Faculty Bibliography

Heart failure (HF) is a leading cause of morbidity, healthcare costs, and mortality. Guideline based segmentation of HF into distinct subtypes is coarse and unlikely to reflect the heterogeneity of etiologies and disease trajectories of patients. While analyses of electronic health records show promise in expanding our understanding of complex syndromes like HF in an evidence-driven way, limitations in data quality have presented challenges for large-scale EHR-based insight generation and decision-making. We present a hypothesis-free approach to generating real-world characteristics and progression patterns of HF. Patient disease state snapshots are extracted from the complaints mentioned in unstructured clinical notes. Typical disease states are generated by clustering and characterized in terms of their distinguishing features, temporal relationships, and risk of important clinical events. Our analysis generates a comprehensive "disease phenome" of real-world patients computed from large, noisy, secondary-use EHR datasets created in a routine clinical setting.

Importance/UNASSIGNED:Accurate assessment of wound area and percentage of granulation tissue (PGT) are important for optimizing wound care and healing outcomes. Artificial intelligence (AI)-based wound assessment tools have the potential to improve the accuracy and consistency of wound area and PGT measurement, while improving efficiency of wound care workflows. Objective/UNASSIGNED:To develop a quantitative and qualitative method to evaluate AI-based wound assessment tools compared with expert human assessments. Design, Setting, and Participants/UNASSIGNED:This diagnostic study was performed across 2 independent wound centers using deidentified wound photographs collected for routine care (site 1, 110 photographs taken between May 1 and 31, 2018; site 2, 89 photographs taken between January 1 and December 31, 2019). Digital wound photographs of patients were selected chronologically from the electronic medical records from the general population of patients visiting the wound centers. For inclusion in the study, the complete wound edge and a ruler were required to be visible; circumferential ulcers were specifically excluded. Four wound specialists (2 per site) and an AI-based wound assessment service independently traced wound area and granulation tissue. Main Outcomes and Measures/UNASSIGNED:The quantitative performance of AI tracings was evaluated by statistically comparing error measure distributions between test AI traces and reference human traces (AI vs human) with error distributions between independent traces by 2 humans (human vs human). Quantitative outcomes included statistically significant differences in error measures of false-negative area (FNA), false-positive area (FPA), and absolute relative error (ARE) between AI vs human and human vs human comparisons of wound area and granulation tissue tracings. Six masked attending physician reviewers (3 per site) viewed randomized area tracings for AI and human annotators and qualitatively assessed them. Qualitative outcomes included statistically significant difference in the absolute difference between AI-based PGT measurements and mean reviewer visual PGT estimates compared with PGT estimate variability measures (ie, range, standard deviation) across reviewers. Results/UNASSIGNED:A total of 199 photographs were selected for the study across both sites; mean (SD) patient age was 64 (18) years (range, 17-95 years) and 127 (63.8%) were women. The comparisons of AI vs human with human vs human for FPA and ARE were not statistically significant. AI vs human FNA was slightly elevated compared with human vs human FNA (median [IQR], 7.7% [2.7%-21.2%] vs 5.7% [1.6%-14.9%]; P < .001), indicating that AI traces tended to slightly underestimate the human reference wound boundaries compared with human test traces. Two of 6 reviewers had a statistically higher frequency in agreement that human tracings met the standard area definition, but overall agreement was moderate (352 yes responses of 583 total responses [60.4%] for AI and 793 yes responses of 1166 total responses [68.0%] for human tracings). AI PGT measurements fell in the typical range of variation in interreviewer visual PGT estimates; however, visual PGT estimates varied considerably (mean range, 34.8%; mean SD, 19.6%). Conclusions and Relevance/UNASSIGNED:This study provides a framework for evaluating AI-based digital wound assessment tools that can be extended to automated measurements of other wound features or adapted to evaluate other AI-based digital image diagnostic tools. As AI-based wound assessment tools become more common across wound care settings, it will be important to rigorously validate their performance in helping clinicians obtain accurate wound assessments to guide clinical care.

