and scored at least 8 points on the GerdQ. We used ATLAS.TI software to code focus group transcripts, generating frequencies at the domain, subdomain, and primary code levels. Results informed the construction of a conceptual framework, and guided development of 26 draft items. These items and subsequent revisions were reviewed by two expert panels and debriefed with cognitive interview participants. These efforts resulted in PRISM, a 21-item instrument that underwent psychometric evaluation during a Phase IIB drug trial (Protocol SPD557-206). Results: A diverse sample participated in focus groups (N=38; mean age=50.8; 55%M), saturating the codebook and supporting a preliminary conceptual framework. Following drafting of items, a new sample of patients participated in cognitive interviews for purposes of item revision (N=20; Age=52.6; 55%M). During the trial, an exploratory item assessment (n=104) revealed that all but one of the PRISM items met expectations for internal and external indices. A confirmatory factor analysis (n=220) resulted in a 4-factor model displaying the highest goodness-of-fit scores. All domains had a high inter-item correlation (Cronbach's α>0.8). Test-retest reliability was excellent with intraclass correlation also exceeding 0.8. Evidence for convergent validity was strong, with highly significant (p<.01) correlations between average weekly PRISM scores and severity anchors, and significant (p<.05) correlations with RDQ sub-scales. Three of 4 PRISM sub-scale scores were highly responsive to changes in both self-reported severity and symptom relief. Cumulative distribution functions revealed significant differences between responders and non-responders. Conclusion: PRISM demonstrates strong validity and psychometric properties in a clinical trial setting. Developed in line with FDA guidance on PROs, PRISM represents an new outcome measure for GERD patients with partial or nonresponse to PPI therapy.

ABSTRACT 506

Background and aims: The outcome of patients presented with cardiac failure admitted to Pediatric Intensive Care Unit (PICU) depends on the underlying disease. Aims: The aim of this study was to examine clinical characteristics, severity of illness, and assess the outcome of children diagnosed with cardiac disease (CD) in Albania. Methods: This is a retrospective study. Are included children diagnosed with cardiac disease during the period January 2011-december 2012. We evaluated clinical characteristics, severity of illness according scoring systems (PRISM, PELOD, GCS) and mortality. Results: During the study period, 81 children with cardiac disease are admitted at our PICU (7.1% of all admissions). Congenital heart diseases 81.5%, cardiomyopathy (12.3%), and supraventricular tachycardia (6.2%) were the leading causes of CD among patients without MV. Surgical correction has underwent 57.7% of congenital heart disease with 11% mortality. Age (p< 0.01) and GCS (p< 0.4) differed among disease sub cohorts, but none of the severity of illness or intervention scores. The PRISM and PELOD scores (p< 0.01) were significantly higher and the GCS lower (p< 0.001) in patients who needed intubation. Cardiac patients admitted to our multidisciplinary PICU had low mortality (2% of admission). Cardiomyopathy and complex congenital heart disease were the main causes of mortality. Conclusions: In patients with CD severity of illness, intervention scores are higher among those who need intubation. Mortality, age and support differ among various causes of CD. Prenatal diagnosis, innovations in cardiothoracic surgical techniques, and improvements in perioperative management have contributed to increased survival of infants with CD.

Predictors of stress among parents in pediatric intensive care unit: a prospective observational study.

To determine the sociodemographic and clinical factors leading to stress among parents whose children are admitted in pediatric intensive care unit (PICU). A prospective observational study was conducted in PICU of a tertiary care hospital of north India. Parents of children admitted to PICU for at least 48 h duration were eligible for participation. At the end of 48 h, parental stress was assessed using parental stress scale (PSS:PICU) questionnaire which was administered to the parents. Baseline demographic and clinical parameters of children admitted to PICU were recorded. The parental stress was compared with demographic and clinical characteristics of children using appropriate statistical methods. A total of 49 parents were finally eligible for participation. Mean (SD) parental stress scores was highest in domains of procedures [1.52 (0.66)] and behavior and emotional [1.32 (0.42)] subscales. Mean (SD) total parental stress score among intubated children [1.31 (0.25)] was significantly more than among non intubated children [0.97 (0.26)] (p < 0.001). However, parental stress score were comparable in terms of gender (p = 0.15) and socioeconomic status (p = 0.32). On subscale analysis, it was found that professional communication is a significant stressor in age groups 0-12 mo [0.61(0.41)] (p = 0.02). It was observed that parents of intubated children were significantly stressed by the physical appearance of their children (p < 0.001), procedures performed on them (p = 0.008) and impairment in parental role (p = 0.002). Total parental stress score had a positive correlation with PRISM score (r = 0.308). Indian parents are stressed maximally with environment of PICU. Factor leading to parental stress was intubation status of the child and was not affected by gender or socio demographic profile of the parents.

