ABSTRACT
BACKGROUND: Acute kidney injury (AKI) is a common complication of critical illness and associated with high morbidity and mortality. Optimal timing of continuous kidney replacement therapy (CKRT) in children is unknown. We aimed to measure the association between timing of initiation and mortality. METHODS: This is a single-center retrospective cohort study of pediatric patients receiving CKRT from 2013 to 2019. The primary exposure, time to CKRT initiation, was measured from onset of stage 3 AKI during hospitalization (defined using Kidney Disease: Improving Global Outcomes creatinine and urine output criteria) and analyzed as both a continuous and categorical variable. The primary outcome was ICU mortality. RESULTS: Ninety-nine patients met criteria for analysis. Overall mortality was 39% (39/99). Median time from stage 3 AKI onset to CKRT initiation was 1.5 days in survivors and 5.5 days in nonsurvivors (p < 0.001). In multivariable analysis, increased time to CKRT initiation was independently associated with mortality [OR 1.02 per hour (95% CI 1.01-1.04), p < 0.001]. Longer time to CKRT initiation was associated with higher odds of mortality in ascending time intervals. Patients started on CKRT > 2 days compared to < 2 days after stage 3 AKI onset had higher mortality (65% vs. 5%, p < 0.001), longer median ICU length of stay (25 vs. 12 d, p < 0.001), longer median CKRT duration (11 vs. 5 d, p < 0.001), and fewer AKI-free days (0 vs. 14 d, p < 0.001). CONCLUSIONS: Longer time to initiation of CKRT after development of severe AKI is independently associated with mortality. Consideration of early CKRT in this high-risk population may be a strategy to reduce mortality and improve recovery of kidney function. However, there remains significant heterogeneity in the definition of early versus late initiation and the optimal timing of CKRT remains unknown.
Subject(s)
Acute Kidney Injury , Continuous Renal Replacement Therapy , Critical Illness , Time-to-Treatment , Humans , Retrospective Studies , Female , Male , Acute Kidney Injury/mortality , Acute Kidney Injury/therapy , Acute Kidney Injury/etiology , Critical Illness/mortality , Critical Illness/therapy , Child , Child, Preschool , Continuous Renal Replacement Therapy/methods , Infant , Time-to-Treatment/statistics & numerical data , Adolescent , Hospital Mortality , Time Factors , Intensive Care Units, Pediatric/statistics & numerical dataABSTRACT
BACKGROUND: Unplanned extubation (UE) is defined as unintentional dislodgement of an endotracheal tube (ETT) from the trachea. UEs can lead to instability, cardiac arrest, and may require emergent tracheal re-intubation. As part of our hospital-wide quality improvement (QI) work, a multidisciplinary committee reviewed all UEs to determine contributing factors and evaluation of clinical outcomes to develop QI interventions aimed to minimize UEs. The objective was to investigate occurrence, contributing factors, and clinical outcomes of UEs in the pediatric ICU (PICU), cardiac ICU (CICU), and neonatal ICU (NICU) in a large academic children's hospital. We hypothesized that these would be substantially different across 3 ICUs. METHODS: A single-center retrospective review of UEs in the PICU, CICU, and NICU was recorded in a prospective database for the last 5 y. Consensus-based standardized operational definitions were developed to capture contributing factors and adverse events associated with UEs. Data were extracted through electronic medical records by 3 respiratory therapists and local Virtual Pediatric Systems (VPS) database. Consistency of data extraction and classification were evaluated. RESULTS: From January 2016-December 2021, 408 UEs in 339 subjects were reported: PICU 52 (13%), CICU 31 (7%), and NICU 325 (80%). The median (interquartile range) of age and weight was 2.0 (0-4.0) months and 5.3 (3.0-8.0) kg. Many UE events were not witnessed (54%). Common contributing factors were routine nursing care (no. = 70, 18%), ETT retaping (no. = 62, 16%), and being held (no. = 15, 3.9%). The most common adverse events with UE were desaturation < 80% (33%) and bradycardia (22.8%). Cardiac arrest occurred in 12%. Sixty-seven percent of UEs resulted in re-intubation within 72 h. The proportion of re-intubation across 3 units was significantly different: PICU 62%, CICU 35%, NICU 71%, P < .001. CONCLUSIONS: UEs occurred commonly in a large academic children's hospital. Whereas UE was associated with adverse events, re-intubation rates within 72 h were < 70% and variable across the units.
