Sepsis Subclasses: A Framework for Development and Interpretation.

Sepsis is defined as a dysregulated host response to infection that leads to life-threatening acute organ dysfunction. It afflicts approximately 50 million people worldwide annually and is often deadly, even when evidence-based guidelines are applied promptly. Many randomized trials tested therapies for sepsis over the past 2 decades, but most have not proven beneficial. This may be because sepsis is a heterogeneous syndrome, characterized by a vast set of clinical and biologic features. Combinations of these features, however, may identify previously unrecognized groups, or "subclasses" with different risks of outcome and response to a given treatment. As efforts to identify sepsis subclasses become more common, many unanswered questions and challenges arise. These include: 1) the semantic underpinning of sepsis subclasses, 2) the conceptual goal of subclasses, 3) considerations about study design, data sources, and statistical methods, 4) the role of emerging data types, and 5) how to determine whether subclasses represent "truth." We discuss these challenges and present a framework for the broader study of sepsis subclasses. This framework is intended to aid in the understanding and interpretation of sepsis subclasses, provide a mechanism for explaining subclasses generated by different methodologic approaches, and guide clinicians in how to consider subclasses in bedside care.

Evaluation of Repeated Quick Sepsis-Related Organ Failure Assessment Measurements Among Patients With Suspected Infection.

OBJECTIVES: Among patients with suspected infection, a single measurement of the quick Sepsis-related Organ Failure Assessment has good predictive validity for sepsis, yet the increase in validity from repeated measurements is unknown. We sought to determine the incremental predictive validity for sepsis of repeated quick Sepsis-related Organ Failure Assessment measurements over 48 hours compared with the initial measurement. DESIGN: Retrospective cohort study. SETTING: Twelve hospitals in southwestern Pennsylvania in 2012. PATIENTS: All adult medical and surgical encounters in the emergency department, hospital ward, postanesthesia care unit, and ICU. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Among 1.3 million adult encounters, we identified those with a first episode of suspected infection. Using the maximum quick Sepsis-related Organ Failure Assessment score in each 6-hour epoch from onset of suspected infection until 48 hours later, we characterized repeated quick Sepsis-related Organ Failure Assessment with: 1) summary measures (e.g., mean over 48 hr), 2) crude trajectory groups, and 3) group-based trajectory modeling. We measured the predictive validity of repeated quick Sepsis-related Organ Failure Assessment using incremental changes in the area under the receiver operating characteristic curve for in-hospital mortality beyond that of baseline risk (age, sex, race/ethnicity, and comorbidity). Of 37,591 encounters with suspected infection, 1,769 (4.7%) died before discharge. Both the mean quick Sepsis-related Organ Failure Assessment at 48 hours (area under the receiver operating characteristic, 0.86 [95% CI, 0.85-0.86]) and crude trajectory groups (area under the receiver operating characteristic, 0.83 [95% CI, 0.83-0.83]) improved predictive validity compared with initial quick Sepsis-related Organ Failure Assessment (area under the receiver operating characteristic, 0.79 [95% CI, 0.78-0.80]) (p < 0.001 for both). Group-based trajectory modeling found five trajectories (quick Sepsis-related Organ Failure Assessment always low, increasing, decreasing, moderate, and always high) with greater predictive validity than the initial measurement (area under the receiver operating characteristic, 0.85 [95% CI, 0.84-0.85]; p < 0.001). CONCLUSIONS: Repeated measurements of quick Sepsis-related Organ Failure Assessment improve predictive validity for sepsis using in-hospital mortality compared with a single measurement of quick Sepsis-related Organ Failure Assessment at the time a clinician suspects infection.

Children with Chronic Disease Bear the Highest Burden of Pediatric Sepsis.

