Implementation of clinical practice changes in the PICU: a qualitative study using and refining the iPARIHS framework.

BACKGROUND: Like in many settings, implementation of evidence-based practices often fall short in pediatric intensive care units (PICU). Very few prior studies have applied implementation science frameworks to understand how best to improve practices in this unique environment. We used the relatively new integrated Promoting Action on Research Implementation in Health Services (iPARIHS) framework to assess practice improvement in the PICU and to explore the utility of the framework itself for that purpose. METHODS: We used the iPARIHS framework to guide development of a semi-structured interview tool to examine barriers, facilitators, and the process of change in the PICU. A framework approach to qualitative analysis, developed around iPARIHS constructs and subconstructs, helped identify patterns and themes in provider interviews. We assessed the utility of iPARIHS to inform PICU practice change. RESULTS: Fifty multi-professional providers working in 8 U.S. PICUs completed interviews. iPARIHS constructs shaped the development of a process model for change that consisted of phases that include planning, a decision to adopt change, implementation and facilitation, and sustainability; the PICU environment shaped each phase. Large, complex multi-professional teams, and high-stakes work at near-capacity impaired receptivity to change. While the unit leaders made decisions to pursue change, providers' willingness to accept change was based on the evidence for the change, and provider's experiences, beliefs, and capacity to integrate change into a demanding workflow. Limited analytic structures and resources frustrated attempts to monitor changes' impacts. Variable provider engagement, time allocated to work on changes, and limited collaboration impacted facilitation. iPARIHS constructs were useful in exploring implementation; however, we identified inter-relation of subconstructs, unique concepts not captured by the framework, and a need for subconstructs to further describe facilitation. CONCLUSIONS: The PICU environment significantly shaped the implementation. The described process model for implementation may be useful to guide efforts to integrate changes and select implementation strategies. iPARIHS was adequate to identify barriers and facilitators of change; however, further elaboration of subconstructs for facilitation would be helpful to operationalize the framework. TRIAL REGISTRATION: Not applicable, as no health care intervention was performed.

Factors Influencing Implementation of Blood Transfusion Recommendations in Pediatric Critical Care Units.

Purpose: Risks of red blood cell transfusion may outweigh benefits for many patients in Pediatric Intensive Care Units (PICUs). The Transfusion and Anemia eXpertise Initiative (TAXI) recommendations seek to limit unnecessary and potentially harmful transfusions, but use has been variable. We sought to identify barriers and facilitators to using the TAXI recommendations to inform implementation efforts. Materials and Methods: The integrated Promoting Action on Research Implementation in Health Services (iPARIHS) framework guided semi-structured interviews conducted in 8 U.S. ICUs; 50 providers in multiple ICU roles completed interviews. Adapted Framework analysis, a form of content analysis, used the iPARIHS innovation, recipient, context and facilitation constructs and subconstructs to categorize data and identify patterns as well as unique informative statements. Results: Providers perceived that the TAXI recommendations would reduce transfusion rates and practice variability, but adoption faced challenges posed by attitudes around transfusion and care in busy and complex units. Development of widespread buy-in and inclusion in implementation, integration into workflow, designating committed champions, and monitoring outcomes data were expected to enhance implementation. Conclusions: Targeted activities to create buy-in, educate, and plan for use are necessary for TAXI implementation. Recognition of contextual challenges posed by the PICU environment and an approach that adjusts for barriers may optimize adoption.

Vasoplegia after pediatric cardiac transplantation in patients supported with a continuous flow ventricular assist device.