Subject(s)
Airway Extubation , Heart Arrest , Infant, Newborn , Humans , Child , Airway Extubation/methods , Risk Factors , Intensive Care Units, Neonatal , Intensive Care Units, Pediatric , Intubation, Intratracheal/adverse effects , Heart Arrest/therapyABSTRACT
OBJECTIVES: Mechanically ventilated children post-hematopoietic cell transplant (HCT) have increased morbidity and mortality compared with other mechanically ventilated critically ill children. Tracheal intubation-associated adverse events (TIAEs) and peri-intubation hypoxemia universally portend worse outcomes. We investigated whether adverse peri-intubation associated events occur at increased frequency in patients with HCT compared with non-HCT oncologic or other PICU patients and therefore might contribute to increased mortality. DESIGN: Retrospective cohort between 2014 and 2019. SETTING: Single-center academic noncardiac PICU. PATIENTS: Critically ill children who underwent tracheal intubation (TI). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Data from the local airway management quality improvement databases and Virtual Pediatric Systems were merged. These data were supplemented with a retrospective chart review for HCT-related data, including HCT indication, transplant-related comorbidity status, and patient condition at the time of TI procedure. The primary outcome was defined as the composite of hemodynamic TIAE (hypo/hypertension, arrhythmia, cardiac arrest) and/or peri-intubation hypoxemia (oxygen saturation < 80%) events. One thousand nine hundred thirty-one encounters underwent TI, of which 92 (4.8%) were post-HCT, while 319 (16.5%) had history of malignancy without HCT, and 1,520 (78.7%) had neither HCT nor malignancy. Children post-HCT were older more often had respiratory failure as an indication for intubation, use of catecholamine infusions peri-intubation, and use of noninvasive ventilation prior to intubation. Hemodynamic TIAE or peri-intubation hypoxemia were not different across three groups (HCT 16%, non-HCT with malignancy 10%, other 15). After adjusting for age, difficult airway feature, provider type, device, apneic oxygenation use, and indication for intubation, we did not identify an association between HCT status and the adverse TI outcome (odds ratio, 1.32 for HCT status vs other; 95% CI, 0.72-2.41; p = 0.37). CONCLUSIONS: In this single-center study, we did not identify an association between HCT status and hemodynamic TIAE or peri-intubation hypoxemia during TI.
Subject(s)
Critical Illness , Hematopoietic Stem Cell Transplantation , Child , Humans , Retrospective Studies , Critical Illness/therapy , Intubation, Intratracheal/adverse effects , Intubation, Intratracheal/methods , Hypoxia/epidemiology , Hypoxia/etiology , Intensive Care Units, PediatricABSTRACT
OBJECTIVES: To determine the association between preintubation respiratory support and outcomes in patients with acute respiratory failure and to determine the impact of immunocompromised (IC) diagnoses on outcomes after adjustment for illness severity. DESIGN: Retrospective multicenter cohort study. SETTING: Eighty-two centers in the Virtual Pediatric Systems database. PATIENTS: Children 1 month to 17 years old intubated in the PICU who received invasive mechanical ventilation (IMV) for greater than or equal to 24 hours. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: High-flow nasal cannula (HFNC) or noninvasive positive-pressure ventilation (NIPPV) or both were used prior to intubation in 1,825 (34%) of 5,348 PICU intubations across 82 centers. When stratified by IC status, 50% of patients had no IC diagnosis, whereas 41% were IC without prior hematopoietic cell transplant (HCT) and 9% had prior HCT. Compared with patients intubated without prior support, preintubation exposure to HFNC (adjusted odds ratio [aOR], 1.33; 95% CI, 1.10-1.62) or NIPPV (aOR, 1.44; 95% CI, 1.20-1.74) was associated with increased odds of PICU mortality. Within subgroups of IC status, preintubation respiratory support was associated with increased odds of PICU mortality in IC patients (HFNC: aOR, 1.50; 95% CI, 1.11-2.03; NIPPV: aOR, 1.76; 95% CI, 1.31-2.35) and HCT patients (HFNC: aOR, 1.75; 95% CI, 1.07-2.86; NIPPV: aOR, 1.85; 95% CI, 1.12-3.02) compared with IC/HCT patients intubated without prior respiratory support. Preintubation exposure to HFNC/NIPPV was not associated with mortality in patients without an IC diagnosis. Duration of HFNC/NIPPV greater than 6 hours was associated with increased mortality in IC HCT patients (HFNC: aOR, 2.41; 95% CI, 1.05-5.55; NIPPV: aOR, 2.53; 95% CI, 1.04-6.15) and patients compared HCT patients with less than 6-hour HFNC/NIPPV exposure. After adjustment for patient and center characteristics, both preintubation HFNC/NIPPV use (median, 15%; range, 0-63%) and PICU mortality varied by center. CONCLUSIONS: In IC pediatric patients, preintubation exposure to HFNC and/or NIPPV is associated with increased odds of PICU mortality, independent of illness severity. Longer duration of exposure to HFNC/NIPPV prior to IMV is associated with increased mortality in HCT patients.
