A Delphi Process to Identify Relevant Outcomes That May Be Associated With a Predictive Analytic Tool to Detect Hemodynamic Deterioration in the Intensive Care Unit.

Background The critical care literature has seen an increase in the development and validation of tools using artificial intelligence for early detection of patient events or disease onset in the intensive care unit (ICU). The hemodynamic stability index (HSI) was found to have an AUC of 0.82 in predicting the need for hemodynamic intervention in the ICU. Future studies using this tool may benefit from targeting those outcomes that are more relevant to clinicians and most achievable. Methods A three-round Delphi study was conducted with a panel of 10 critical care physicians and three nurses in the United States to identify outcomes that may be most relevant and achievable with the HSI when evaluated for use in the ICU. To achieve criteria for relevance, at least 65% of panelists had to rate an outcome as a 4 or 5 on a 5-point scale. Results Nineteen of 24 outcomes that may be associated with the HSI achieved consensus for relevance. The Kemeny-Young approach was used to develop a matrix depicting the distribution of outcomes considering both relevance and achievability. "Reduces time spent in hemodynamic instability" and "reduces times to recognition of hemodynamic instability" were the highest-ranking outcomes considering both relevance and achievability. Conclusion This Delphi study was a feasible method to identify relevant outcomes that may be associated with an appropriate predictive analytic tool in the ICU. These findings can provide insight to researchers looking to study such tools to impact outcomes relevant to critical care practitioners. Future studies should test these tools in the ICU that target the most clinically relevant and achievable outcomes, such as time spent hemodynamically unstable or time until actionable nursing assessment or treatment.

Sepsis as 2 problems: Identifying sepsis at admission and predicting onset in the hospital using an electronic medical record-based acuity score.

PURPOSE: Early identification and treatment improve outcomes for patients with sepsis. Current screening tools are limited. We present a new approach, recognizing that sepsis patients comprise 2 distinct and unequal populations: patients with sepsis present on admission (85%) and patients who develop sepsis in the hospital (15%) with mortality rates of 12% and 35%, respectively. METHODS: Models are developed and tested based on 258 836 adult inpatient records from 4 hospitals. A "present on admission" model identifies patients admitted to a hospital with sepsis, and a "not present on admission" model predicts postadmission onset. Inputs include common clinical measurements and the Rothman Index. Sepsis was determined using International Classification of Diseases, Ninth Revision, codes. RESULTS: For sepsis present on admission, area under the curves ranged from 0.87 to 0.91. Operating points chosen to yield 75% and 50% sensitivity achieve positive predictive values of 17% to 25% and 29% to 40%, respectively. For sepsis not present on admission, at 65% sensitivity, positive predictive values ranged from 10% to 20% across hospitals. CONCLUSIONS: This approach yields good to excellent discriminatory performance among adult inpatients for predicting sepsis present on admission or developed within the hospital and may aid in the timely delivery of care.

The epidemiology of spontaneous fever and hypothermia on admission of brain injury patients to intensive care units: a multicenter cohort study.

OBJECTIVES: Fever and hypothermia (dysthermia) are associated with poor outcomes in patients with brain injuries. The authors sought to study the epidemiology of dysthermia on admission to the intensive care unit (ICU) and the effect on in-hospital case fatality in a mixed cohort of patients with brain injuries. METHODS: The authors conducted a multicenter retrospective cohort study in 94 ICUs in the United States. Critically ill patients with neurological injuries, including acute ischemic stroke (AIS), aneurysmal subarachnoid hemorrhage (aSAH), intracerebral hemorrhage (ICH), and traumatic brain injury (TBI), who were older than 17 years and consecutively admitted to the ICU from 2003 to 2008 were selected for analysis. RESULTS: In total, 13,587 patients were included in this study; AIS was diagnosed in 2973 patients (22%), ICH in 4192 (31%), aSAH in 2346 (17%), and TBI in 4076 (30%). On admission to the ICU, fever was more common among TBI and aSAH patients, and hypothermia was more common among ICH patients. In-hospital case fatality was more common among patients with hypothermia (OR 12.7, 95% CI 8.4-19.4) than among those with fever (OR 1.9, 95% CI 1.7-2.1). Compared with patients with ICH (OR 2.0, 95% CI 1.8-2.3), TBI (OR 1.5, 95% CI 1.3-1.8), and aSAH (OR 1.4, 95% CI 1.2-1.7), patients with AIS who developed fever had the highest risk of death (OR 3.1, 95% CI 2.5-3.7). Although all hypothermic patients had an increased mortality rate, this increase was not significantly different across subgroups. In a multivariable analysis, when adjusted for all other confounders, exposure to fever (adjusted OR 1.3, 95% CI 1.1-1.5) or hypothermia (adjusted OR 7.8, 95% CI 3.9-15.4) on admission to the ICU was found to be significantly associated with in-hospital case fatality. CONCLUSIONS: Fever is frequently encountered in the acute phase of brain injury, and a small proportion of patients with brain injuries may also develop spontaneous hypothermia. The effect of fever on mortality rates differed by neurological diagnosis. Both early spontaneous fever and hypothermia conferred a higher risk of in-hospital death after brain injury.

Development and simultaneous application of multiple care protocols in critical care: a multicenter feasibility study.

OBJECTIVE: To test the feasibility of and interactions among three software-driven critical care protocols. DESIGN: Prospective cohort study. SETTING: Intensive care units in six European and American university hospitals. PATIENTS: 174 cardiac surgery and 41 septic patients. INTERVENTIONS: Application of software-driven protocols for cardiovascular management, sedation, and weaning during the first 7 days of intensive care. MEASUREMENTS AND RESULTS: All protocols were used simultaneously in 85% of the cardiac surgery and 44% of the septic patients, and any one of the protocols was used for 73 and 44% of study duration, respectively. Protocol use was discontinued in 12% of patients by the treating clinician and in 6% for technical/administrative reasons. The number of protocol steps per unit of time was similar in the two diagnostic groups (n.s. for all protocols). Initial hemodynamic stability (a protocol target) was achieved in 26+/-18 min (mean+/-SD) in cardiac surgery and in 24+/-18 min in septic patients. Sedation targets were reached in 2.4+/-0.2h in cardiac surgery and in 3.6 +/-0.2h in septic patients. Weaning protocol was started in 164 (94%; 154 extubated) cardiac surgery and in 25 (60%; 9 extubated) septic patients. The median (interquartile range) time from starting weaning to extubation (a protocol target) was 89 min (range 44-154 min) for the cardiac surgery patients and 96 min (range 56-205 min) for the septic patients. CONCLUSIONS: Multiple software-driven treatment protocols can be simultaneously applied with high acceptance and rapid achievement of primary treatment goals. Time to reach these primary goals may provide a performance indicator.