Faculty Bibliography

Background Following the emergence of coronavirus disease 2019 (COVID-19), early reports suggested a decrease in stroke and acute coronary syndrome (ACS). We sought to provide descriptive statistics for stroke and ACS from a sample of hospitals throughout the United States, comparing data from March 2020 to similar months pre-COVID. Methods We performed a retrospective analysis of 65 academic and community hospitals in the Vizient Clinical Data Base. The primary outcome is monthly count of stroke and ACS, and acute procedures for both, from February and March in 2020 compared to the same months in 2018 and 2019. Results are aggregated for all hospitals and reported by Census Region. Results We identified 51,246 strokes (42,780 ischemic, 8,466 hemorrhagic), 1,043 mechanical thrombectomies (MT), 836 tissue plasminogen activator (tPA) administrations, 36,551 ACS, and 3,925 percutaneous coronary interventions (PCI) for ACS. In February 2020, relative to February 2018 and 2019, hospitalizations with any discharge diagnosis of stroke and ACS increased by 9.8% and 12.1%, respectively, while in March 2020 they decreased 18.5% and 7.5%, relative to March 2018 and 2019. When only including hospitalizations with the primary discharge diagnosis of stroke or ACS, in March 2020 they decreased 17.6% and 25.7%, respectively. In March 2020, tPA decreased 3.3%, MT increased 18.8%, although in February 2020 it had increased 36.8%, and PCI decreased 14.7%. These decreases were observed in all Census regions. Conclusions Following greater recognition of the risks of COVID-19, hospitalizations with stroke and ACS were markedly diminished in a geographically diverse sample of United States hospitals. Because the most likely explanation is that some patients with stroke and ACS did not seek medical care, the underlying reasons for this decrease warrant additional study to inform public health efforts and clinical care during this and future pandemics.

OBJECTIVE:Postencephalitic epilepsy is often resistant to antiseizure medications, leading to evaluation for epilepsy surgery. Characterizing its localization carries implications for optimal surgical approach. We aimed to determine whether a prior history of encephalitis is associated with specific epileptogenic networks among patients with drug resistant epilepsy undergoing stereotactic EEG (SEEG). METHODS:We conducted a retrospective cohort study of drug resistant epilepsy, with and without a prior history of encephalitis. We analyzed SEEG recordings to identify patterns of seizure onset and organization. Seventeen patients with a history of encephalitis (of infectious etiology in two subjects) were identified from a database of patients undergoing SEEG and were compared to seventeen drug-resistant epilepsy controls without a history of encephalitis matched for confounding variables including pre-implantation hypotheses, epilepsy duration, age, and sex. RESULTS:Independent bilateral seizures were noted in 65% of the postencephalitic epilepsy cohort. We identified four SEEG-ictal patterns in patients with a prior history of encephalitis: (1) anteromesial temporal onset (24%), (2) anteromesial temporal onset with early spread to the perisylvian region (29%), (3) perisylvian (59%) and (4) synchronized anteromesial temporal and perisylvian (29%) onsets. Patterns 3 and 4, with perisylvian involvement at onset, were unique to the encephalitis group (p = 0.0003 and 0.04 respectively) and exhibited a "patchwork" organization. None of the encephalitis patients vs 5/7 matched controls had Engel I outcome (p = 0.0048). CONCLUSIONS:Postencephalitic epilepsies involve anteromesial temporal and perisylvian networks, often in a bilateral independent manner. Unique ictal patterns involving the perisylvian regions was identified in the encephalitis group, but not in the matched control group. SIGNIFICANCE/CONCLUSIONS:These findings may reflect a selective vulnerability of the perisylvian regions to epilepsy resulting from encephalitis, significantly mitigating the chances of success with SEEG-guided temporal resections.