Subject(s)
Hematopoietic Stem Cell Transplantation , Noninvasive Ventilation , Respiratory Distress Syndrome , Respiratory Insufficiency , Cannula , Child , Cohort Studies , Humans , Intubation, Intratracheal/adverse effects , Oxygen Inhalation Therapy , Retrospective StudiesABSTRACT
BACKGROUND: It is not known whether brain magnetic resonance imaging (MRI) abnormalities in pediatric sepsis are associated with clinical outcomes. Study objectives were to (1) determine the prevalence and type of sepsis-related neuroimaging abnormalities evident on clinically indicated brain MRI in children with sepsis and (2) test the association of these abnormalities with mortality, new disability, length of stay (LOS), and MRI indication. METHODS: Retrospective cohort study of 140 pediatric patients with sepsis and a clinically indicated brain MRI obtained within 60 days of sepsis onset at a single, large academic pediatric intensive care unit (PICU). Two radiologists systematically reviewed the first post-sepsis brain MRI and determined which abnormalities were sepsis-related. Outcomes compared in patients with versus without sepsis-related MRI abnormalities. RESULTS: PICU mortality was 7%. Thirty patients had one or more sepsis-related MRI abnormality, yielding a prevalence of 21% (95% confidence interval 15%, 28%). Among those, 53% (16 of 30) had sepsis-related white matter signal abnormalities; 53% (16 of 30) sepsis-related ischemia, infarction, or thrombosis; and 27% (eight of 30) sepsis-related posterior reversible encephalopathy. Patients with one or more sepsis-related MRI abnormality had increased mortality (17% vs 5%; P = 0.04), new neurological disability at PICU discharge (32% vs 11%; P = 0.03), and longer PICU LOS (median 18 vs 11 days; P = 0.04) compared with patients without. CONCLUSIONS: In children with sepsis and a clinically indicated brain MRI, 21% had a sepsis-related MRI abnormality. Sepsis-related MRI abnormalities were associated with increased mortality, new neurological disability, and longer PICU LOS.
