key: cord-0765235-y45so2tj authors: Ragheb, J.; McKinney, A.; Zierau, M.; Brooks, J.; Hill-Caruthers, M.; Iskander, M.; Ahmed, Y.; Lobo, R.; Vlisides, P. E. title: Delirium and Post-Discharge Neuropsychological Outcomes in Critically Ill Patients with COVID-19: an Institutional Case Series date: 2020-11-04 journal: nan DOI: 10.1101/2020.11.03.20225466 sha: 8469cbf5adea6df266389cd307aea9e085596137 doc_id: 765235 cord_uid: y45so2tj Delirium is a serious and common complication among critically ill patients with COVID-19. The objective of this study was to characterize the clinical course of delirium for COVID-19 patients in the intensive care unit, including post-discharge cognitive outcomes. A retrospective chart review was conducted for patients diagnosed with COVID-19 (n=148) admitted to an intensive care unit at Michigan Medicine between 3/1/2020 and 5/31/2020. Delirium was identified in 107/148 (72%) patients in the study cohort, with median (interquartile range) duration lasting 10 (4 - 17) days. Sedative regimens, inflammation, deviation from delirium prevention protocols, and hypoxic-ischemic injury were likely contributing factors, and the most common disposition for delirious patients was a skilled care facility (41/148, 38%). Among patients who were delirious during hospitalization, 4/17 (24%) later tested positive for delirium at home based on caretaker assessment, 5/22 (23%) demonstrated signs of questionable cognitive impairment or cognitive impairment consistent with dementia, and 3/25 (12%) screened positive for depression within two months after discharge. Overall, patients with COVID-19 commonly experience a prolonged course of delirium in the intensive care unit, likely with multiple contributing factors. Furthermore, neuropsychological impairment may persist after discharge. The outbreak of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), the virus that causes Coronavirus Disease (COVID-19), emerged as a public health threat in December 2019 and was declared a pandemic by World Health Organization in March 2020. Several studies have reported the serious nature of complications associated with this disease. Major neurological complications, such as encephalopathy, delirium, strokes, seizures, and ataxia, have all been observed. [1] [2] [3] [4] [5] Delirium appears to be a common complication, with previous investigations demonstrating an incidence of approximately 65-80% in the intensive care unit (ICU). 1, 4 Delirium may occur due to direct coronavirus invasion into central nervous system, 6 and systemic inflammatory responses may further exacerbate neurocognitive impairment. In the ICU, synergistic factors such as sedation regimen, social isolation, and deviation from standard care protocols may further increase risk. Delirium is also associated with prolonged hospitalization, long-term cognitive and functional impairment, and increased mortality. [7] [8] [9] As such, there is a critical need to advance understanding of this syndrome in patients with COVID-19. While a high incidence of delirium has been previously reported in COVID-19 patients, fundamental questions persist. The clinical course of delirium, including average duration and post-discharge cognitive trajectory, remains unknown. Pathophysiologic drivers of delirium (e.g., polypharmacy, inflammation, cerebrovascular events) are incompletely understood, and the extent to which standard prevention protocols are implemented is unclear. Such detailed understanding will contribute to delirium All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint phenotyping of COVID-19 patients and provide insight into the clinical and neurocognitive burden associated with COVID-19. In this context, the objective of this study was to determine granular details associated with delirium in ICU patients with COVID-19. Specifically, the clinical course of delirium, presence of exacerbating factors, nature of prevention strategy implementation, and post-discharge cognitive outcomes were all characterized. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint Baseline characteristics are presented in Table 1 . The majority of patients were African-American and non-Hispanic, and the most common comorbidities were hypertension, Diabetes mellitus, and obesity. Overall, age, demographics, and comorbidity profiles were similar between patients with delirium and those who did not experience delirium during hospitalization. Delirium incidence was high in the cohort (107/148, 72%), and median (interquartile range) duration was 10 (4 -17) days ( Table 2 ). The most common delirium prevention measure, based on the ABCDEF ICU liberation bundle, 10, 11 was assessment and treatment for pain; bringing familiar objects from home and spontaneous awakening trials were the least common (Table 2) . Overall, the total number ICU liberation bundle activities charted was fewer than expected for the median duration of ICU admission (see Table 2 legend for description of prevention protocol activity schedule). New antidepressant use was more common for those with delirium (26/107, 24%) compared to those without delirium (4/41, 9.8%; p=0.049). Similarly, a psychiatry consult was obtained for 21/107 (20%) delirious patients compared to 0/41 (0%) in the non-delirium group (p=0.002). Lastly, no evidence of reversal or improvement was reported for nearly 30% of patients during index hospitalization. