NutriSighT: Interpretable Transformer Model for Dynamic Prediction of Hypocaloric Enteral Nutrition in Mechanically Ventilated Patients.

Achieving adequate enteral nutrition among mechanically ventilated patients is challenging, yet critical. We developed NutriSighT, a transformer model using learnable positional coding to predict which patients would achieve hypocaloric nutrition between days 3-7 of mechanical ventilation. Using retrospective data from two large ICU databases (3,284 patients from AmsterdamUMCdb - development set, and 6,456 from MIMIC-IV - external validation set), we included adult patients intubated for at least 72 hours.

Effect of Tiered Implementation of Clinical Decision Support System for Acute Kidney Injury and Nephrotoxin Exposure in Cardiac Surgery Patients.

Background:  Nephrotoxin exposure may worsen kidney injury and impair kidney recovery if continued in patients with acute kidney injury (AKI).

Objectives:  This study aimed to determine if tiered implementation of a clinical decision support system (CDSS) would reduce nephrotoxin use in cardiac surgery patients with AKI.

Methods:  We assessed patients admitted to the cardiac surgery intensive care unit at a tertiary care center from January 2020 to December 2021, and August 2022 to September 2023.

Extracting International Classification of Diseases Codes from Clinical Documentation using Large Language Models.

Background: Large language models (LLMs) have shown promise in various professional fields, including medicine and law. However, their performance in highly specialized tasks, such as extracting ICD-10-CM codes from patient notes, remains underexplored.

Objective: The primary objective was to evaluate and compare the performance of ICD-10-CM code extraction by different LLMs with that of human coder.

Benchmarking Large Language Models for Extraction of International Classification of Diseases Codes from Clinical Documentation.

Background: Healthcare reimbursement and coding is dependent on accurate extraction of International Classification of Diseases-tenth revision - clinical modification (ICD-10-CM) codes from clinical documentation. Attempts to automate this task have had limited success. This study aimed to evaluate the performance of large language models (LLMs) in extracting ICD-10-CM codes from unstructured inpatient notes and benchmark them against human coder.

Managing End of Life Care for the Critically Ill: A Novel Program to Deliver Bedside Critical Care Without Transfer to the Intensive Care Unit.

Navigating medical care at the end of life can be a challenging experience for patients. There are also significant resource burdens, including intensive care unit (ICU) admissions, accompanying terminal illness. For actively dying patients, developing a care plan based on patient goals and delivering care at the bedside can enhance patient well-being, avoid inappropriate transfers or interventions, and improve resource management.

Role of artificial intelligence in critical care nutrition support and research.

Nutrition plays a key role in the comprehensive care of critically ill patients. Determining optimal nutrition strategy, however, remains a subject of intense debate. Artificial intelligence (AI) applications are becoming increasingly common in medicine, and specifically in critical care, driven by the data-rich environment of intensive care units.

Machine learning derived serum creatinine trajectories in acute kidney injury in critically ill patients with sepsis.

Background: Current classification for acute kidney injury (AKI) in critically ill patients with sepsis relies only on its severity-measured by maximum creatinine which overlooks inherent complexities and longitudinal evaluation of this heterogenous syndrome. The role of classification of AKI based on early creatinine trajectories is unclear.

Methods: This retrospective study identified patients with Sepsis-3 who developed AKI within 48-h of intensive care unit admission using Medical Information Mart for Intensive Care-IV database.

The COVID-19 Tracheostomy Experience at a Large Academic Medical Center in New York during the First Year.

New York City was the epicenter of the initial surge of the COVID-19 pandemic in the United States. Tracheostomy is a critical procedure in the care of patients with COVID-19. We hypothesized that early tracheostomy would decrease the length of time on sedation, time on mechanical ventilation, intensive care unit length of stay, and mortality.

Real-Time Machine Learning Alerts to Prevent Escalation of Care: A Nonrandomized Clustered Pragmatic Clinical Trial.

Objectives: Machine learning algorithms can outperform older methods in predicting clinical deterioration, but rigorous prospective data on their real-world efficacy are limited. We hypothesized that real-time machine learning generated alerts sent directly to front-line providers would reduce escalations.

Design: Single-center prospective pragmatic nonrandomized clustered clinical trial.

Bayesian methods: a potential path forward for sepsis trials.

Background: Given the success of recent platform trials for COVID-19, Bayesian statistical methods have become an option for complex, heterogenous syndromes like sepsis. However, study design will require careful consideration of how statistical power varies using Bayesian methods across different choices for how historical data are incorporated through a prior distribution and how the analysis is ultimately conducted. Our objective with the current analysis is to assess how different uses of historical data through a prior distribution, and type of analysis influence results of a proposed trial that will be analyzed using Bayesian statistical methods.

Expanding access to difficult urinary catheter insertion services through a novel nurse practitioner-led team.

Background: Within the United States health care system, one of the most common procedures performed daily is urinary catheterization. Oftentimes, the urinary catheter is placed by nursing personnel without any difficulty. Although the procedure is usually simple and routine, there are instances in which placement can be problematic.