Subject(s)
Brain Diseases/diagnostic imaging , Brain Diseases/epidemiology , Magnetic Resonance Imaging , Sepsis/complications , Sepsis/diagnostic imaging , Adolescent , Age Factors , Child , Child, Preschool , Female , Humans , Infant , Length of Stay , Male , Neuroimaging , Predictive Value of Tests , Retrospective Studies , Sepsis/mortality , Survival RateABSTRACT
OBJECTIVES: The primary objective was to determine the prevalence and characteristics associated with malpositioned temporary, nontunneled central venous catheters (CVCs) placed via the internal jugular (IJ) and subclavian (SC) veins in pediatric patients. DESIGN: Single-center retrospective cohort study. SETTING: Quaternary academic PICU. PATIENTS: Children greater than 1 month to less than 18 years who had a CVC placed between January 2014 and December 2018. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The primary outcome was the CVC tip position located on the first postprocedural radiograph. CVC tip was defined as follows: "recommended" (tip location between the carina and two vertebral bodies inferior to the carina), "high" (tip location between one and four vertebral bodies superior to the carina), "low" (tip position three or more vertebral bodies inferior to the carina), and "other" (tip grossly malpositioned). Seven hundred eighty-one CVCs were included: 481 (61.6%) were in "recommended" position, 157 (20.1%) were "high," 131 (16.8%) were "low," and 12 (1.5%) were "other." Multiple multinomial regression (referenced to "recommended" position) showed that left-sided catheters (adjusted odds ratio [aOR], 2.00, 95% CI 1.17-3.40) were associated with "high" CVC tip positions, whereas weight greater than or equal to 40 kg had decreased odds of having a "high" CVC tip compared with the reference (aOR, 0.45; 95% CI, 0.24-0.83). Further, weight category 20-40 kg (aOR, 2.42; 95% CI, 1.38-4.23) and females (aOR, 1.51; 95% CI, 1.01-2.26) were associated with "low" CVC tip positions. There was no difference in rates of central line-associated blood stream infection, venous thromboembolism, or tissue plasminogen activator usage or dose between the CVCs with tips outside and those within the recommended location. CONCLUSIONS: The prevalence of IJ and SC CVC tips outside of the recommended location was high. Left-sided catheters, patient weight, and sex were associated with malposition. Malpositioned catheters were not associated with increased harm.
Subject(s)
Catheterization, Central Venous , Central Venous Catheters , Catheterization, Central Venous/adverse effects , Central Venous Catheters/adverse effects , Child , Female , Humans , Retrospective Studies , Subclavian Vein , Tissue Plasminogen ActivatorABSTRACT
AIMS: To determine the prevalence of pulmonary hypertension (PH) among children with in-hospital cardiac arrest (IHCA) and its association with survival. METHODS: Children (<18 years) admitted to ICUs participating in the Virtual Pediatric Systems multicenter registry between January 2011 and December 2017 who had an IHCA during their hospitalization were included. Patients were classified by whether they had a documented diagnosis of PH at the time of IHCA. Clinical characteristics were compared between patients with and without PH. After propensity score matching, conditional logistic regression within the matched cohort determined the association between PH and survival to hospital discharge. RESULTS: Of 18,575 children with IHCA during the study period, 1,590 (8.6%) had a pre-arrest diagnosis of PH. Patients with PH were more likely to be 29 days to 2 years of age, female, Black/African American, and American Indian/Alaskan Native, and to be treated in a cardiac ICU or mixed PICU/cardiac ICU. At ICU admission, PH patients had a lower probability of death as determined by the Pediatric Index of Mortality 2 (PIM-2) score. Patients with PH were more likely to be receiving inhaled nitric oxide (13.0% vs. 2.1%; p < 0.001). Propensity score matching successfully matched 1,302 PH patients with 3,604 non-PH patients. Patients with PH were less likely to survive to hospital discharge (aOR 0.83; 95% CI: 0.72-0.95; p = 0.01) than non-PH patients. CONCLUSIONS: In this large multicenter study, 8.6% of children with IHCA had pre-existing documented PH. These children were less likely to survive to hospital discharge than those without PH.
Subject(s)
Cardiopulmonary Resuscitation , Heart Arrest , Hypertension, Pulmonary , Child , Female , Heart Arrest/epidemiology , Heart Arrest/therapy , Hospitals , Humans , Hypertension, Pulmonary/epidemiology , Patient DischargeABSTRACT
INTRODUCTION: In many centers, children with diabetic ketoacidosis (DKA) receive care either in an endocrinology ward or a pediatric intensive care unit (PICU). We conducted a quality improvement (QI) initiative to reduce potentially avoidable PICU admissions of children with DKA without increasing endocrinology ward-to-PICU transfers. METHODS: A survey of providers demonstrated opportunities to increase awareness of institutional criteria for PICU admissions of children with DKA. We created an electronic health record (EHR) dot-phrase, prepopulated with these criteria, and placed a note in the EHR for all patients with DKA as a reference for all providers. An EHR-based data report was created to monitor the disposition of DKA patients and the use of the dot-phrase (process measure). The primary outcome measure was the potentially avoidable PICU admissions for patients with DKA. Endocrinology ward-to-PICU transfers were tracked as a balancing measure to ensure safe disposition. RESULTS: After the implementation of the dot-phrase, use was variable, but averaged 33.4% over 1 year. The percentage of DKA admissions classified as potentially avoidable PICU stays decreased from 4.1% to 0.5%, with a concurrent decrease in the total percentage of PICU admissions for DKA from 19.1% to 8.4%. The percentage of endocrinology ward-to-PICU transfers also declined from 0.8% to 0%. CONCLUSIONS: A novel EHR-based intervention increasing awareness and documentation of established pediatric DKA management guidelines can be used to safely reduce PICU admissions for DKA without increasing the rate of endocrinology ward-to-intensive care unit transfers.