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint In terms of hospitalization (Table 3) , median length of hospitalization was 25 (13 -48) days, and median length of ICU stay 15 (7 -31) . Length of hospitalization, ICU length of stay, and duration of mechanical ventilation were all significantly prolonged in patients experiencing delirium (Table 3) . Correspondingly, sedative-hypnotic use was higher in patients with delirium. Delirious patients demonstrated higher white blood cell counts and higher levels of c-reactive protein, lactate dehydrogenase, and d-dimer compared to non-delirious patients. Less than half of patients were ultimately discharged home, and the most common disposition for those with delirium was a skilled care facility (41/107, 38%) after discharge (Table 3) . Neuropsychological outcomes after discharge are reported in Table 4 . Among patients who were still alive and available to complete survey materials, nearly 25% of patients In total, 47 patients underwent neuroimaging during hospitalization. The majority of imaging results were unremarkable or demonstrated incidental findings unrelated to All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint COVID-19. However, some notable findings were present. A brain MRI was ordered for a 59-year-old female with COVID-19 pneumonia and recent ECMO decannulation due to severe encephalopathy (i.e., no response to commands or noxious stimulus) with preservation of brainstem reflexes. Imaging revealed abnormal fluid attenuated inversion recovery (FLAIR) hyperintensity affecting the occipital and temporal lobes (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint In a cohort of ICU patients with COVID-19, delirium was a common complication, affecting more than 70% of patients. Furthermore, delirium was associated with prolonged hospitalization, increased length of ICU stay, discharge to skilled care facilities, and positive screens for neuropsychological impairment up to two months after discharge. Delirium occurred in the setting of multiple sedative-hypnotic agents, acute inflammatory responses, deviation from delirium prevention protocols, and cerebrovascular events, which are all factors that could have further catalyzed delirium precipitation. While ICU liberation activities were conducted, many of the interventions were infrequently implemented, particularly those that reduce delirium risk (e.g., spontaneous awakening trials). Overall, the burden of cognitive impairment was high in patients with COVID-19, as was the risk of related complications. These results align with previous data demonstrating a high incidence of delirium in critically ill patients with COVID-19. [1] [2] [3] [4] Cognitive dysfunction may occur as a result of direct coronavirus invasion of the central nervous system 6 or other indirect mechanisms, such as polypharmacy, systemic inflammatory responses, or cerebrovascular events. Indeed, benzodiazepine sedation was common in this patient cohort, with nearly 60% of patients receiving midazolam sedation at one point during ICU admission. Lorazepam was a common sedation agent as well, and benzodiazepine use is associated with delirium in critically ill patients. [12] [13] [14] Inflammation may have also contributed to delirium risk. Inflammatory markers (e.g., c-reactive protein, ferritin, interleukin-6, lactate dehydrogenase) were considerably elevated in this patient cohort. In fact, serum levels All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint observed in this study aligned with -or exceeded -previously reported values in patients with severe COVID-19, 2,5 and there was MRI evidence of neuroinflammation for at least two patients in this series. Specific inflammatory markers (e.g., c-reactive protein, lactate dehydrogenase) were elevated in delirium patients and may reflect a specific in neuroinflammatory process. Indeed, c-reactive protein increases blood-brain barrier permeability in basic science models, 15 and lactate dehydrogenase has been associated with neurocognitive disorders. 16, 17 However, these were unadjusted, bivariable analyses, and further covariate adjustment may have revealed a reduced -or absent -association. 5 Lastly, cerebral ischemia may also contribute to delirium risk in patients with COVID-19. Severe hypoxic-ischemic injury occurred in a patient who experienced multiple cardiopulmonary arrests during the course of illness. Stroke has previously been reported in patients with COVID-19, 18 as thromboembolic phenomena and cerebral malperfusion may both occur during the clinical course of COVID-19. As such, multiple processes likely contribute to delirium in patients with COVID-19. Targeted case-control studies can address some of these potential risk factors, such sedative-hypnotic regimens and inflammatory profiles. Delirium prevention and management are inherently challenging for COVID-19 patients in the ICU. While delirium prevention bundles have been consistently demonstrated to reduce risk, 19,20 unique challenges posed by COVID-19 hinder the implementation of standard prevention practices. Spontaneous awakening and breathing trials, for example, were often not possible due to illness severity and associated ventilator requirements. Indeed, the median number of total spontaneous awakening trials charted All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint was two across the entire cohort. Clinicians may have also been limited in terms of sedation regimen. Agitation was commonly observed, and nearly one-third of patients required antipsychotics. Agitation and hyperactive delirium likely prompted additional sedation and prolonged use of physical restraints. Early mobility was limited given illness severity, and family