Purpose of study The COVID-19 pandemic led to an unprecedented rapid transmission of healthcare information. This information was critical to enact frequently changing patient care protocols and to inform staff about redistribution of hospital resources at New York University Langone Hospital- Long Island. In this investigation, we analyze our hospital clinicians' methods of mass communication to front-line health care workers, with particular interest in assessing how communication was informed by real-time clinical findings. At the height of the pandemic (March 25th- April 15th), a mass broadcast email disseminated daily from the Director of Pulmonary and Critical Care was effective in informing treatment protocols that were clinically observed to improve patient outcomes. We analyzed over thirty broadcast emails and identified three major categories of information that were routinely addressed and/or updated: (i) reallocation of resources, (ii) clinical protocol changes, (iii) recommended lab tests for monitoring patient clinical course. We also interviewed key hospital clinicians and administrators on their experience working during the height of the pandemic. We found treatment protocols in these emails included information regarding the use of steroids and monoclonal antibody therapy, ventilators, and patient repositioning. In addition, the hospital's first autopsy results on COVID related deaths gave further insight into the disease process and manner of death for many patients (diffuse alveolar damage and evidence of hypercoagulability). So, too, did clinical findings around this time support what was seen grossly on autopsy-patients with more severe disease often presented with serial d-dimer levels >6x the normal limit. The information through these different conduits was synthesized and subsequently communicated in the aforementioned mass emails as an anticoagulation treatment protocol. Through continuous input of data, this protocol was updated and adjusted over the course of three weeks. We found that real-time communication amongst hospital staff regarding patient treatment protocols was a dynamic process that required synthesis of lab values, autopsy findings, and observed response to treatments. Successful treatment of patients depended on continuous review and communication of this information. Methods used The COVID-19 pandemic led to an unprecedented rapid transmission of healthcare information. This information was critical to enact frequently changing patient care protocols and to inform staff about redistribution of hospital resources at New York University Langone Hospital-- Long Island. In this investigation, we analyze our hospital clinicians' methods of mass communication to front-line health care workers, with particular interest in assessing how communication was informed by real-time clinical findings. At the height of the pandemic (March 25th- April 15th), a mass broadcast email disseminated daily from the Director of Pulmonary and Critical Care was effective in informing treatment protocols that were clinically observed to improve patient outcomes. Summary of results We analyzed over thirty broadcast emails and identified three major categories of information that were routinely addressed and/or updated: (i) reallocation of resources, (ii) clinical protocol changes, (iii) recommended lab tests for monitoring patient clinical course. We also interviewed key hospital clinicians and administrators on their experience working during the height of the pandemic. We found treatment protocols in these emails included information regarding the use of steroids and monoclonal antibody therapy, ventilators, and patient repositioning. In addition, the hospital's first autopsy results on COVID related deaths gave further insight into the disease process and manner of death for many patients (diffuse alveolar damage and evidence of hypercoagulability). So, too, did clinical findings around this time support what was seen grossly on autopsy- patients with more severe disease often presented with serial d-dimer levels >6x the normal limit. The information through these different conduits was synthesized and subsequently communicated in the aforementioned mass emails as an anticoagulation treatment protocol. Through continuous input of data, this protocol was updated and adjusted over the course of three weeks. Conclusions We found that real-time communication amongst hospital staff regarding patient treatment protocols was a dynamic process that required synthesis of lab values, autopsy findings, and observed response to treatments. Successful treatment of patients depended on continuous review and communication of this information

As a leading cause of death and morbidity, heart failure (HF) is responsible for a large portion of healthcare and disability costs worldwide. Current approaches to define specific HF subpopulations may fail to account for the diversity of etiologies, comorbidities, and factors driving disease progression, and therefore have limited value for clinical decision making and development of novel therapies. Here we present a novel and data-driven approach to understand and characterize the real-world manifestation of HF by clustering disease and symptom-related clinical concepts (complaints) captured from unstructured electronic health record clinical notes. We used natural language processing to construct vectorized representations of patient complaints followed by clustering to group HF patients by similarity of complaint vectors. We then identified complaints that were significantly enriched within each cluster using statistical testing. Breaking the HF population into groups of similar patients revealed a clinically interpretable hierarchy of subgroups characterized by similar HF manifestation. Importantly, our methodology revealed well-known etiologies, risk factors, and comorbid conditions of HF (including ischemic heart disease, aortic valve disease, atrial fibrillation, congenital heart disease, various cardiomyopathies, obesity, hypertension, diabetes, and chronic kidney disease) and yielded additional insights into the details of each HF subgroup's clinical manifestation of HF. Our approach is entirely hypothesis free and can therefore be readily applied for discovery of novel insights in alternative diseases or patient populations.

Organoids are becoming widespread in drug-screening technologies but have been used sparingly for cell therapy as current approaches for producing self-organized cell clusters lack scalability or reproducibility in size and cellular organization. We introduce a method of using hydrogels as sacrificial scaffolds, which allow cells to form self-organized clusters followed by gentle release, resulting in highly reproducible multicellular structures on a large scale. We demonstrated this strategy for endothelial cells and mesenchymal stem cells to self-organize into blood-vessel units, which were injected into mice, and rapidly formed perfusing vasculature. Moreover, in a mouse model of peripheral artery disease, intramuscular injections of blood-vessel units resulted in rapid restoration of vascular perfusion within seven days. As cell therapy transforms into a new class of therapeutic modality, this simple method-by making use of the dynamic nature of hydrogels-could offer high yields of self-organized multicellular aggregates with reproducible sizes and cellular architectures.

