key: cord-0762139-lsj92c9k
authors: Romagnolo, Alberto; Imbalzano, Gabriele; Artusi, Carlo Alberto; Balestrino, Roberta; Ledda, Claudia; De Rosa, Francesco Giuseppe; Riccardini, Franco; Montanaro, Elisa; Bozzali, Marco; Rizzone, Mario Giorgio; Zibetti, Maurizio; Lopiano, Leonardo
title: Neurological comorbidities and COVID-19-related case fatality: A cohort study
date: 2021-08-05
journal: J Neurol Sci
DOI: 10.1016/j.jns.2021.117610
sha: 68e4f06669fd0fcd8183ed9213e36751ee830349
doc_id: 762139
cord_uid: lsj92c9k

BACKGROUND: Neurological involvement in Coronavirus disease-2019 (COVID-19) is widely recognized. However, the role of pre-existing neurological comorbidities in modulating COVID-19-related mortality still remains unclear. This cohort study evaluates the COVID-19-related case fatality rate (CFR) of patients with pre-existing neurological diseases. METHODS: We retrospectively evaluated all patients consecutively admitted to our hospital with a diagnosis of COVID-19 between March and April 2020. We used a multivariate regression analysis to estimate the association between pre-existing neurological diseases and COVID-19-related mortality. Then, we compared the CFR and survival curves of two cohorts (patients suffering vs. those not suffering from pre-existing neurological disease), matched trough the propensity score (PS). Age and other comorbidities were considered for PS calculation. We applied a 1:1 matching for the entire neurological cohort and, separately, for cerebrovascular, neurodegenerative, and other neurological diseases. RESULTS: Among 332 patients, 75 (22.6%) were affected by pre-existing neurological disease (n = 29 cerebrovascular, n = 26 neurodegenerative, n = 20 others). From the multivariate regression analysis, they resulted with a significant increase of COVID-19-related mortality (OR:2.559; 95%CI 1.181–5.545; p < 0.017). From the cohort analysis, CFR resulted 2-fold higher in patients with neurological disease (48.0% vs. 24.0%; p = 0.002). CFR was significantly higher in patients with neurodegenerative diseases compared to matched individuals (73.9% vs. 39.1%; p = 0.017), while CFR increase in patients with cerebrovascular diseases did not reach statistical significance (48.3% vs. 41.4%; p = 0.597). CONCLUSIONS: Pre-existing neurological comorbidities, in particular neurodegenerative diseases, increase significantly COVID-19-related case fatality, indicating a clear priority for viral screening, access to care facilities and vaccination in these populations.

Since its first discovery in December 2019, Coronavirus disease-2019 (COVID- 19) has affected more than 190 million people and caused over 4 million deaths worldwide [1] .

Although the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection mostly presents with mild or no symptoms, elderly individuals and patients suffering from chronic diseases such as hypertension, diabetes, malignancy, lung disease, and renal failure, show significantly higher risk for severe infection, hospitalization, admittance to intensive care units (ICUs), and death [2] [3] [4] . Neurological symptoms have been reported in 35-50% of COVID-19 cases and found to be associated with a more severe disease [5, 6] and higher mortality [7] . Conversely, the role of pre-existing neurological conditions in modulating the severity and mortality of COVID-19 remains to be clarified. A recent study from our group [8] reported neurological comorbidities, in particular dementia and cerebrovascular diseases, as independent risk factors for a more severe course of SARS-CoV-2 infection, while another study from Spain [9] reported an association between a preexistent neurological condition and COVID-19 mortality. In addition, some studies investigating specific neurological diseases, reported that dementia [4, 6, 10] , stroke [4, [11] [12] [13] , and Parkinson's disease (PD) [14, 15] In the descriptive statistics, mean, standard deviation and range were used for continuous variables, while frequencies were used for categorical data. The Mann-Whitney and Pearson χ 2 tests were used for comparisons between groups when appropriate. A univariate binary logistic regression was used to estimate the odd ratio (OR) of death (dependent variable) entering the following demographic and clinical features as independent variables: gender, age, severe COVID-19 at onset, hospitalization regimen (i.e. no mechanical respiratory support/non-invasive mechanical support/invasive mechanical support in ICU vs. discharge at home), the presence of neurological diseases, arterial hypertension, diabetes, renal failure, COPD, neoplastic disease. This was followed by a second multivariate binary logistic regression analysis that included as independent variables every feature resulting significantly associated with death. Then, we performed a 1:1 matching between patients with and without neurological diseases, calculating the PS with a multivariable logistic regression model, which considered exclusively the variables that resulted statistically significant at the first univariate logistic regression analysis. The samples were matched via a nearest neighbor approach (individuals matched according to a similar PS); caliper matching without replacement was used, with a caliper width set at 0.15 times the standard deviation of the PS.