Epilepsy surgery is considered to reduce the risk of epilepsy-related mortality, including sudden unexpected death in epilepsy (SUDEP), though data from existing surgical series are conflicting. We retrospectively examined all-cause mortality and SUDEP in a population of 590 epilepsy surgery patients and a comparison group of 122 patients with pharmacoresistant focal epilepsy who did not undergo surgery, treated at Columbia University Medical Center between 1977 and 2014. There were 34 deaths in the surgery group, including 14 cases of SUDEP. Standardized mortality ratio (SMR) for the surgery group was 1.6, and SUDEP rate was 1.9 per 1000 patient-years. There were 13 deaths in the comparison group, including 5 cases of SUDEP. Standardized mortality ratio for the comparison group was 3.6, and SUDEP rate was 4.6 per 1000 patient-years. Both were significantly greater than in the surgery group (p < 0.05). All but one of the surgical SUDEP cases, and all of the comparison group SUDEP cases, had a history of bilateral tonic-clonic seizures (BTCS). Of postoperative SUDEP cases, one was seizure-free, and two were free of BTCS at last clinical follow-up. Time to SUDEP in the surgery group was longer than in the comparison group (10.1 vs 5.9 years, p = 0.013), with 10 of the 14 cases occurring >10 years after surgery. All-cause mortality was reduced after epilepsy surgery relative to the comparison group. There was an early benefit of surgery on the occurrence of SUDEP, which was reduced after 10 years. A larger, multicenter study is needed to further investigate the time course of postsurgical SUDEP.

BACKGROUND:We have described the clinical stages of the brain aging and Alzheimer's disease (AD) continuum. In terms of the pre-dementia stages of AD, we introduced the terminology "mild cognitive impairment" (MCI) for the first pre-dementia stage and "subjective cognitive decline" (SCD) for the pre-MCI stage. We now report the characteristics of a pre-SCD condition eventuating in likely AD. OBJECTIVE:The aim of this study was to characterize a pre-SCD condition eventuating in AD. METHOD/METHODS:Sixty healthy persons with "no cognitive decline" (NCD) were recruited and 47 were followed (mean baseline age, 64.1 ± 8.9 years; mean follow-up time, 6.7 ± 3.1 years). Outcome was determined at the final assessment prior to 2002 as "decliner," if SCD or worse, or "nondecliner" if NCD. RESULTS:After controlling for age, gender, years of education, and follow-up time, there was a between-group difference in the decline rate (p < 0.001). Also, after controlling for demographic variables and follow-up time, the combinatorial psychometric score was lower at baseline in the future decliners (p = 0.035). Of the 9 psychometric variables, after controlling for demographic variables and follow-up time, 3 were significantly lower at baseline in future decliners. Since AD is known to be age related and all subjects in this study were otherwise healthy, we also did an analysis without controlling for age. The combinatorial psychometric score was highly significantly better at baseline in the future nondecliners than in the future decliners (p = 0.008). CONCLUSION/CONCLUSIONS:This is ostensibly the first study to link psychometric cognitive decline to the subsequent SCD stage of eventual AD.

As all other aspects in times of the coronavirus disease (COVID)-19 pandemic, carrying-out quality clinical research has been challenging. Many well-established paradigms have shifted as a consequence of the rapid demand for new knowledge. New treatments are fast-moving, informed consent forms are difficult to obtain, a competitive invitation from researchers to participate in different studies is common, and non-COVID-19 research protocols are suffering continuity. However, these challenges should not imply taking shortcuts or accepting deficiencies in bioethical standards, but rather enhance the alertness for rigorous ethical approaches despite these less than ideal circumstances. In this manuscript, we point out some interrogates in COVID-19 research and outline possible strategies to overcome the difficult task to continue with high-quality research without violating the ethical principles.

BACKGROUND:Coronavirus (CoV) disease (COVID)-19 poses difficult situations in which the ethical course of action is not clear, or choices are made between equally unacceptable responses. METHODS:A web search was performed using the terms "bioethics; COVID-19; ethics; severe acute respiratory syndrome CoV-2; emergent care; pandemic; and public health emergencies." RESULTS:Protection from COVID-19 has resulted in the cancellation of necessary medical interventions, lengthened suffering, and potential non-COVID-19 deaths. Prolonged lockdown reduced well-being, triggering or aggravating mental illnesses and violence, and escalated medical risks. Collateral damage includes restrictions on visitations to hospitals, alienation from the deceased relative, or lack of warm caring of patients. Finally, in a public health crisis, public health interest overrides individual rights if it results in severe harm to the community. CONCLUSION:Balancing ethical dilemmas are one more challenge in the COVID-19 pandemic.

