Faculty Bibliography

During the coronavirus disease 2019 (COVID-19) pandemic, rapid and accurate triage of patients at the emergency department is critical to inform decision-making. We propose a data-driven approach for automatic prediction of deterioration risk using a deep neural network that learns from chest X-ray images and a gradient boosting model that learns from routine clinical variables. Our AI prognosis system, trained using data from 3661 patients, achieves an area under the receiver operating characteristic curve (AUC) of 0.786 (95% CI: 0.745-0.830) when predicting deterioration within 96 hours. The deep neural network extracts informative areas of chest X-ray images to assist clinicians in interpreting the predictions and performs comparably to two radiologists in a reader study. In order to verify performance in a real clinical setting, we silently deployed a preliminary version of the deep neural network at New York University Langone Health during the first wave of the pandemic, which produced accurate predictions in real-time. In summary, our findings demonstrate the potential of the proposed system for assisting front-line physicians in the triage of COVID-19 patients.

Spike protein mutations E484K and N501Y carried by SARS-CoV-2 variants have been associated with concerning changes of the virus, including resistance to neutralizing antibodies and increased transmissibility. While the concerning variants are fast spreading in various geographical areas, identification and monitoring of these variants is lagging far behind, due in large part to the slow speed and insufficient capacity of viral sequencing. In response to the unmet need for a fast and efficient screening tool, we developed a single-tube duplex molecular assay for rapid and simultaneous identification of E484K and N501Y mutations from nasopharyngeal swab (NS) samples within 2.5 h from sample preparation to report. Using this tool, we screened a total of 1135 clinical NS samples collected from COVID patients at 8 hospitals within the Hackensack Meridian Health network in New Jersey between late December 2020 and March 2021. Our data revealed dramatic increases in the frequencies of both E484K and N501Y over time, underscoring the need for continuous epidemiological monitoring.

BACKGROUND:Neurologists need to be adept at disclosing prognosis and breaking bad news. Objective structured clinical examinations (OSCE) allow trainees to practice these skills. METHODS:In 2017, in conjunction with the NYU School of Medicine Simulation Center, neurology faculty designed an OSCE case in which a resident had to inform a standardized patient (SP) her father had severe global hypoxic ischemic injury. The residents were surveyed on the experience using a Likert scale from 1 (worst) to 5 (best). The SP completed a behavioral anchored checklist and marked items as "not done," "partly done," or "well done". RESULTS:57 third and fourth year neurology residents completed the case from 2018 to 2020, 54 (95%) of whom completed the post-OSCE survey. Residents reported feeling moderately prepared for the simulation (mean Likert score 3.7/5), and thought their performance was average (3.4/5). Overall, they found the case to be very helpful (4.6/5). The residents performed well in the realms of maintaining professionalism (64% rated "well done"), developing a relationship (62% rated "well done"), and information gathering (61% rated "well done"). There was room for improvement in the realms of providing education and presenting the bad news (39% and 37% rated "partly/not done," respectively). CONCLUSIONS:OSCE cases can be used to teach neurology trainees how to discuss prognosis and break bad news. Feedback about this simulation was positive, though its efficacy has yet to be evaluated and could be a future direction of study.

Fibromyalgia (FM) is characterized by widespread chronic pain, fatigue, and somatic symptoms. The influence of phenotypic changes in monocytes on symptoms associated with FM are not fully understood. The primary aim of this study was to take a comprehensive whole-body to molecular approach in characterizing relationships between monocyte phenotype and FM symptoms in relevant clinical populations. LPS-evoked and spontaneous secretion of IL-5 and other select cytokines from circulating monocytes was higher in women with FM compared to women without pain. Additionally, greater secretion of IL-5 was significantly associated with pain and other clinically relevant psychological and somatic symptoms of FM. Further, higher levels of pain and pain-related symptoms were associated with a lower percentage of intermediate monocytes (CD14/CD16) and a greater percentage of non-classical monocytes (CD14/CD16) in women with FM. Based on findings from individuals with FM, we examined the role of IL-5, an atypical cytokine secreted from monocytes, in an animal model of widespread muscle pain. Results from the animal model show that IL-5 produces analgesia and polarizes monocytes toward an anti-inflammatory phenotype (CD206). Taken together, our data suggest that monocyte phenotype and their cytokine profiles are associated with pain-related symptoms in individuals with FM. Furthermore, our data show that IL-5 has a potential role in analgesia in an animal model of FM. Thus, targeting anti-inflammatory cytokines such as IL-5 in secreted by circulating leukocytes could serve as a promising intervention to control pain and other somatic symptoms associated with FM.