After creation of the PS-matched samples, we assessed the covariate balance between them, considering it satisfactory when the means of the covariates across the two groups presented with an absolute difference of <10%; continuous and categorical variables were compared using the Mann-Whitney or Pearson χ 2 tests, as appropriate.

The CFR between the matched groups was compared using the Pearson χ2 test; Kaplan-Meier survival curves and Cox regression analysis were used to evaluate differences in the survival distributions.

We applied the same approach with PS for both the entire sample of neurological patients and the different types of neurological diseases. In order to minimize the loss of statistical power, J o u r n a l P r e -p r o o f Journal Pre-proof we arbitrarily considered individuals suffering from a specific neurological disease as a separate group if their number was  20.

Patients with any missing data on their medical history or COVID-19 course were excluded from the analyses. Samuels and colleagues [17] reported a general mortality rate of 16% in patients evaluated in the Florida Memorial Healthcare System facilities, with neurological diseases associated with a 4-fold higher mortality rate; based on these observations, we estimated that a sample of 54 neurological patients would have been sufficiently large to detect an increased CFR of at least 2.5-fold in patients with neurological compared to those without a neurological diseases, with 80% power at the 5% level of significance.

All reported p-values are referred to two-tailed tests, with p<0.05 considered as cut-off for statistical significance. Data were analyzed using the Statistical Package for the Social Sciences (SPSS 26 for Mac, Chicago, IL); PS was calculated with the PS matching SPSS Rextension utility.

This study received approval from the ethical standards committee on human experimentation (Comitato Etico Interaziendale AOU Città della Salute e della Scienza di Torino, AO Ordine and clinical features of the remaining 332 patients who entered the final analyses, as well as the differences between patients with or without preexisting neurological diseases, are summarized in Table 1. J o u r n a l P r e -p r o o f Seventy-five patients (22.6%) were suffering from pre-existing neurological diseases. They were older and presented with a higher burden of comorbidity than those without pre-existing neurological conditions (Table 1 ); in particular, they had a significant higher rate of arterial hypertension, renal failure, and malignancy, while the rates of diabetes and chronic obstructive pulmonary diseases (COPD) were higher, but not statistically different between the two groups ( Table 1 ). The CFR rate in patients with neurological disease was more than 4-fold higher than that observed in patients with no neurological comorbidity.

The univariate binary logistic regression analysis ( J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f .886

J o u r n a l P r e -p r o o f [16] . Two observational studies reported a 4.6-fold [17] and a 2.9-fold [18] higher COVID-19-related mortality in neurological patients, evaluated by a multivariate approach, while other studies reported, in univariate analyses only, a significant association between mortality and neurological diseases [23] or cerebrovascular and neurodegenerative disorders [11, 12] .