Background There is growing recognition that positive health beliefs may promote blood pressure (BP) reduction, which is critical to stroke prevention but remains a persistent challenge. Yet, studies that examine the association between positive health beliefs and BP among stroke survivors are lacking. Methods and Results Data came from the DESERVE (Discharge Educational Strategies for Reduction of Vascular Events) study, a randomized controlled trial of a skills-based behavioral intervention to reduce vascular risk in a multiethnic cohort of 552 transient ischemic attack and mild/moderate stroke patients in New York City. The exposure was perception that people can protect themselves from having a stroke (ie, prevention self-efficacy) at baseline. The association between systolic BP (SBP) reduction at 12-month follow-up and self-efficacy was examined using linear regression adjusted for key confounders, overall and stratified by age, sex, race/ethnicity, and intervention trial arm. Approximately three quarters endorsed self-efficacy. These participants had, on average, 5.6 mm Hg greater SBP reduction compared with those who did not endorse it (95% CI, 0.5-10.7 mm Hg; P=0.032). Self-efficacy was significantly associated with greater SBP reduction, particularly among female versus male, younger versus older, and Hispanic versus non-Hispanic white patients. Sensitivity analysis adjusting for baseline SBP instead of elevated BP yielded no association between self-efficacy and SBP reduction, but showed sex differences in this association (women: β=5.3; 95% CI, -0.2 to 10.8; P=0.057; men: β=-3.3; 95% CI, -9.4 to 2.9; P=0.300; interaction P=0.064). Conclusions Self-efficacy was linked with greater SBP reduction among female stroke survivors. Targeted strategies to improve health beliefs after stroke may be important for risk factor management. REGISTRATION URL: https://www.clini​caltr​ials.gov; Unique identifier: NCT01836354.

The offline "replay" of neural firing patterns underlying waking experience, previously observed in non-human animals, is thought to be a mechanism for memory consolidation. Here, we test for replay in the human brain by recording spiking activity from the motor cortex of two participants who had intracortical microelectrode arrays placed chronically as part of a brain-computer interface pilot clinical trial. Participants took a nap before and after playing a neurally controlled sequence-copying game that consists of many repetitions of one "repeated" sequence sparsely interleaved with varying "control" sequences. Both participants performed repeated sequences more accurately than control sequences, consistent with learning. We compare the firing rate patterns that caused the cursor movements when performing each sequence to firing rate patterns throughout both rest periods. Correlations with repeated sequences increase more from pre- to post-task rest than do correlations with control sequences, providing direct evidence of learning-related replay in the human brain.

Deep learning (DL) methods have in recent years yielded impressive results in medical imaging, with the potential to function as clinical aid to radiologists. However, DL models in medical imaging are often trained on public research cohorts with images acquired with a single scanner or with strict protocol harmonization, which is not representative of a clinical setting. The aim of this study was to investigate how well a DL model performs in unseen clinical datasets-collected with different scanners, protocols and disease populations-and whether more heterogeneous training data improves generalization. In total, 3117 MRI scans of brains from multiple dementia research cohorts and memory clinics, that had been visually rated by a neuroradiologist according to Scheltens' scale of medial temporal atrophy (MTA), were included in this study. By training multiple versions of a convolutional neural network on different subsets of this data to predict MTA ratings, we assessed the impact of including images from a wider distribution during training had on performance in external memory clinic data. Our results showed that our model generalized well to datasets acquired with similar protocols as the training data, but substantially worse in clinical cohorts with visibly different tissue contrasts in the images. This implies that future DL studies investigating performance in out-of-distribution (OOD) MRI data need to assess multiple external cohorts for reliable results. Further, by including data from a wider range of scanners and protocols the performance improved in OOD data, which suggests that more heterogeneous training data makes the model generalize better. To conclude, this is the most comprehensive study to date investigating the domain shift in deep learning on MRI data, and we advocate rigorous evaluation of DL models on clinical data prior to being certified for deployment.