Sociodemographic Factors Associated with Do-Not-Resuscitate Order Utilization in the Surgical Intensive Care Unit: An Observational Study.

The use of a do-not-resuscitate (DNR) order is a powerful tool in outlining end-of-life care. This study explores sociodemographic factors associated with selection of a DNR order and assigning a healthcare proxy in the Surgical Intensive Care Unit (SICU). A retrospective chart review of 312 patients who expired in the SICU over a 7-year period was conducted.

Critical Care and Postoperative Management of the First Human Long Segment Orthotopic Tracheal Transplantation.

Unlabelled: To outline the postoperative management of a long segment tracheal transplant in the ICU setting.

Design: The recipient required reconstruction of a long segment tracheal defect from a previous prolonged intubation. A male donor was chosen for a female recipient to enable analysis of the reepithelialization kinetics using fluorescence in situ hybridization to analyze the source of the new ciliated epithelium.

The fourth wave: vaccination status and intensive care unit mortality at a large hospital system in New York City.

Background: We aim to describe the demographics and outcomes of patients with severe disease with the Omicron variant. The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus continues to mutate, and the availability of vaccines and boosters continue to rise, it is important to understand the health care burden of new variants. We analyze patients admitted to intensive care units (ICUs) in a large Academic Health System during New York City's fourth surge beginning on November 27, 2021.

Adaptations in vascular access procedures as an initiative to protect healthcare workers throughout the COVID-19 pandemic.

COVID-19 has led to procedural changes in vascular access services to protect healthcare workers and patients from further spread of the virus. Operational changes made by the vascular access service at a healthcare system in New York City during the first wave of the COVID-19 (SARS-CoV-2) pandemic included a team-based approach as well as consideration for types of lines placed to address the increase in patient volume while providing safety to healthcare workers and conserving personal protective equipment. The study consists of two samples of adult inpatients admitted to Mount Sinai Hospital in New York City in need of vascular access.

Postoperative Nursing Care of a Deceased Donor Tracheal Transplant Recipient.

Introduction: Certain airway disorders, such as tracheal stenosis, can severely affect the ability to breathe, reduce quality of life, and increase morbidity and mortality. Treatment options for long-segment tracheal stenosis include multistage tracheal replacement with biosynthetic material, autotransplantation, and allotransplantation. These interventions have not demonstrated long-term dependable results because of lack of adequate blood supply to the organ and ciliated epithelium.

Novel Multidisciplinary Vascular Access Team Helps to Improve ICU Workflow Amidst Covid-19 Pandemic.

Objective: The surge in critically ill patients has pressured hospitals to expand their intensive care unit capacities and critical care staff. This was difficult given the country's shortage of intensivists. This paper describes the implementation of a multidisciplinary central line placement team and its impact in reducing the vascular access workload of ICU physicians during the height of the COVID-19 pandemic.

The Third Wave: Comparing Seasonal Trends in COVID-19 Patient Data at a Large Hospital System in New York City.

Unlabelled: The third wave of COVID-19 is unique in that vaccines have been widely available; however, the highly transmissible Delta variant has been the predominant strain. Temporal changes of hospitalized patient characteristics should continue to be analyzed as COVID-19 progresses.

Objectives: Compare the demographics and outcomes of hospitalized patients during New York City's third wave of COVID-19 to the first two waves.

Development of a Peripheral Intravenous Access Training Program for Nurses in the Pediatric Intensive Care Units.

Purpose: Obtaining vascular access in the pediatric population can be challenging, with insertion success rates varying widely based on patient and practitioner associated factors. Difficulty establishing peripheral intravenous access can delay treatment, which can be detrimental in emergent situations. Nurses who are trained in vascular access yield a much higher first attempt success rate, which decreases resource utilization, time to intervention, and complication rate.

Prognostic Value of Electrocardiographic QRS Diminution in Patients Hospitalized With COVID-19 or Influenza.

During the clinical care of hospitalized patients with COVID-19, diminished QRS amplitude on the surface electrocardiogram (ECG) was observed to precede clinical decompensation, culminating in death. This prompted investigation into the prognostic utility and specificity of low QRS complex amplitude (LoQRS) in COVID-19. We retrospectively analyzed consecutive adults admitted to a telemetry service with SARS-CoV-2 (n = 140) or influenza (n = 281) infection with a final disposition-death or discharge.

Critical Care Simulation Education Program During the COVID-19 Pandemic.

Background: Coronaviruses are important emerging human and animal pathogens. SARS-CoV-2, the virus that causes COVID-19, is responsible for the current global pandemic. Early in the course of the pandemic, New York City became one of the world's "hot spots" with more than 250,000 cases and more than 15,000 deaths.

The Mount Sinai Hospital Institute for critical care medicine response to the COVID-19 pandemic.

Background: The coronavirus disease 2019 (COVID-19) pandemic resulted in a surge of critically ill patients. This was especially true in New York City. We present a roadmap for hospitals and healthcare systems to prepare for a Pandemic.