348

Introduction: Critically ill children with status asthmaticus frequently present with fever and abnormal chest radiographs. These children are often placed on broad-spectrum antibiotic therapy for clinical suspicion of pneumonia. However, many of these children do not have pneumonia. While previous studies have shown that serum procalcitonin(PCT) levels can predict bacterial infection and decrease antibiotic overuse, there is limited data on PCT levels in children with status asthmaticus. For example, it is not known whether corticosteroids affect PCT levels. The specific aims of this study are to measure PCT levels in critically ill children with status asthmaticus and determine whether increased PCT correlates with presence of co-existing bacterial pneumonia. Methods: With IRB approval, we conducted a 12-month, prospective, cross-sectional study of patients from 1 to 21 years of age who were admitted to the PICU with status asthmaticus. We excluded patients with a known bacterial infection outside the lung, chronic lung disease, cystic fibrosis, bronchiectasis or immunodeficiency. After informed consent, we obtained PCT levels within 8 hours of admission. Patient demographic and clinical data including PRISM scores, radiographic and laboratory results and hospital and ICU length of stay were also collected. A blinded clinician and radiologist reviewed the charts retrospectively for the diagnosis of bacterial pneumonia. Statistical analysis included descriptive statistics, the Mann-Whitney U test, and regression analysis. Results: 61 patients with a median age of 7.3 years (IQR 4, 10) were included in the study. 51% were male. Average PRISM score was 2.7(SD 2.9). 3 patients (5%) were determined to have bacterial pneumonia, whereas 52 (85%) did not. 6 patients (10%) were indeterminate. The mean PCT level for all patients was 0.65ng/ml (SD 1.54). There was no significant difference in the mean PCT levels between the patients with and without bacterial pneumonia (0.33(SD 0.36) vs. 0.69(SD1.67), p=0.44). Regression analysis revealed a significant association with presence of fever with increased PCT level (p=0.004), but no significant correlation between PCT level and presence of pneumonia, length of stay PRISM scores or receipt of antibiotics. Conclusions: Status asthmaticus is not associated with a significant elevation in procalcitonin. Although our study did not find an association with co-existing bacterial pneumonia and PCT elevation, the small sample size may have biased our results - further studies are indicated.

653

Introduction: Critically ill children suffer from post-traumatic stress disorder and deficits in attention, and spatial and verbal memory, following discharge. The impact of critical illness on executive function, which is necessary for purposeful, goal-directed, and problem-solving behavior in addition to higher learning, has not been studied in children. Methods: In this prospective cohort study, we enrolled patients aged 5-17 years admitted to the pediatric intensive care unit (PICU) with 1 of the following: mechanical ventilation (MV), vasopressor therapy, a Pediatric Risk of Mortality (PRISM III) score over 15, or a neurologic admission diagnosis. We collected demographic data upon enrollment and in-hospital data until PICU discharge. Our primary outcome was executive function, which is measured by the Behavior Rating Inventory of Executive Function (BRIEF). The BRIEF was completed upon enrollment and then at 3 and 6 months following discharge. The BRIEF assesses 8 domains of executive function and is reported as a Global Executive Composite (GEC) T-score. For the GEC, a T-score of 50 reflects the age/gender adjusted norm for that population, and a 10 point increase indicates a 1-standard deviation (SD) worsening in outcome. The Child PTSD Symptom Scale was completed in children at least 8 years of age. Linear regression assessed the relationship between duration of mechanical ventilation with GEC outcomes at 3 months and 6 months separately, adjusting for a priori determined covariates (PRISM and baseline GEC scores). Results: The median PRISM score for the 99 patients enrolled was 6 (2.5, 11) reflecting moderate illness, with 67.7% of patients requiring MV. Median executive function scores (GEC) were 0.5 SD worse than population norms at 3 months [54.6 (p=0.1)] and 6 months [57.1 (p=0.08)] following discharge, though were not significantly different from baseline. In MV patients, however, the GEC was significantly worse at 3 months compared to baseline (p=0.03). For descriptive purposes the 8 subdomains of executive function were analyzed. In our cohort, working memory was significantly worse at 3 months [T-score 55.8 (p=0.02)], and the ability to initiate thought was significantly worse only at 6 months [T score 55.9 (p=0.03)] when compared to baseline. Among patients requiring MV, working memory was significantly worse at 3 months [T-score 55.8 (p=0.01)], and the ability to initiate thought was significantly worse at 3 months [T-score of 52.5 (p=0.03)] and 6 months [53.9 (p=0.05)] following discharge. In patients > 8 years, PTSD occurred in 7% of patients at 3 months, with only 5% having ongoing symptoms at 6 months. In our multivariable regressions, baseline executive function was the strongest predictor of worse GEC scores at 3 and 6 months. Conclusions: Children suffer from cognitive dysfunction and PTSD following critical illness. Larger studies are needed to determine the prevalence of these serious morbidities and to identify potentially modifiable risk factors to target interventional therapies.