engagement was often not possible due to visitor policy restrictions. In-person interactions with clinicians were also limited given the intent of reducing virus transmission. As such, the culmination of disease severity, limited faceto-face time spent with patients, and visitor restriction policies likely hindered ICU liberation bundle implementation. Indeed, given that the median length of ICU stay was 15 days across the cohort, most interventions were charted less than once daily ( Table 2 ). Limited implementation of delirium prevention bundles could have increased risk for delirium incidence, duration, and severity. Interestingly, prevention bundle activities were charted more commonly in delirious patients. This may have reflected clinician awareness of delirium and attempts to reduce risk. Novel strategies for implementing delirium prevention bundles in this patient population may help to further mitigate risk and should be tested in prospective trials. Neuropsychological impairment after discharge was also present for some patients based on subjective reporting, caretaker assessment, and objective testing for depression and cognitive impairment. Furthermore, all patients that screened positive for possible impairment also experienced delirium in the hospital. These estimates may have been even higher, given that approximately 70% patients that were called for postdischarge cognitive assessment were still admitted to skilled care facilities, refused All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint participation, or were unable to be reached. Whether post-discharge cognitive impairment was related specifically to COVID-19 or critical illness more broadly is unclear. Indeed, cognitive impairment is common at discharge for patients who experienced delirium while in the ICU, and delirium is present for nearly 20% of patients newly admitted to acute care facilities. 7, 21 Moreover, cognitive impairment can be present for months-to-years after acute respiratory distress syndrome and sepsis, [22] [23] [24] and symptoms of depression and post-traumatic stress disorder are commonly reported among ICU survivors. 25 Neuropsychological impairment after discharge may, in part, reflect critical illness, rather than pathophysiologic insults specific to COVID-19. Nonetheless, ICU patients with COVID-19 experience considerable neuropsychological burden, and related complications, both during and after hospitalization. The strengths of this study include granular data with respect to delirium, potential risk factors, nature and frequency of prevention strategies, related complications, and postdischarge outcomes. Data were representative of an academic tertiary care center with nearly 150 patients. A validated chart review method was used to identify delirium, 26 and the study measures used to characterize cognitive function, such as the FAM-CAM, Short Blessed Test, and PROMIS assessments, are validated measures that increase confidence in the results. In terms of limitations, this is a this was a single center analysis, and the results are restricted to the institution studied. The study was not conducted with a matched control cohort, as the nature of this study was descriptive. Lastly, data were limited for post-discharge cognitive outcomes, as more than half of patients called were unavailable to complete assessments. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint In summary, delirium is common complication of COVID-19 with multiple contributing factors. Furthermore, neuropsychological impairment may persist in some patients after discharge. Further research should aim to identify independent risk factors in this population and novel, effective prevention strategies. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint This was a single-center, institutional case series from Michigan Medicine. Detailed chart review data were collected from critically ill patients with COVID-19 (3/1/2020 -5/31/2020), and post-discharge telephone surveys were conducted to determine if cognitive impairment persisted after discharge. All study operations were conducted at Michigan Medicine, Ann Arbor MI USA, and approval was obtained from the University of Michigan Medical School Institutional Review Board (HUM00182646). All patients with a COVID-19 diagnosis admitted to a Michigan Medicine ICU between 03/01/2020 -05/31/2020 were eligible for study inclusion. ICU patients admitted during this time, without a diagnosis of COVID-19, were not eligible for study inclusion. The primary outcome was delirium presence (yes/no, %) at any point during admission. Delirium was evaluated via chart review method (described below). Several secondary outcomes were also collected in relation to delirium and overall clinical trajectory. These outcomes included the following: duration of delirium (days), antipsychotic administration, length of hospital stay, length of ICU stay, number of days requiring ventilator support, inflammatory laboratory values (white blood cell count, c-reactive protein, ferritin, lactate dehydrogenase, d-dimer, and interleukin-6), new psychiatry consults, new antidepressant use, and final disposition (e.g., home, long-term care All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint facility, death). Delirium prevention strategies, based on the ABCDEF ICU liberation bundle, 10, 11 were also recorded. These included the following: assess and treat for pain, discontinue physical restraints, structured mobility exercises, removal of temporary medical lines or devices, place familiar objects from home