OBJECTIVE: To describe the contemporary epidemiology of pediatric sepsis in children with chronic disease, and the contribution of chronic diseases to mortality. We examined the incidence and hospital mortality of pediatric sepsis in a nationally representative sample and described the contribution of chronic diseases to hospital mortality. STUDY DESIGN: We analyzed the 2013 Nationwide Readmissions Database using a retrospective cohort design. We included non-neonatal patients <19 years of age hospitalized with sepsis. We examined patient characteristics, the distribution of chronic disease, and the estimated national incidence, and described hospital mortality. We used mixed effects logistic regression to explore the association between chronic diseases and hospital mortality. RESULTS: A total of 16 387 admissions, representing 14 243 unique patients, were for sepsis. The national incidence was 0.72 cases per 1000 per year (54 060 cases annually). Most (68.6%) had a chronic disease. The in-hospital mortality was 3.7% overall-0.7% for previously healthy patients and 5.1% for patients with chronic disease. In multivariable analysis, oncologic, hematologic, metabolic, neurologic, cardiac and renal disease, and solid organ transplantation were associated with increased in-hospital mortality. CONCLUSIONS: More than 2 of 3 children admitted with sepsis have ≥1 chronic disease and these patients have a higher in-hospital mortality than previously healthy patients. The burden of sepsis in hospitalized children is greatest in pediatric patients with chronic disease.

Prompt Administration of Antibiotics and Fluids in the Treatment of Sepsis: A Murine Trial.

OBJECTIVES: Sepsis, the acute organ dysfunction caused by a dysregulated host response to infection, poses a serious public health burden. Current management includes early detection, initiation of antibiotics and fluids, and source control as necessary. Although observational data suggest that delays of even a few hours in the initiation of antibiotics or IV fluids is associated with survival, these findings are controversial. There are no randomized data in humans, and prior animal studies studied time from experimental manipulation, not from the onset of clinical features of sepsis. Using a recently developed murine cecal ligation and puncture model that precisely monitors physiologic deterioration, we hypothesize that incremental hourly delays in the first dose of antibiotics, in the first bolus of fluid resuscitation, or a combination of the two at a clinically relevant point of physiologic deterioration during polymicrobial sepsis will shorten survival. DESIGN: Randomized laboratory animal experimental trial. SETTING: University basic science laboratory. SUBJECTS: Male C57BL/6J, female C57BL/6J, aged (40-50 wk old) male C57BL/6J, and BALB/C mice. INTERVENTIONS: Mice (n = 200) underwent biotelemetry-enhanced cecal ligation and puncture and were randomized after meeting validated criteria for acute physiologic deterioration. Treatment groups consisted of a single dose of imipenem/cilastatin, a single bolus of 30 mL/kg fluid resuscitation, or a combination of the two. Mice were allocated to receive treatment at the time of meeting deterioration criteria, after a 2-hour delay or after a 4-hour delay. MEASUREMENTS AND MAIN RESULTS: Hourly delays in the initiation of antibiotic therapy led to progressively shortened survival in our model (p < 0.001). The addition of fluid resuscitation was unable to rescue animals, which received treatment 4 hours after meeting enrollment criteria. Systemic inflammation was increased, and host physiology was increasingly deranged with hourly delays to antibiotics. CONCLUSIONS: We conclude that antibiotic therapy is highly time sensitive, and efforts should be made to deliver this critical therapy as early as possible in sepsis, perhaps extending into the first point of medical contact outside the hospital.

Use of Biotelemetry to Define Physiology-Based Deterioration Thresholds in a Murine Cecal Ligation and Puncture Model of Sepsis.

OBJECTIVES: Murine models of critical illness are commonly used to test new therapeutic interventions. However, these interventions are often administered at fixed time intervals after the insult, perhaps ignoring the inherent variability in magnitude and temporality of the host response. We propose to use wireless biotelemetry monitoring to define and validate criteria for acute deterioration and generate a physiology-based murine cecal ligation and puncture model that is more similar to the conduct of human trials of sepsis. DESIGN: Laboratory and animal research. SETTING: University basic science laboratory. SUBJECTS: Male C57BL/6 mice. INTERVENTIONS: Mice underwent cecal ligation and puncture, and an HD-X11 wireless telemetry monitor (Data Sciences International) was implanted that enabled continuous, real-time measurement of heart rate, core temperature, and mobility. We performed a population-based analysis to determine threshold criteria that met face validity for acute physiologic deterioration. We assessed construct validity by temporally matching mice that met these acute physiologic deterioration thresholds with mice that had not yet met deterioration threshold. We analyzed matched blood samples for blood gas, inflammatory cytokine concentration, cystatin C, and alanine aminotransferase. MEASUREMENTS AND MAIN RESULTS: We observed that a 10% reduction in both heart rate and temperature sustained for greater than or equal to 10 minutes defined acute physiologic deterioration. There was significant variability in the time to reach acute deterioration threshold across mice, ranging from 339 to 529 minutes after cecal ligation and puncture. We found adequate construct validity, as mice that met criteria for acute deterioration had significantly worse shock, systemic inflammation (elevated tumor necrosis factor-α, p = 0.003; interleukin-6, p = 0.01; interleukin-10, p = 0.005), and acute kidney injury when compared with mice that had not yet met acute deterioration criteria. CONCLUSIONS: We defined a murine threshold for acute physiologic deterioration after cecal ligation and puncture that has adequate face and construct validity. This model may enable a more physiology-based model for evaluation of novel therapeutics in critical illness.

