Funded by: National Institute of General Medical Sciences (NIGMS)

CRISMA Project Team: Derek Angus, PI; John Kellum; David Huang, Lisa Weissfeld, and Lan Kong, Co-Is; Tammy Young, Project Manager; Jodi Gigler, Project Assistant; Melinda Carter, Laboratory Manager; Xia Tang, Programmer.

Severe sepsis is the syndrome of acute organ dysfunction secondary to infection. It affects 750,000 Americans each year, with a mortality of 30%. Despite considerable understanding of the pathophysiology of sepsis, current efforts to improve care are hampered by limited empiric data regarding the amount and timing of sepsis therapies. We have amassed for this Center proposal a multidisciplinary group of investigators and consortium of leading institutions. Our goal is to address the overarching hypothesis that there are ‘golden hours’ in the initial management of sepsis and septic shock where prompt, rigorous, standardized care can reduce unwanted downstream consequences and improve clinical outcomes.

Efforts will capitalize on the findings of a recent ‘proof-of-concept’ trial by Rivers et al. which demonstrated in a single center randomized trial that 6h of protocolized resuscitation for subjects presenting to the Emergency Department (ED) with early septic shock dramatically improved mortality when compared to usual care. While this study was revolutionary, it left unanswered whether the findings were generalizable and whether all elements of the protocol are necessary, especially the use of central venous catheterization and blood transfusion. The apparent success of the Rivers protocol also prompts questions about the mechanisms by which resuscitation techniques affect outcome. And, there are important pragmatic questions regarding the logistic and economic constraints to widespread implementation of protocolized resuscitation across all EDs in the US.

These questions will be addressed through execution of a large multicenter trial which will randomize 1950 subjects who present to the ED in septic shock to 3 arms (650/arm): the ‘Rivers’ protocol; a simpler, less invasive protocol (esophageal Doppler monitoring and no blood transfusion); and usual care. Protocols will be implemented using best evidence regarding guideline dissemination. We have organized our efforts under 3 integrated Subprojects. Subproject #1 (Clinical Efficacy) will conduct the trial and test whether protocolized care improves mortality compared to usual care and whether the Rivers protocol is superior to the simpler protocol. Subproject #2 (Mechanisms of Action) will measure concentrations over time of carefully selected circulating markers of four fundamental pathways implicated in sepsis-related organ dysfunction (cellular hypoxia, oxidative stress, inflammation, and coagulation/thrombosis) and test whether protocolized resuscitation reduces expression of these markers and whether the clinical efficacy of these protocols is associated with reduced expression of these markers. Subproject #3 (Costs and Cost-effectiveness) will measure the incremental costs and resource use of protocolized resuscitation and determine the value, or cost-effectiveness, of the alternative strategies.

This project will generate new, important, and comprehensive data on the clinical, biologic, and pragmatic aspects of standard, prompt, rigorous resuscitation for septic shock. Our findings will aid scientists, clinicians, families and policymakers and will immediately affect care of the critically ill. As the number of Americans dying with sepsis is similar to that of acute myocardial infarction, the proposed study has enormous implications for the public health of the country and is consistent with the recent NIH emphasis on translational research (nihroadmap.nih.gov).

Funded by: National Heart, Lung, and Blood Institute (NHLBI).

CRISMA Project Team: John Kellum, PI; Gilles Clermont, Lisa Weissfeld, and Lan Kong, Co-Is; Michele Elder, Project Manager; Melinda Carter. Laboratory Technician; Xia Tang, Programmer.

This project is a Bioengineering Research Partnership (BRP) put together to design and test an extracorporeal device for the treatment of severe sepsis, based on the principle of hemoadsorption. The multi-disciplinary team comprises basic science and clinical researchers, bioengineering and biomaterials experts, and experts in complex systems modeling. The project will carefully integrate this diverse group of researchers from both technical and biomedical fields, crossing the boundaries of multiple scientific disciplines and leveraging the capabilities of both industry and academia.