at the bedside, family education and reassurance, promote use of visual and hearing aids, sleep promotion protocols (e.g., lights off at night, melatonin tablets), and spontaneous awakening trials. The total number of times a prevention strategy was charted was recorded for each patient. The median of these total values was then reported for each prevention strategy. Neuroimaging data were also collected and reviewed. Lastly, a telephone survey was conducted between 30-60 days post-discharge to determine whether subjective or objective signs of cognitive impairment were present. During telephone interviews, the following tests were conducted: the Patient-Reported (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint teams to retrospectively search for patient cohorts. Charts that screened positive were then manually reviewed by study team members to confirm study eligibility. Charts were then reviewed in further detail for outcome abstraction. Delirium was assessed via a validated chart review method. 26 Briefly, any instance of an acute confusional state was recorded in the instrument. The source of information was recorded, along with the date and time. The total number of days with acute confusion was also included in the instrument, along with any evidence of reversibility or improvement of the confusion state. Other clinical outcomes, along with laboratory values, were collected directly from the charts. Neuroimaging studies were manually reviewed by a board certified radiologist with a Certificate of Added Qualification in neuroradiology (R.L.). All analyses were performed using IBM SPSS version 27 (Armonk, NY USA). Exploratory data analysis techniques were used to assess the distribution of dependent measures for determining the appropriate analytical strategy. The Shapiro-Wilk test was used to assess the distribution of continuous outcomes, and Independent t-tests or Mann-Whitney U tests were used as appropriate. Mean (standard deviation) or median (interquartile range) was reported for parametric and non-parametric data, respectively. For binary outcomes and proportions, The Chi-Square Test or Fisher's Exact Test were used, as appropriate. The threshold for significance was set to p<0.05 across all tests. For post-discharge cognitive outcomes, descriptive statistics were reported. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint Ntaios, G. et al. Characteristics and outcomes in patients with COVID-19 and acute ischemic stroke: The global COVID-19 stroke registry. Stroke. 51, e254-e258 (2020). All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint Table 3 Neurologic features in severe SARS-CoV-2 infection Neurologic manifestations of hospitalized patients with coronavirus disease 2019 in Wuhan, China Neurological and neuropsychiatric complications of COVID-19 in 153 patients: a UK-wide surveillance study Delirium and encephalopathy in severe COVID-19: a cohort analysis of ICU patients Frequent neurologic manifestations and encephalopathy-associated morbidity in COVID-19 patients Coronavirus infection of the central nervous system: Host-virus stand-off Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit Delirium as a predictor of long-term cognitive impairment in survivors of critical illness Delirium in the ICU and subsequent long-term disability among survivors of mechanical ventilation The ABCDEF bundle in critical care No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Liberation and animation for ventilated ICU patients: The ABCDE bundle for the back-end of critical care Benzodiazepine and opioid use and the duration of intensive care unit delirium in an older population Benzodiazepine-associated delirium in critically ill adults Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients C-reactive protein increases BBB permeability: Implications for obesity and neuroinflammation Serum iron parameters, HFE C282Y genotype, and cognitive performance in older adults: Results from the FACIT study Redox proteomic identification of 4-hydroxy-2-nonenal-modified brain proteins in amnestic mild cognitive impairment: Insight into the role of lipid peroxidation in the progression and pathogenesis of Alzheimer's disease We would like to acknowledge Dr. Michael Kenes (PharmD, BCPS, BCCCP) and Ms.Margaret Diehl for assistance with medical chart data extraction. The study was originally conceived by J. Competing Interests Statement: The authors declare no competing interests. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint New psychiatry consults, n (%) 21 (14) 21 (20) 0 (0) 0.002Delirium prevention measures are based on the standard ICU liberation bundle protocols (see text for details). Per institutional protocol, clinicians conduct pain assessments every two hours (and more frequently as needed for procedures), range of motion exercises are scheduled three times daily, spontaneous awakening/breathing trials occur daily (if medically appropriate), and family updates are scheduled once daily. Given this schedule, these prevention measures were charted less frequently than expected based on a median ICU course of 15 (7 -31) days. IQR interquartile range.All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint Reasons for not completing a test included the following: patient deceased, patient remains admitted to an inpatient facility, patient (or family member) refused survey participation, or patient was unable to be reached.All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted November 4, 2020. ; https://doi.org/10.1101/2020.11.03.20225466 doi: medRxiv preprint