Creating an infrastructure for comparative effectiveness research in emergency medical services.

OBJECTIVES: Emergency medical services (EMS) providers deliver the initial care for millions of people in the United States each year. The Institute of Medicine noted a deficit in research necessary to improve prehospital care, created by the existence of data silos, absence of long-term outcomes, and limited stakeholder engagement in research. This article describes a regional effort to create a high-performing infrastructure in southwestern Pennsylvania addressing these fundamental barriers. METHODS: Regional EMS records from 33 agencies in January 2011 were linked to hospital-based electronic health records (EHRs) in a single nine-hospital system, with manual review of matches for accuracy. The use of community stakeholder engagement was included to guide scientific inquiry, as well as 2-year follow up for patient-centered outcomes. RESULTS: Local EMS medicine stakeholders emphasized the limits of single-agency EMS research and suggested that studies focus on improving cross-cutting, long-term outcomes. Guided by this input, more than 95% of EMS records (2,675 of 2,800) were linked to hospital-based EHRs. More than 80% of records were linked to 2-year mortality, with more deaths among EMS patients with prehospital hypotension (30.5%) or respiratory distress (19.5%) than chest pain (5.4%) or nonspecific complaints (9.4%). CONCLUSIONS: A prehospital comparative effectiveness research infrastructure composed of patient-level EMS data, EHRs at multiple hospitals, long-term outcomes, and community stakeholder perspectives is feasible and may be scalable to larger regions and networks. The lessons learned and barriers identified offer a roadmap to answering community and policy-relevant research questions in prehospital care.

Accuracy of prehospital transport time estimation.

OBJECTIVES: Estimates of prehospital transport times are an important part of emergency care system research and planning; however, the accuracy of these estimates is unknown. The authors examined the accuracy of three estimation methods against observed transport times in a large cohort of prehospital patient transports. METHODS: This was a validation study using prehospital records in King County, Washington, and southwestern Pennsylvania from 2002 to 2006 and 2005 to 2011, respectively. Transport time estimates were generated using three methods: linear arc distance, Google Maps, and ArcGIS Network Analyst. Estimation error, defined as the absolute difference between observed and estimated transport time, was assessed, as well as the proportion of estimated times that were within specified error thresholds. Based on the primary results, a regression estimate was used that incorporated population density, time of day, and season to assess improved accuracy. Finally, hospital catchment areas were compared using each method with a fixed drive time. RESULTS: The authors analyzed 29,935 prehospital transports to 44 hospitals. The mean (± standard deviation [±SD]) absolute error was 4.8 (±7.3) minutes using linear arc, 3.5 (±5.4) minutes using Google Maps, and 4.4 (±5.7) minutes using ArcGIS. All pairwise comparisons were statistically significant (p < 0.01). Estimation accuracy was lower for each method among transports more than 20 minutes (mean [±SD] absolute error was 12.7 [±11.7] minutes for linear arc, 9.8 [±10.5] minutes for Google Maps, and 11.6 [±10.9] minutes for ArcGIS). Estimates were within 5 minutes of observed transport time for 79% of linear arc estimates, 86.6% of Google Maps estimates, and 81.3% of ArcGIS estimates. The regression-based approach did not substantially improve estimation. There were large differences in hospital catchment areas estimated by each method. CONCLUSIONS: Route-based transport time estimates demonstrate moderate accuracy. These methods can be valuable for informing a host of decisions related to the system organization and patient access to emergency medical care; however, they should be employed with sensitivity to their limitations.