OBJECTIVE: To determine the association between continuous flow ventricular assist devices and the incidence of vasoplegia following orthotopic heart transplant in children. Moreover, to propose a novel clinical definition of vasoplegia for use in pediatric populations. METHODS: This is a single-center, retrospective cohort study set in the cardiovascular intensive care unit of a tertiary children's hospital. All patients aged 3 years and older who underwent orthotopic heart transplant at Stanford University between April 1, 2014, and July 31, 2017, were included. Vasoplegia was defined by the use of vasoconstrictive medication, diastolic hypotension, preserved systolic heart function, and absence of infection or right atrial pressure or central venous pressure <5 mm Hg. RESULTS: Of 44 eligible patients, 21 were supported using a continuous flow ventricular assist device. Following heart transplant, 14 patients (32%) developed vasoplegia by the study definition. Development of vasoplegia was associated with pretransplant use of a continuous flow ventricular assist device (52% vs 13%) with a relative risk of 4.02 (95% confidence interval, 1.30-12.45; P = .009). No other variables were predictive of vasoplegia in univariable analysis. Presence of vasoplegia was not associated with adverse outcomes, although there were trends towards higher incidence of acute kidney injury and increased length of hospital stays. CONCLUSIONS: Children receiving continuous flow ventricular assist device support are at increased risk for vasoplegia following orthotopic heart transplant, using a novel definition of vasoplegia. Anticipation of this complication will allow for prompt intervention, thereby minimizing hemodynamic instability and impact on patient outcomes.

Factors Associated With Mortality in Neonates Requiring Extracorporeal Membrane Oxygenation for Cardiac Indications: Analysis of the Extracorporeal Life Support Organization Registry Data.

OBJECTIVES: Survival among neonates supported with extracorporeal membrane oxygenation for cardiac indications is 39%. Previous single-center studies have identified factors associated with mortality, but a comprehensive multivariate analysis is not available for this population. Understanding factors associated with mortality may help design treatment strategies, determine optimal timing for cannulation, and inform patient selection. This study identifies factors associated with mortality in neonates supported with extracorporeal membrane oxygenation for cardiac indications. DESIGN: Retrospective cohort study. SETTING: Two hundred and thirty U.S. and international centers reporting extracorporeal membrane oxygenation data to the Extracorporeal Life Support Organization. SUBJECTS: Four thousand and four seventy one neonates with congenital and acquired cardiac disease supported with extracorporeal membrane oxygenation for cardiac indications during 2001-2011. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: The primary outcome measure was mortality prior to hospital discharge. Overall hospital mortality was 59%. Demographic and preextracorporeal membrane oxygenation factors associated with mortality were evaluated in a multivariable model. Factors associated with death prior to hospital discharge included lower body weight, earlier era, single ventricle physiology, lower preextracorporeal membrane oxygenation arterial pH, and longer time from intubation to extracorporeal membrane oxygenation cannulation. Lower pH was associated with increased mortality regardless of cardiac diagnosis and surgical complexity. The majority of survivors separated from extracorporeal membrane oxygenation less than 8 days after extracorporeal membrane oxygenation deployment. CONCLUSIONS: Mortality for neonates supported with extracorporeal membrane oxygenation for cardiac indications is high. Severity of preextracorporeal membrane oxygenation acidosis was independently associated with increased risk of mortality. Earlier initiation of extracorporeal membrane oxygenation may reduce the degree and duration of acidosis and may improve survival. Further studies are needed to determine optimal timing of cannulation in this population.

Association of graft ischemic time with survival after heart transplant among children in the United States.

BACKGROUND: Previous studies have found no association between graft ischemic time (IT) and survival in pediatric heart transplant (HTx) recipients. However, previous studies were small or analyzed risk only at the extremes of IT, where observations are few. We sought to determine whether graft IT is independently associated with graft survival in a large cohort of children with no a priori assumptions about where the risk threshold may lie. METHODS: All children aged <18 years in the U.S. undergoing primary HTx (1987 to 2008) were included. The primary end point was graft loss (death or retransplant) within 6 months. Multivariate analysis was performed to analyze the association between graft IT and graft loss within 6 months after transplant. A secondary end point of longer-term graft loss was assessed among recipients who survived the first 6 months after transplant. RESULTS: Of 4,716 pediatric HTxs performed, the median IT was 3.5 hours (interquartile range, 2.7-4.3 hours). Adjusted analysis showed that children with an IT > 3.5 hours were at increased risk of graft loss within 6 months after transplant (hazard ratio, 1.3; 95% confidence interval, 1.1-1.5; p = 0.002). Among 6-month survivors, IT was not associated with longer-term graft loss. CONCLUSIONS: IT beyond 3.5 hours is associated with a 30% increase in risk of graft loss within 6 months in pediatric HT recipients. Although the magnitude of risk associated with IT is small compared with the risk associated with recipient factors, these findings may be important during donor assessment for high-risk transplant candidates.