Objective/UNASSIGNED:Given the high mortality and prolonged duration of mechanical ventilation of COVID-19 patients, we evaluated the safety and efficacy of hyperbaric oxygen for COVID-19 patients with respiratory distress. Methods/UNASSIGNED:This is a single-center clinical trial of COVID-19 patients at NYU Winthrop Hospital from March 31 to April 28, 2020. Patients in this trial received hyperbaric oxygen therapy at 2.0 atmospheres of pressure in monoplace hyperbaric chambers for 90 minutes daily for a maximum of five total treatments. Controls were identified using propensity score matching among COVID-19 patients admitted during the same time period. Using competing-risks survival regression, we analyzed our primary outcome of inpatient mortality and secondary outcome of mechanical ventilation. Results/UNASSIGNED:We treated 20 COVID-19 patients with hyperbaric oxygen. Ages ranged from 30 to 79 years with an oxygen requirement ranging from 2 to 15 liters on hospital days 0 to 14. Of these 20 patients, two (10%) were intubated and died, and none remain hospitalized. Among 60 propensity-matched controls based on age, sex, body mass index, coronary artery disease, troponin, D-dimer, hospital day, and oxygen requirement, 18 (30%) were intubated, 13 (22%) have died, and three (5%) remain hospitalized (with one still requiring mechanical ventilation). Assuming no further deaths among controls, we estimate that the adjusted subdistribution hazard ratios were 0.37 for inpatient mortality (p=0.14) and 0.26 for mechanical ventilation (p=0.046). Conclusion/UNASSIGNED:Though limited by its study design, our results demonstrate the safety of hyperbaric oxygen among COVID-19 patients and strongly suggests the need for a well-designed, multicenter randomized control trial.

Chronic nonhealing wounds cause significant morbidity and mortality and remain a challenging condition to treat. Regenerative wound surgery involves operative debridement of wounds to remove dead and healing-impaired tissue and bacterial contamination and, subsequently, the application of regenerative medicine treatments to accelerate healing. Regenerative treatments aim to restore native tissue structure and function by targeting biological mechanisms underlying impaired healing. A wide range of regenerative modalities are used for treating chronic and complex wounds, including decellularized scaffolds, living engineered donor tissues, autologous stem cells, and recombinant growth factors. Each of these modalities has specific and sometimes complex requirements for implementation. The advanced wound care team, including OR staff members, should be aware of how these products are used and regulated. This article highlights some of the common and emerging regenerative treatments that are applied in wound surgery and focuses on how the products are used practically in the OR.

GENERAL PURPOSE/UNASSIGNED:To describe the development of an evidence-based wound electronic medical record (WEMR) framework for providers to execute timely, protocol-based, best-practice care for patients with chronic, nonhealing wounds. TARGET AUDIENCE/BACKGROUND:This continuing education activity is intended for physicians, physician assistants, nurse practitioners, and nurses with an interest in skin and wound care. LEARNING OBJECTIVES/OUTCOMES/UNASSIGNED:After completing this continuing education activity, you should be better able to: ABSTRACT: The care of patients with nonhealing wounds involves a host of treatment modalities. The authors developed a wound-specific framework to enhance provider management of these wounds and a summary sheet to involve patients and caregivers in their own healthcare to improve treatment adherence and outcomes. Implementing evidence-based practice for chronic wounds enables corrective actions to optimize care.

GENERAL PURPOSE/UNASSIGNED:To provide information about the diagnosis and treatment of diabetic myonecrosis (DMN).This continuing education activity is intended for physicians, physician assistants, nurse practitioners, and nurses with an interest in skin and wound care.After participating in this educational activity, the participant should be better able to:1. Cite the incidence and symptomatology of diabetic myonecrosis.2. Identify the diagnostic tests associated with DMN.3. Summarize the evidence-based treatments for DMN.Diabetic myonecrosis is a rare complication of poorly controlled diabetes mellitus that presents similarly to many common conditions such as cellulitis, abscess, and fasciitis. Therefore, a high index of suspicion is required for diagnosis. Magnetic resonance imaging is the investigative test of choice. Treatment includes antiplatelet therapy, nonsteroidal anti-inflammatory agents, and glycemic control.

Merkel cell carcinoma (MCC) is a rare, aggressive carcinoma that usually arises in sun-exposed regions. MCC is a primary neuroendocrine tumor that arises in the skin. This report describes an unusual case of MCC on the buttocks that was treated with excision, radiation and chemotherapy. Physicians should consider MCC as a differential diagnosis when encountering a rapidly growing, painless lesion. Early diagnosis and treatment may improve patient survival rates.