Development and Evaluation of a Machine Learning Model for the Early Identification of Patients at Risk for Sepsis.

STUDY OBJECTIVE: The Third International Consensus Definitions (Sepsis-3) Task Force recommended the use of the quick Sequential [Sepsis-related] Organ Failure Assessment (qSOFA) score to screen patients for sepsis outside of the ICU. However, subsequent studies raise concerns about the sensitivity of qSOFA as a screening tool. We aim to use machine learning to develop a new sepsis screening tool, the Risk of Sepsis (RoS) score, and compare it with a slate of benchmark sepsis-screening tools, including the Systemic Inflammatory Response Syndrome, Sequential Organ Failure Assessment (SOFA), qSOFA, Modified Early Warning Score, and National Early Warning Score. METHODS: We used retrospective electronic health record data from adult patients who presented to 49 urban community hospital emergency departments during a 22-month period (N=2,759,529). We used the Rhee clinical surveillance criteria as our standard definition of sepsis and as the primary target for developing our model. The data were randomly split into training and test cohorts to derive and then evaluate the model. A feature selection process was carried out in 3 stages: first, we reviewed existing models for sepsis screening; second, we consulted with local subject matter experts; and third, we used a supervised machine learning called gradient boosting. Key metrics of performance included alert rate, area under the receiver operating characteristic curve, sensitivity, specificity, and precision. Performance was assessed at 1, 3, 6, 12, and 24 hours after an index time. RESULTS: The RoS score was the most discriminant screening tool at all time thresholds (area under the receiver operating characteristic curve 0.93 to 0.97). Compared with the next most discriminant benchmark (Sequential Organ Failure Assessment), RoS was significantly more sensitive (67.7% versus 49.2% at 1 hour and 84.6% versus 80.4% at 24 hours) and precise (27.6% versus 12.2% at 1 hour and 28.8% versus 11.4% at 24 hours). The sensitivity of qSOFA was relatively low (3.7% at 1 hour and 23.5% at 24 hours). CONCLUSION: In this retrospective study, RoS was more timely and discriminant than benchmark screening tools, including those recommend by the Sepsis-3 Task Force. Further study is needed to validate the RoS score at independent sites.

The voice conveys emotion in ten globalized cultures and one remote village in Bhutan.

With data from 10 different globalized cultures and 1 remote, isolated village in Bhutan, we examined universals and cultural variations in the recognition of 16 nonverbal emotional vocalizations. College students in 10 nations (Study 1) and villagers in remote Bhutan (Study 2) were asked to match emotional vocalizations to 1-sentence stories of the same valence. Guided by previous conceptualizations of recognition accuracy, across both studies, 7 of the 16 vocal burst stimuli were found to have strong or very strong recognition in all 11 cultures, 6 vocal bursts were found to have moderate recognition, and 4 were not universally recognized. All vocal burst stimuli varied significantly in terms of the degree to which they were recognized across the 11 cultures. Our discussion focuses on the implications of these results for current debates concerning the emotion conveyed in the voice.

Ethical considerations in embedding a surgeon in a military or civilian tactical team.

Tactical emergency medical services (TEMS) bring immediate medical support to the inner perimeter of special weapons and tactics team activations. While initially envisioned as a role for an individual dually trained as a police officer and paramedic, TEMS is increasingly undertaken by physicians and paramedics who are not police officers. This report explores the ethical underpinnings of embedding a surgeon within a military or civilian tactical team with regard to identity, ethically acceptable actions, triage, responsibility set, training, certification, and potential future refinements of the role of the tactical police surgeon.

Massive transfusion of blood in the surgical patient.

Hemorrhage remains a leading cause of morbidity and death in both civilian and military trauma. Restoration of effective end-organ perfusion by stopping hemorrhage and restoring intravascular volume in such a way as to minimize acidosis, hypothermia, and coagulopathy, almost always requires the use of blood and/or blood-component therapy. The best method to manage life-threatening hemorrhage is to avoid the circumstance that prompted it or to mitigate blood loss early in the injury cycle; otherwise, blood replacement must suffice. This article reviews current understanding of massive transfusion, along with its attendant unintended consequences, in the management of patients with profound hemorrhage.

The evaluation of morbidity and mortality in abdominal aortic aneurysm repair patients as related to body mass index.

BACKGROUND: This study evaluates the relationship between body mass index (BMI) and other comorbidities on the overall morbidity and mortality of abdominal aortic aneurysm (AAA) repair. METHODS: A database of all nonemergent open and endovascular AAA repairs performed at our center from 2004 to 2008 was created. The outcomes at the predefined time intervals were then evaluated for each group of patients. RESULTS: One hundred forty-three patients qualified for this study with a 3:2 stent graft-to-open ratio. A trend relating patient BMI with early mortality was noted. Age>80 years was a strong predictor of mortality in all time intervals. CONCLUSIONS: The outcomes for this population show a significant trend toward early mortality in open AAA repair patients with an elevated preoperative BMI. Appropriate patient selection and preoperative optimization are recommended for all AAA repair candidates; however, some innate characteristics such as patient age, may play the largest role in determining outcomes.

A single center experience in the use of polyurethaneurea arteriovenous grafts.

BACKGROUND: Polytetrafluoroethylene (PTFE) grafts often require temporary catheters, which increase costs as well as risk of infection, bleeding, and poor blood flow. Polyurethaneurea (PUU) grafts can be cannulated within 24 hours of placement, avoiding the need for temporary catheters. METHODS: Hemodialysis patients who were not candidates for arteriovenous fistula (AVF) were eligible for the study. Those requiring urgent (<48 hours) hemodialysis access received PUU grafts; all others received PTFE grafts. One surgeon performed all implantations. RESULTS: 133 PUU grafts were implanted, 31 (24%) patients died (unrelated to graft placement). Within the first 4 days after graft placement, 108 of 133 grafts (81%) were cannulated. None required a temporary catheter. Of 102 surviving patients, 50 (49%) had graft thrombosis; 47/50 (94%) underwent successful percutaneous thrombectomy (PT). Primary patency at 6 months and at 1 year was 51% and 33%, respectively. Secondary patency at 6 months and 1 year was 78% and 61%, respectively. Mean bleeding time (after withdrawal of dialysis needle) was 4.0 (range 1-14) minutes in PUU grafts vs. 9.2 (1-30) minutes in PTFE grafts. CONCLUSIONS: Polyurethane graft placement allowed early graft cannulation without temporary catheters. Patency rates are comparable with previous reports. Decreased bleeding time with PUU grafts reduced aggregate blood exposure for patients and staff, lowering the risk of infection and anemia exacerbation and speeding time between treatments.