739

Introduction: Critically ill children and especially infants in the pediatric intensive care unit (PICU) are at high risk of underfeeding. Methods: Retrospective chart review of patients admitted to PICU for at least >48 h between (10/12-12/12). All intravenous fluids, parenteral and enteral nutrition for the first 8D were analyzed. Resting energy expenditure (REE) and protein needs were estimated by the Schoefield equation (Hum Nutr Clin Nutr 39, 1985) and ASPEN Clinical Guidelines (JPEN 33, 2009), respectively. Results are mean ± SD. Results: There were 156 (sex M/F 79/77) patients identified, with a total length of hospital stay of 12 (8.5-19)(median (interquartile range)) days, with age 1.3 (0.3-4.6) yr; PRISM score 5 (2.5-8.0); and Pediatric Logistic Organ Dysfunction (PELOD) 11 (2-11). Estimated REE by Schoefield for <2 y, n=93; 2-5 y, n=30; >6 y, n=33; was 52.8 ± 9.1, 54.7 ± 7.7, 36 and 33.5 ± 7.4 kcal/kg/d, respectively. Average 8D energy intake was 46.6 ± 22.5, 33.4 ± 13.3, and 18.0 ± 12.3 kcal/kg/d, respectively (paired t-test all p <0.01). Thus only 93, 62 and 53 %, respectively of 8D energy needs were met. Patients met <90% of estimated REE: <2 y: 51%; 2-6 y: 87% and >6y: 91% (Chi2 p <0.001). Average energy deficit was 0.2, 2.3 and 2.0 times the daily energy needs in <2, 2-5 and >6y old children. Estimated protein needs by ASPEN guidelines for <2 y, 2-5 y and >6 y was 2.5 ± 0.1, 1.8 ± 0.1 and 1.6 ± 0.1 g/kg/d; average 8D protein intake was 1.0 ± 0.7, 0.9 ± 0.6 and 0.5 ± 0.5 g/kg/d, respectively (all p <0.001). Thus only 41, 51 and 33 %, respectively of protein needs were met. Patients met <90% of estimated protein needs: <2 y: 94%; 2-6 y: 87% and >6y: 94% (Chi2 p <0.43). Average protein deficit was 3.2, 2.8 and 3.1 times the daily energy need in <2, 2-5 and >6y old children. Conclusions: Only 53 to 93% of energy and 33 to 51 % of protein needs were met during the first 8 days of PICU stay. Critically ill children are at significant risk for underfeeding, especially protein underfeeding. Further efforts should be made to meet energy and protein needs of PICU patients.

FRI0285 Prognostic factors associated with mortality in pediatric systemic lupus erythematosus (SLE).

BackgroundSystemic lupus erythematosus (SLE) is an autoimmune disease with significant morbidity and mortality. Recent studies suggest that childhood-onset SLE is more frequent and severe in nonwhite populations, especially Hispanics.ObjectivesTo assess...

1083

Introduction: Erythropoietin (EPO) is frequently used to enhance erythropoiesis in anemic patients to avoid unnecessary transfusions. However, data to support its utility in pediatric intensive care unit (PICU) patients is lacking. This is in part due to the fact that most transfusions occur within the first 48 hours of admission. There may be a subgroup of patients who may benefit from EPO therapy to prevent later transfusions. The purpose of this study is to explore the clinical rationale and assess the efficacy of EPO in certain PICU patients at risk of requiring blood transfusion. Hypothesis: EPO administration will result in significant rise in hemoglobin in anemic patients in the PICU. Methods: A retrospective review of all patients admitted to a tertiary care PICU in 2010 and 2011 that received EPO. Diagnosis, PRISM scores, hemoglobin values (Hgb), number of EPO doses, and transfusions were recorded. Mann-Whitney test and regression analysis were utilized. Values were expressed in means and standard deviations. Results: Of 2595 admissions, 110 received EPO. The average age of the 110 patients (51 females and 59 males) was 7.4 ± 5.4 years and the median PRISM score was 3. Of these, 8 patients had ongoing blood loss and were excluded, 33 were transfused (Group A) and 69 were not (Group B). Pre EPO Hgb was 7.6 gm/dl ± 1.4 vs. 8.8 gm/dl ± 1.2 for group A and B respectively (p =.78). Regression analysis revealed that a higher PRISM score predicted need for transfusions (p =.024). In group B, Hgb rise following EPO therapy was significant in trauma (1.0 gm/dl, p =.018, n = 9) and post-operative (± 0.87 gm/dl, p =.012, n = 31) patients, but not in patients with infections (0.23 gm/dl, p =.573, n = 20). Patients receiving? 3 doses of EPO had higher Hgb rise (1.06 gm/dl ± 1.52) compared to those receiving fewer doses (0.3 gm/dl ± 1.4), p =.05. There were no reported complications. Conclusions: EPO induced a significant hemoglobin rise in sub-acute trauma and post-operative patients but not in patients with infections. Fewer than 3 doses of EPO were not effective, and PRISM score predicted transfusion requirement. This data will be helpful in targeting specific populations in future prospective studies.