Severe sepsis (acute onset organ failure in the setting of infection) is a major health problem that kills nearly 250,000 Americans each year and costs billions of dollars. Available therapies for sepsis, including those recently approved, are suboptimal and new therapies are urgently needed. However, the complexities of the inflammatory response network and the high cost of clinical trials, particularly in the critically ill, renders the traditional drug/device development paradigm obsolete. We have previously developed and tested an extracorporeal blood purification device for treatment in chronic renal disease and have adapted this device for the treatment of acute inflammatory diseases. We have also developed and partially calibrated—in both rodents and humans—a mathematical model of sepsis. This project proposes to integrate these two achievements and, through an iterative design process, develop a device that can be used to treat severe sepsis.

The goal of this proposal is to design an extracorporeal blood purification device for the treatment of severe sepsis. This proposal will bring together investigators from the following departments at the University of Pittsburgh: Bioengineering, Chemical Engineering, Critical Care Medicine, Surgery, Medicine, and Mathematics. We will also call on the expertise of two companies, one specializing in adsorbent polymer technology (MedaSorb Technologies, LLC) and the other in complex systems modeling (Immunetrics, Inc.).

Funded by: National Institute of Diabetes & Digestive & Kidney Diseases (NIDDK)

CRISMA Project Team: John Kellum, PI; Ramesh Venkataraman, Derek Angus, and Lisa Weissfeld, Co-Is; Michele Elder, Project Manager; Melinda Carter. Laboratory Technician; Xia Tang, Programmer.

BioMaRK is an ancillary study to the ARF Trial Network (ATN) Study (CSP 530), a 27-center randomized controlled trial jointly funded by the Department of Veterans’ Affairs, Cooperative Studies Program and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). While ATN seeks to determine whether intensive RRT or standard therapy is superior in terms of 60-day mortality, BioMaRK seeks to better understand why.

The central goal of this study is to better understand the role of inflammation and oxidative stress in the recovery from ARF—both in terms of survival and recovery of renal function. To do so, we will adjust for clinical characteristics including underlying severity of illness, concurrent conditions including sepsis, and treatment variables. Finally, in keeping with the NIH roadmap, in order to understand clinical utility of this work we will build a clinical risk prediction model that will consider plasma mediator levels, urine markers, and clinical variables.

Funded by: Agency for Healthcare Research and Quality (AHRQ)/National Heart, Lung, and Blood Institute (NHLBI)

CRISMA Project Team: Derek Angus, PI; Gilles Clermont, Lisa Weissfeld, and Lan Kong, Co-Is; Michael Coughlin, Project Manager; Tony Dremsizov, Senior Research Assistant; Kim Fusko, Patient Interviewer; Xia Tang, Programmer.

The pulmonary artery catheter (PAC) is a commonly used device that provides hemodynamic data to guide care of the critically ill, such as patients with acute respiratory distress syndrome or acute lung injury (ARDS/ALI). Clinicians believe PAC use improves decision-making and patient outcomes but evidence is lacking and recent data suggest the PAC may increase mortality as well as considerably increase costs. In response, the NHLBI funded a large multicenter trial (Fluid And Catheter Treatment Trial (FACTT)) in which ARDS/ALI patients will be randomized to receive a PAC or the less invasive central venous catheter (CVC) and to receive a liberal or conservative fluid management protocol in response to data provided by the PAC or CVC. The primary end-point of FACTT will be in-patient mortality.

We are complementing FACTT with a concurrent economic analysis of the PAC. Our aims are to: 1.) compare differences between study arms in long-term survival, quality of life, and quality-adjusted survival; 2.) compare differences between study arms in acute care and long-term costs; 3.) calculate the cost-efficacy of PAC use (i.e., the balance of costs and effects under the controlled environment of the FACTT trial), and 4.) estimate cost-effectiveness under more real-world conditions and produce life-time cost-effectiveness ratios, thereby facilitating comparison of our results to other cost-effectiveness analyses. The results of our adjunct to FACTT will substantially amplify the value of the data being collected and provide, for the first time, robust estimates from randomized data of the economic effects of the widespread application of this important technology.

Funded by: National Institute of General Medical Sciences (NIGMS)

CRISMA Project Team: Derek Angus, PI; John Kellum Co-PI; Lisa Weissfeld, and Lan Kong, Co-Is; Tammy Young, Project Manager; Jodi Gigler, Project Assistant; Melinda Carter, Laboratory Manager; Xia Tang, Programmer.