ABSTRACT
OBJECTIVES: To assess the prevalence of immunocompromised diagnoses among children with severe sepsis and septic shock, and to determine the association between immunocompromised diagnoses and clinical outcomes after adjustment for demographics and illness severity. DESIGN: Retrospective multicenter cohort study. SETTING: Eighty-three centers in the Virtual Pediatric Systems database. PATIENTS: Children with severe sepsis or septic shock admitted to a participating PICU between January 1, 2012, and December 31, 2016. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Across 83 centers, we identified 10,768 PICU admissions with an International Classification of Diseases, 9th Revision, Clinical Modification code for severe sepsis or septic shock; 3,021 of these patients (28%) had an immunocompromised diagnosis. To evaluate variation across centers and determine factors associated with PICU mortality, we used mixed-effect logistic regression models. Among patients without hematopoietic cell transplant, congenital immunodeficiency (adjusted odds ratio, 1.90; 95% CI, 1.24-2.92), multiple prior malignancies (adjusted odds ratio, 1.86; 95% CI, 1.15-2.99), and hemophagocytic lymphohistiocytosis (adjusted odds ratio, 3.09; 95% CI, 1.91-4.98) were associated with an increased odds of PICU mortality. Among patients with prior hematopoietic cell transplant, liquid malignancy (adjusted odds ratio, 3.15; 95% CI, 2.09-4.74), congenital immunodeficiency (adjusted odds ratio, 6.94; 95% CI, 3.84-12.53), multiple prior malignancies (adjusted odds ratio, 3.54; 95% CI, 1.80-6.95), and hemophagocytic lymphohistiocytosis (adjusted odds ratio, 2.79; 95% CI, 1.36-5.71) were associated with an increased odds of PICU mortality. PICU mortality varied significantly by center, and a higher mean number of sepsis patients per month in a center was associated with lower PICU mortality (adjusted odds ratio, 0.94; 95% CI, 0.90-0.98). PICU resource utilization varied by immunocompromised diagnosis and history of hematopoietic cell transplant, and among survivors immunocompromised patients have shorter median PICU length of stay compared with patients without immunocompromised diagnoses (p < 0.001). CONCLUSIONS: Immunocompromised diagnoses are present in 28% of children with severe sepsis or septic shock. Multiple prior malignancies, hemophagocytic lymphohistiocytosis, congenital immunodeficiency, and hematopoietic cell transplant are independently associated with an increased odds of PICU mortality in children with severe sepsis or septic shock. Significant variation exists in PICU mortality among centers despite adjustment for immunocompromised diagnoses, known risk factors for sepsis-related mortality, and center-level sepsis volume.
Subject(s)
Immunocompromised Host , Sepsis/mortality , Shock, Septic/mortality , Adolescent , Child , Child, Preschool , Female , Humans , Infant , Infant, Newborn , Intensive Care Units, Pediatric , Male , Retrospective Studies , Risk FactorsABSTRACT
OBJECTIVES: To evaluate the translation of a paper high-risk checklist for PICU patients at risk of clinical deterioration to an automated clinical decision support tool. DESIGN: Retrospective, observational cohort study of an automated clinical decision support tool, the PICU Warning Tool, adapted from a paper checklist to predict clinical deterioration events in PICU patients within 24 hours. SETTING: Two quaternary care medical-surgical PICUs-The Children's Hospital of Philadelphia and Cincinnati Children's Hospital Medical Center. PATIENTS: The study included all patients admitted from July 1, 2014, to June 30, 2015, the year prior to the initiation of any focused situational awareness work at either institution. INTERVENTIONS: We replicated the predictions of the real-time PICU Warning Tool by retrospectively querying the institutional data warehouse to identify all patients that would have flagged as high-risk by the PICU Warning Tool for their index deterioration. MEASUREMENTS AND MAIN RESULTS: The primary exposure of interest was determination of high-risk status during PICU admission via the PICU Warning Tool. The primary outcome of interest was clinical deterioration event within 24 hours of a positive screen. The date and time of the deterioration event was used as the index time point. We evaluated the sensitivity, specificity, positive predictive value, and negative predictive value of the performance of the PICU Warning Tool. There were 6,233 patients evaluated with 233 clinical deterioration events experienced by 154 individual patients. The positive predictive value of the PICU Warning Tool was 7.1% with a number needed to screen of 14 patients for each index clinical deterioration event. The most predictive of the individual criteria were elevated lactic acidosis, high mean airway pressure, and profound acidosis. CONCLUSIONS: Performance of a clinical decision support translation of a paper-based tool showed inferior test characteristics. Improved feasibility of identification of high-risk patients using automated tools must be balanced with performance.