BACKGROUND:Correctional facilities increasingly offer medications for opioid use disorder (OUD), including buprenorphine. Nevertheless, retention in treatment post-incarceration is suboptimal and overdose mortality remains high. Our objectives were to understand how incarcerated patients viewed buprenorphine treatment and identify modifiable factors that influenced treatment continuation post-release. METHODS:We conducted semi-structured interviews with 22 men receiving buprenorphine treatment in an urban jail. Interviews were audio recorded, professionally transcribed, and analyzed using a grounded-theory approach. Team members constructed preliminary case memos from transcripts, and then interactively discussed themes within respective memos. We established participant 'typologies' by consensus. RESULTS:Distinct typologies emerged based on participants' post-release treatment goals: (1) those who viewed buprenorphine treatment as a cure for OUD; (2) those who thought buprenorphine would help manage opioid-related problems; and (3) those who did not desire OUD treatment. Participants also described common social structural barriers to treatment continuation and community re-integration. Participants reported that post-release housing instability, unemployment, and negative interactions with parole contributed to opioid use relapse and re-incarceration. CONCLUSION:Participants had different goals for post-release buprenorphine treatment continuation, but their prior experiences suggested that social structural issues would complicate these plans. Incarceration can intensify marginalization, which when combined with heightened legal supervision, reinforced cycles of release, relapse, and re-incarceration. Participants valued buprenorphine treatment, but other structural and policy changes will be necessary to reduce incarceration-related inequities in opioid overdose mortality.

In a previous study, dasotraline demonstrated efficacy at a dose of 8 mg/day in adults with attention deficit hyperactivity disorder (ADHD). The aim of the current study was to evaluate the efficacy and safety of dasotraline in doses of 4 and 6 mg/day. Adults meeting Diagnostic and Statistical Manual of Mental Disorders, 5th edition criteria for ADHD were randomized to 8 weeks of double-blind, once-daily, fixed-dose treatment with dasotraline 4 mg/day, 6 mg/day, or placebo. The primary efficacy endpoint was changed in the ADHD Rating Scale, Version IV (ADHD RS-IV) total score. Secondary efficacy endpoints included the Clinical Global Impression, Severity (CGI-S) Scale. Least squares mean reduction at week 8 in the ADHD RS-IV HV total score was not significantly greater (vs. placebo) in the dasotraline 4 mg/day group (-15.0 vs. -13.9; n.s.; or in the dasotraline 6 mg/day group (-16.5 vs. -13.9; P = 0.074; Hochberg correction). Treatment with dasotraline 6 mg/day was significant at week 8 (uncorrected) on the ADHD RS-IV total score (P = 0.037) and the CGI-S score (P = 0.011). Treatment with the 4 mg/day dose of dasotraline was NS. Treatment with dasotraline was generally well tolerated. The results provide additional evidence that supports the potential efficacy of dasotraline, in doses of 6 mg/day, in adults with ADHD.

In this Emerging Technology Review, a novel, wireless, wearable Doppler ultrasound patch is described as a tool for resuscitation. The device is designed, foremost, as a functional hemodynamic monitor-a simple, fast, and consistent method for measuring hemodynamic change with preload variation. More generally, functional hemodynamic monitoring is a paradigm that helps predict stroke volume response to additional intravenous volume. Because Doppler ultrasound of the left ventricular outflow tract noninvasively measures stroke volume in realtime, it increasingly is deployed for this purpose. Nevertheless, Doppler ultrasound in this manner is cumbersome, especially when repeat assessments are needed. Accordingly, peripheral arteries have been studied and various measures from the common carotid artery Doppler signal act as windows to the left ventricle. Yet, handheld Doppler ultrasound of a peripheral artery is susceptible to human measurement error and statistical limitations from inadequate beat sample size. Therefore, a wearable Doppler ultrasound capable of continuous assessment minimizes measurement inconsistencies and smooths inherent physiologic variation by sampling many more cardiac cycles. Reaffirming clinical studies, the ultrasound patch tracks immediate SV change with excellent accuracy in healthy volunteers when cardiac preload is altered by various maneuvers. The wearable ultrasound also follows jugular venous Doppler, which qualitatively trends right atrial pressure. With further clinical research and the application of artificial intelligence, the monitoring modalities with this new technology are manifold.