Over 450,000 cases of sepsis occur per year in the US and frequently progress to organ dysfunction and death. Although experimental studies using cells and animals have greatly improved our understanding of the pathophysiology of sepsis, there remains a remarkable paucity of biochemical and genetic data regarding the natural history of this important public health problem.

Because pneumonia is the most common cause of sepsis, patients with this infection represent an excellent clinical model for studying sepsis in a relatively homogeneous population. We have enrolled a large cohort of patients (n=2,323) with community-acquired pneumonia (CAP). In addition to collecting detailed clinical data, we have performed genetic analyses, focusing on allelic variations in the coding or noncoding regions of genes whose products are important in the expression and/or regulation of the inflammatory response. We have also obtained measurements over time of the plasma concentrations or cell surface expression of several key inflammatory molecules.

We will determine the influence of specific polymorphisms on the development, course and outcome of pneumonia and sepsis. We will also test whether genetic predisposition to an exuberant inflammatory response protects against infection yet also increases risk for adverse systemic effects and outcome. We will also compare genetic data from patients with data obtained from a cohort of healthy controls (n=300). We are testing several existing hypotheses regarding the association of circulating inflammatory molecules and outcome.

This study is generating: new and valuable information regarding existing lines of inquiry and laboratory investigation; new hypotheses arising from the use of time-dependent modeling, and new clinical decision tools that have immediate and practical value for designing clinical trials and improving patient care.

Funded by: National Heart, Lung, and Blood Institute (NHLBI) and INO Therapeutics

CRISMA Project Team: Derek Angus, PI; R. Scott Watson, Gilles Clermont, Lisa Weissfeld, and Lan Kong, Co-Is; Michael Coughlin, Project Manager; Tony Dremsizov, Senior Research Assistant; Carrie Lushington, Patient Interviewer; Xia Tang, Programmer.

PrONOx has been funded by NHLBI to study the long-term, multi-dimensional outcomes of premature babies enrolled in a randomized controlled trial (RCT) of inhaled nitric oxide (iNO) to prevent chronic lung disease. Because iNO (a selective pulmonary vasodilator that improves gas exchange) also has lung-specific anti-inflammatory properties, NHLBI funded a large, multi-center RCT to assess the effect of iNO in 800 pre-term newborns with respiratory failure. We are extending and enhancing the follow-up of the RCT by following subjects for an average of 4½ years to evaluate survival, developmental, family and economic effects of iNO and contemporary medical care for premature babies. Specifically, we are assessing: 1.) long-term clinical and childhood developmental outcomes, 2.) family effects, 3.) healthcare costs of prematurity-associated respiratory failure, and 4.) the cost-effectiveness of iNO. Our follow-up program consists of structured telephone interviews every 3 to 6 months, yearly clinic visits at sites for comprehensive developmental evaluation, and analysis of detailed hospital bills.

This study is generating contemporary data on the long-term consequences of prematurity-associated respiratory failure and the effects of iNO on babies and their families. It will also aid in the design of future trials by providing information about the appropriate duration of follow-up and the robustness of surrogate endpoints.

Funded by: National Institute of General Medical Sciences (NIGMS)

CRISMA Project Team: Gilles Clermont, PI.

The initial response of the body to acute biological stress such as bacterial infection or tissue trauma is an acute inflammatory response. This response involves a cascade of events mediated by a large array of cells and molecules that locate invading pathogens or damaged tissue, alert and recruit other cells and molecules, eliminate the offending agents, and finally restore the body to equilibrium. Although much has been learned in the last several years on the complex cellular and molecular mechanisms of the acute inflammatory response, the association between these mechanisms and patient outcome has remained elusive.