Subject(s)
Clinical Deterioration , Decision Support Systems, Clinical , Heart Arrest/epidemiology , Intensive Care Units, Pediatric , Cardiopulmonary Resuscitation/statistics & numerical data , Checklist , Child , Electronic Health Records , Heart Arrest/diagnosis , Hospital Rapid Response Team/statistics & numerical data , Hospitalization , Humans , Retrospective Studies , Risk Factors , Sensitivity and SpecificityABSTRACT
Importance: Guidelines for declaration of brain death in children were revised in 2011 by the Society of Critical Care Medicine, American Academy of Pediatrics, and Child Neurology Society. Despite widespread medical, legal, and ethical acceptance, ongoing controversies exist with regard to the concept of brain death and the procedures for its determination. Objectives: To determine the epidemiology and clinical characteristics of pediatric patients declared brain dead in the United States. Design, Setting, and Participants: This study involved the abstraction of all patient deaths from the Virtual Pediatric Systems national multicenter database between January 1, 2012, and June 30, 2017. All patients who died in pediatric intensive care units (PICUs) were included. Main Outcomes and Measures: Patient demographics, preillness developmental status, severity of illness, cause of death, PICU medical and physical length of stay, and organ donation status, as well as comparison between patients who were declared brain dead vs those who sustained cardiovascular or cardiopulmonary death. Results: Of the 15â¯344 patients who died, 3170 (20.7%) were declared brain dead; 1861 of these patients (58.7%) were male, and 1401 (44.2%) were between 2 and 12 years of age. There was a linear association between PICU size and number of patients declared brain dead per year, with an increase of 4.27 patients (95% CI, 3.46-5.08) per 1000-patient increase in discharges (P < .001). The median (interquartile range) of patients declared brain dead per year ranged from 1 (0-3) in smaller PICUs (defined as those with <500 discharges per year) to 10 (7-15) for larger PICUs (those with 2000-4000 discharges per year). The most common causative mechanisms of brain death were hypoxic-ischemic injury owing to cardiac arrest (1672 of 3170 [52.7%]), shock and/or respiratory arrest without cardiac arrest (399 of 3170 [12.6%]), and traumatic brain injury (634 of 3170 [20.0%]). Most patients declared brain dead (681 of 807 [84.4%]) did not have preexisting neurological dysfunction. Patients who were organ donors (1568 of 3144 [49.9%]) remained in the PICU longer after declaration of brain death compared with those who were not donors (median [interquartile range], 29 [6-41] hours vs 4 [1-8] hours; P < .001). Conclusions and Relevance: Brain death occurred in one-fifth of PICU deaths. Most children declared brain dead had no preexisting neurological dysfunction and had an acute hypoxic-ischemic or traumatic brain injury. Brain death determinations are infrequent, even in large PICUs, emphasizing the importance of ongoing education for medical professionals and standardization of protocols to ensure diagnostic accuracy and consistency.
