Risk Factors for New Neurologic Diagnoses in Hospitalized Patients With COVID-19: A Case-Control Study in New York City

Abstract

Background and objectives: There have been numerous reports of neurologic manifestations identified in hospitalized patients infected with SARS-CoV-2, the virus that causes COVID-19. Here, we identify the spectrum of associated neurologic symptoms and diagnoses, define the time course of their development, and examine readmission rates and mortality risk posthospitalization in a multiethnic urban cohort.

Methods: We identify the occurrence of new neurologic diagnoses among patients with laboratory-confirmed SARS-CoV-2 infection in New York City. A retrospective cohort study was performed on 532 cases (hospitalized patients with new neurologic diagnoses within 6 weeks of positive SARS-CoV-2 laboratory results between March 1, 2020, and August 31, 2020). We compare demographic and clinical features of the 532 cases with 532 controls (hospitalized COVID-19 patients without neurologic diagnoses) in a case-control study with one-to-one matching and examine hospital-related data and outcomes of death and readmission up to 6 months after acute hospitalization in a secondary case-only analysis.

Results: Among the 532 cases, the most common new neurologic diagnoses included encephalopathy (478, 89.8%), stroke (66, 12.4%), and seizures (38, 7.1%). In the case-control study, cases were more likely than controls to be male (58.6% vs 52.8%, p = 0.05), had baseline neurologic comorbidities (36.3% vs 13.0%, p < 0.0001), and were to be treated in an intensive care unit (62.0% vs 9.6%, p < 0.0001). Of the 394 (74.1%) cases who survived acute hospitalization, more than half (220 of 394, 55.8%) were readmitted within 6 months, with a mortality rate of 23.2% during readmission.

Discussion: Hospitalized patients with SARS-CoV-2 and new neurologic diagnoses have significant morbidity and mortality postdischarge. Further research is needed to define the effect of neurologic diagnoses during acute hospitalization on longitudinal post-COVID-19-related symptoms including neurocognitive impairment.

Figure 1. Case Inclusion

Central boxes with dashed lines represent sources from which cases and controls were identified, including all Neurology units and the COVID-CARES database established at CUIMC/NYP. The workflow on the left details exclusion process for cases; the right side details exclusion process for controls. Bold outlined boxes indicate final sample size for cases and controls. cEEG = continuous electroencephalography; COVID-19 = coronavirus disease 2019; COVID-CARES = Comprehensive database of COVID-19 infectious cases at CUIMC; EMU = epilepsy monitoring unit; LP = lumbar puncture; NICU = neurologic intensive care unit; RT-PCR = reverse transcriptase PCR; SARS-CoV-2 = severe acute respiratory syndrome coronavirus 2.

Figure 2. Time of First Recorded COVID RT-PCR–Positive Test to First Neurologic Diagnosis

Cases with neurologic diagnoses established on the same day of their first positive COVID PCR test (usually coinciding with day of presentation to hospital) are indicated by the bar at zero days. Cases with established primary neurologic diagnoses before their first COVID PCR positivity result are represented by units to the left of zero days. Cases with established primary neurologic diagnoses after their first COVID-positive PCR test are indicated to the right of zero days. All cases are color coded by diagnosis. The blue and green dashed lines indicate median and mean number of days between first COVID-positive PCR test and primary neurologic diagnosis, respectively. RT-PCR = reverse transcriptase PCR.