We hypothesize that a major reason an effective treatment strategy has not been developed is that a good understanding of the global behavior of the acute inflammatory response is lacking. We propose to address this shortcoming by combining mathematical modeling of the acute inflammatory response with controlled companion experiments that are designed to validate and further refine the models. We envision a synergistic environment, in which modeling generates new hypotheses for experiments to test and the experiments generate new phenomena for the model to explain. Given the extreme complexity of the acute inflammatory response, we recognize that it is impossible to develop a completely faithful model. However, we feel that it is possible to produce a hierarchy of models of varying levels of biological detail that are each designed to address a limited set of questions. We will follow a modeling strategy that has been successful in computational neuroscience - namely combining analysis of reduced models with numerical simulations of more complicated and biologically plausible models.

We intend to achieve our objectives through the following specific aims:

Aim 1: Develop a hierarchy of mathematical models of the acute inflammatory response, each designed to capture a set of salient aspects of this response.

Aim 2: Refine and validate the mathematical models through an iterative process of experimentation, statistical analysis, and model development.

Aim 3: Analyze the various modes of behavior in the mathematical models and use these modes to make predictions of outcomes in different experimental scenarios.

Funded by: National Institute of Child Health and Human Development (NICHD)

CRISMA Project Team: Scott Watson, PI; Derek Angus, Mentor; Kim Fusko, Project Assistant.

A quarter of a million children aged 1 to 19 years receive care in an intensive care unit (ICU) in the US each year. Although most survive, the survivors are at risk of impairment in multiple domains. The need for additional research evaluating long-term outcomes and functional status after critical illness in children was highlighted at a recent NIH sponsored conference.

The goals of this proposal are to determine the frequency and extent of patient morbidity, the impact of illness on families, and risk factors for adverse sequelae following critical illness in children. A stratified prospective cohort of 300 previously healthy children who survive critical illness in the Children's Hospital of Pittsburgh Pediatric ICU will be recruited and followed for 1 to 4 years post-discharge. Three primary domains will be assessed: 1) Functional status and quality of life, 2) Neuropsychologic sequelae, and 3) Family effects. Patients and their families will be evaluated in the hospital and followed from 1 to 4 years, with evaluations 1, 6, and 12 months post-discharge and yearly thereafter. Teachers of school-aged children will also be contacted three months post-discharge. This study will have implications for future interventional trials and larger observational studies that may improve the care and outcomes of children with critical illness.

Funded by: National Institute on Aging (NIA)

CRISMA Project Team: Amber Barnato, PI; Derek Angus, Principal Mentor.

The overall goal of this project is to describe variation in hospital-level treatment intensity at the end of life in Pennsylvania, to identify the hospital characteristics associated with this variation, and to determine the effect of this variation on patient survival and inpatient costs. Aggregate hospital-referral region measures elucidate geographic variations in practice patterns, but intra-regional variation can in some instances exceed inter-regional variation. Hospital-level analyses may provide a more meaningful level of granularity because a hospital – or even a service within a hospital – may be the unit within which a practice culture emerges and persists. This project is designed to bridge this gap by studying hospital-level variations in the use of intensive services at the end of life in Pennsylvania hospitals. Hospital-level variation, by definition, aggregates to regional-level variation and hospitals are a major place where decisions are made by administrators, physicians, and patients about the use of intensive services. Studying hospitals in a single state with uniform regulatory and financing structures will help minimize sources of variation. Importantly, this study neither addresses patient preferences nor physician attitudes – clearly targets for further study – but it will provide information about the context (e.g., the culture) in which these more micro-level aspects of patient care and management are couched. The results of my work can be applied to future end-of-life quality-improvement activities in partnership with the Pennsylvania Health Care Cost Containment Council (PHC4).

Funded by: National Heart, Lung, and Blood Institute (NHLBI)

CRISMA Project Team: Eric B. Milbrandt, PI; Derek Angus, Lisa Weissfeld, Mentors.

Delirium is a very common and serious complication of mechanical ventilation that is associated with increased duration of mechanical ventilation and ICU stay; increased medical complications, such as ventilator associated pneumonia; and poor patient outcomes, such as increased 6-month mortality and long-term cognitive deficits. While the poor outcomes associated with delirium have been defined, there is an incomplete understanding of how delirium should be managed as well as the pathophysiology that underlies it. No work has rigorously evaluated the current national guidelines that call for routine delirium screening and treatment with haloperidol, nor examined the contribution of brain injury and impaired central cholinergic transmission to the development of delirium. To address these issues, we are conducting a randomized controlled trial of haloperidol versus placebo in 304 delirious mechanically ventilated patients. This study is nested within a larger cohort study of delirium pathophysiology in 380 mechanically ventilated ICU patients. The two components of this study represent a translational approach to tackling this problem. This work has the potential to significantly advance the understanding of delirium and inform future clinicians in the proper approach to delirium management while also serving as preliminary work for future grant submissions.