Subject(s)
Brain Death , Intensive Care Units, Pediatric/statistics & numerical data , Adolescent , Brain Death/diagnosis , Brain Injuries, Traumatic/mortality , Cause of Death , Child , Child, Preschool , Databases, Factual , Female , Heart Arrest/mortality , Hospital Mortality , Humans , Hypoxia-Ischemia, Brain/mortality , Infant , Length of Stay/statistics & numerical data , Male , Tissue Donors/statistics & numerical data , United States/epidemiologyABSTRACT
OBJECTIVES: To compare the performance of three methods of identifying children with severe sepsis and septic shock from the Virtual Pediatric Systems database to prospective screening using consensus criteria. DESIGN: Observational cohort study. SETTING: Single-center PICU. PATIENTS: Children admitted to the PICU in the period between March 1, 2012, and March 31, 2014. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: During the study period, all PICU patients were prospectively screened daily for sepsis, and those meeting consensus criteria for severe sepsis or septic shock on manual chart review were entered into the sepsis registry. Of 7,459 patients admitted to the PICU during the study period, 401 met consensus criteria for severe sepsis or septic shock (reference standard cohort). Within Virtual Pediatric Systems, patients identified using "Martin" (n = 970; κ = 0.43; positive predictive value = 34%; F1 = 0.48) and "Angus" International Classification of Diseases, 9th Edition, Clinical Modification codes (n = 1387; κ = 0.28; positive predictive value = 22%; F1 = 0.34) showed limited agreement with the reference standard cohort. By comparison, explicit International Classification of Diseases, 9th Edition, Clinical Modification codes for severe sepsis (995.92) and septic shock (785.52) identified a smaller, more accurate cohort of children (n = 515; κ = 0.61; positive predictive value = 57%; F1 = 0.64). PICU mortality was 8% in the reference standard cohort and the cohort identified by explicit codes; age, illness severity scores, and resource utilization did not differ between groups. Analysis of discrepancies between the reference standard and Virtual Pediatric Systems explicit codes revealed that prospective screening missed 66 patients with severe sepsis or septic shock. After including these patients in the reference standard cohort as an exploratory analysis, agreement between the cohort of patients identified by Virtual Pediatric Systems explicit codes and the reference standard cohort improved (κ = 0.73; positive predictive value = 70%; F1 = 0.75). CONCLUSIONS: Children with severe sepsis and septic shock are best identified in the Virtual Pediatric Systems database using explicit diagnosis codes for severe sepsis and septic shock. The accuracy of these codes and level of clinical detail available in the Virtual Pediatric Systems database allow for sophisticated epidemiologic studies of pediatric severe sepsis and septic shock in this large, multicenter database.
Subject(s)
Databases as Topic , Sepsis/diagnosis , Shock, Septic/diagnosis , Adolescent , Child , Child, Preschool , Clinical Coding , Databases as Topic/statistics & numerical data , Female , Humans , Infant , Intensive Care Units/statistics & numerical data , Male , Registries/statistics & numerical data , User-Computer InterfaceABSTRACT
BACKGROUND AND OBJECTIVE: Pediatric in-hospital cardiac arrest most commonly occurs in the pediatric intensive care unit (PICU) and is frequently preceded by early warning signs of clinical deterioration. In this study, we describe the implementation and evaluation of criteria to identify high-risk patients from a paper-based checklist into a clinical decision support (CDS) tool in the electronic health record (EHR). MATERIALS AND METHODS: The validated paper-based tool was first adapted by PICU clinicians and clinical informaticians and then integrated into clinical workflow following best practices for CDS design. A vendor-based rule engine was utilized. Littenberg's assessment framework helped guide the overall evaluation. Preliminary testing took place in EHR development environments with more rigorous evaluation, testing, and feedback completed in the live production environment. To verify data quality of the CDS rule engine, a retrospective Structured Query Language (SQL) data query was also created. As a process metric, preparedness was measured in pre- and postimplementation surveys. RESULTS: The system was deployed, evaluating approximately 340 unique patients monthly across 4 clinical teams. The verification against retrospective SQL of 15-minute intervals over a 30-day period revealed no missing triggered intervals and demonstrated 99.3% concordance of positive triggers. Preparedness showed improvements across multiple domains to our a priori goal of 90%. CONCLUSION: We describe the successful adaptation and implementation of a real-time CDS tool to identify PICU patients at risk of deterioration. Prospective multicenter evaluation of the tool's effectiveness on clinical outcomes is necessary before broader implementation can be recommended.