Funded by: Agency for Healthcare Research and Quality (AHRQ)

CRISMA Project Team: Derek Angus, PI.

Major goal of this project: Development of a broadly based, multidisciplinary health services research training program designed to provide trainees with the skills necessary to meet the rapidly changing demands of health care systems today. The overall goal is to create a rigorous but flexible training program in health services research that can provide the appropriate knowledge, skills and experiences necessary for several different levels of participation and future career paths of the trainees.

Funded by: National Institute of Diabetes & Digestive & Kidney Diseases (NIDDK) and The Department of Veterans Affairs (VA)

CRISMA Project Team: John Kellum, Co-I and ATN Executive Committee Member; Ramesh Venkataraman, Co-I; Jill Aubrecht, Clinical Research Coordinator.

ATN is a multi-center, prospective, randomized, parallel-group trial of an intensive strategy versus conventional strategy of renal replacement therapy for the treatment of acute renal failure (ARF) secondary to acute tubular necrosis in critically ill patients. The primary hypothesis is that the intensive strategy will reduce 60-day all-cause mortality by 10% compared to the conventional strategy - i.e., a reduction from 55% in the conventional arm to 45% in the intensive arm. Secondary outcomes are 60-day in-hospital all-cause mortality, 1-year all-cause mortality, and recovery of renal function by day 28. The study will recruit 1164 patients over a period of 3 years and each patient will be actively followed for 60 days.

The ATN is a collaboration between nephrology and critical care medicine. At the University of Pittsburgh site, the ATN study is jointly conducted by the Department of Medicine, Division of Nephrology, and the Department of Critical Care Medicine, CRISMA Laboratory.

For more information see: http://www.clinicaltrials.gov/show/NCT00076219

Funded by: Health Resources and Services Administration (HRSA)

CRISMA Project Team: John Kellum, PI; Ramesh Venkataraman, Co-I; Michele Elder, Project Manager; Melinda Carter, Laboratory Manager.

Brain death is associated with an acute surge in inflammatory response. This response has been associated with pre-explantation organ dysfunction (PEOD) in the donor and acute organ dysfunction (AOD) post transplantation. The purpose of this project is to determine whether attenuation of this inflammatory response using a hemoadsorption device (CytoSorb) is safe and efficacious in reducing AOD and thereby improving organ recovery.

HIDONOR is a dual-site, 3-study, 3-year project. The protocol for the randomized controlled trial (RCT) will depend upon the findings from the DOSE and OCS studies.

STUDY A:  Dose-finding open-label study in patients not eligible for organ donation (HIDonOR-DOSE, now enrolling)

This study is a two-center randomized, open-label, dose-finding study in which 24 brain dead subjects deemed unsuitable for organ donation by respective organ procurement organizations (OPO) will be randomized to two treatment groups after obtaining consent from families.  All study subjects will be treated with hemoadsorption for 4 hours using CytoSorb. Subjects will be randomized to one of two cartridge sizes.  Fractional clearance of IL-6 and other mediators will be measured every hour during the treatment. The objectives are to determine which dose of CytoSorb (defined by cartridge size) is most effective in reducing the circulating concentrations of a number of key inflammatory molecules (TNF, IL-6, IL-10, S100-B) in potential organ donors, to determine if blood purification using CytoSorb can improve specific measures of organ function in potential organ donors, and to evaluate safety and feasibility of CytoSorb in the organ donor population.

STUDY B:  Observational cohort study (HIDonOR-OCS, enrollment completed)

This study is a two-center prospective observational study in which 30 brain dead organ donors were observed in the ICU to characterize their management and natural history.  After obtaining consent from families all study subjects were observed until they were transferred to the operating room for organ explantation. The objectives of this study were to determine baseline circulating concentrations of a number of key inflammatory molecules in potential organ donors, to characterize management and natural history of organ donors at our sites, and to evaluate baseline safety endpoints (sodium, glucose, complement, T3 and cortisol) in the organ donor population.

STUDY C:  Hemoadsorption to improve donor organ recovery randomized controlled trial (HIDonOR-RCT, enrollment anticipated to begin in September 2005)

This study is a two-center randomized, controlled study in which 60 brain dead organ donors will be randomized to two treatment groups.  After obtaining consent from families, study subjects will be randomized to either receive protocolized donor care (control arm) or protocolized donor care with hemoadsorption using CytoSorb (treatment arm). Hemoadsorption will be started after consent and will be continued for a specific time-frame. Values before and after hemoadsorption in the treatment arm will be compared to time-matched values (immediately after brain death and just prior to transfer to operating room for organ explantation) in the control arm. The objectives of this study are to determine if treatment with CytoSorb results in an increase in organ recovery compared to control arm, to determine if blood purification using CytoSorb can improve specific measures of organ function compared to control arm, and to evaluate safety and feasibility of CytoSorb in the organ donor population.

CRISMA Project Team: David Huang, PI; Derek Angus, Gilles Clermont, Lisa Weissfeld and Lan Kong, Mentors. Dr. Huang began this project as Fellow and is continuing it now as an Assistant Professor of Critical Care Medicine and Emergency Medicine.

Previous studies and experience suggest that superior Intensive Care Unit (ICU) collaboration may improve patient care and outcomes, as well as increase nurse job satisfaction and retention rates. Furthermore, the Society of Critical Care Medicine's official best practice model recommendation endorses a model wherein dedicated ICU personnel...all work as a team. However, the basic question Do safety attitudes independently affect ICU outcomes has never been definitively addressed. Answering this question is the primary objective of ICUTEAMS-USA. Secondary objectives will be to determine the variability of these safety attitudes within a sample of US ICUs and assessing the accuracy of ICU directors' assessments of their own ICUs' safety attitudes.

In collaboration with human factors experts at the University of Texas, we will administer a previously validated survey instrument, the SAQ-ICU (Safety Attitudes Questionnaire - ICU), to the ICUs of the Project Impact - CCM network. The SAQ-ICU measures six distinct safety attitudes, is based on a validated aviation industry survey instrument and has been applied to hundreds' of ICUs in three countries. Survey results will be linked to ICU outcomes from the Project Impact-CCM database to answer the primary research objective.

Funded by: Eisai Medical Research Inc and Eisai Ltd. (Eisai)

ACCESS: A Controlled Comparison of Eritoran Tetrasodium and Placebo in Patients with Severe Sepsis

CRISMA Project Team: Derek Angus, Eric Milbrandt, Co-Principal Investigators; Lisa Weissfeld, Lan Kong, Biostatisticians; Tammy Young, Project Manager; Tony Dremsizov, Project Associate; Xia Tang, David Del Grosso, Honest Brokers.

Eisai is conducting a phase III clinical trial (ACCESS) to understand the effect of eritoran tetrasodium (eritoran) on 28-day mortality in subjects with severe sepsis. CRISMA is overseeing the research segment relevant to the secondary endpoints of 12-month mortality, Pharmacoeconomic and quality-of-life effects of eritoran.

ACCESS is a randomized, double-blind, placebo-controlled study in which subjects with severe sepsis are assigned to either treatment or placebo. Approximately 2000 subjects will be recruited from ~200 sites in the US, Canada, European Union, South America, South Africa, Asia, and Oceania. Enrolment began in June 2006 and is targeted to conclude during or before July 2010.

CRISMA designed the Pharmacoeconomic and quality of life assessments for the trial. These include costs incurred during the stay at the study hospital and after discharge up to one year as well as quality-of-life assessments using the EuroQol instrument at 28 days, 3, 6 and 12 months. The follow-up will be conducted through telephone interviews. This and a petition to the National Death Index will yield 1-mortality data. A subanalysis of long-term outcomes and resource use in the subset of Medicare beneficiaries also is planned.