Using a large health electronic analytic platform (OpenSAFELY) that includes 40% of all Importantly, after dividing the entire sample of patients with neurological comorbidity into different subgroups of neurological diseases, we found variable degrees of association with case fatality in the cohort analysis. Patients with cerebrovascular diseases showed higher CFR with no significant differences in comparison to matched patients without neurological comorbidity. This result is consistent with previous findings [7, 18] , though other Authors reported significantly higher mortality in cerebrovascular patients [6, 11, 12] , in particular in ICU populations [13] . Moreover, a recent meta-analysis demonstrated the role of cerebrovascular diseases in determining poor outcomes in COVID-19 hospitalized patients [24] . The lack of significant difference in the CFR between patients with and without cerebrovascular diseases in our study is likely due to the coexistence, in both groups, of the interaction between an already damaged nervous system and a virus with proven neurotropism [30] might promote a more severe form of infection and poorer outcomes, as previously observed in the SARS-CoV epidemic in 2003 [31] . Our robust result needs to be further investigated in tailored studies to assess biological and social causes of this increased risk of mortality, whose importance goes well beyond our understanding of the effects of COVID-19 pandemic.

In our sample, other neurological comorbidities were not associated with an increased risk of COVID-19-related mortality. This latter subgroup of patients had indeed a similar CFR (15%) in comparison to patients without neurological comorbidity, when considering both the entire non-neurological sample (10.9%) and matched patients in the cohort analysis (20%). In particular, no patients with chronic headache/neuralgia died, while the CFR of epileptic patients was similar to that reported in previous studies [32, 33] .

The main limitations of our study are represented by the relatively small sample size, in particular when considering specific types of neurological diseases, and the single-center retrospective design, which partially restricts the generalizability of our findings. Nonetheless, our statistical design has controlled for major potential biases that may affect this kind of studies and result in uncertain conclusions. Moreover, some important comorbidities, such as

Dr Romagnolo has received grant support and speaker honoraria from AbbVie, speaker honoraria from Chiesi Farmaceutici and travel grants from Lusofarmaco, Chiesi Farmaceutici, Medtronic, and UCB Pharma.

Dr Imbalzano has no financial conflicts to disclose.

Dr Artusi has received travel grants from Zambon and Abbvie.

Dr Balestrino has no financial conflicts to disclose.

Dr Ledda has no financial conflicts to disclose.

Prof De Rosa has no financial conflicts to disclose.

Dr Riccardini has no financial conflicts to disclose.

Dr Montanaro has received travel grant from Ralpharma.

Prof Bozzali is member of the Advisory board of Roche Pharmaceuticals and has received honoraria for lecturing from Biogen Pharma.

Dr Rizzone has received grant support and speaker honoraria from Medtronic and UCB.

Dr Zibetti has received honoraria from Medtronic, Zambon Pharma and AbbVie.

Prof Lopiano has received honoraria for lecturing and travel grants from Medtronic, UCB Pharma, and AbbVie.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors

A. Romagnolo had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

J o u r n a l P r e -p r o o f

Patients with other neurological diseases showed a CFR similar to that of patients without neurological diseases (15.0% vs. 20.0%; p=0.677), as confirmed by Cox regression analysis

World Health Organization (WHO) Coronavirus Disease (COVID-19) Dashboard

Risk factors associated with acute respiratory distress syndrome and death in patients with Coronavirus Disease

Risk factors associated with mortality among patients with COVID-19 in Intensive Care Units in Lombardy

Factors associated with COVID-19-related death using OpenSAFELY

Neurologic manifestations of hospitalized patients with Coronavirus Disease

Neurological features and outcome in COVID-19: dementia can predict severe disease

Neurologic syndromes predict higher in-hospital mortality in COVID-19

Neurological comorbidity and severity of COVID-19

Neurological comorbidity is a predictor of death in Covid-19 disease: a cohort study on 576 patients

Predictors of mortality of COVID-19 in the general population and nursing homes

Older age and comorbidity are independent mortality predictors in a large cohort of 1305 COVID-19 patients in Michigan, United States

Mortality risk assessment in Spain and Italy, insights of the HOPE COVID-19 registry

The swedish covid-19 intensive care cohort: risk factors of ICU admission and ICU mortality

COVID-19 in Parkinson's disease: report on prevalence and outcome

Risk of hospitalization and death for COVID-19 in people with Parkinson's disease or parkinsonism

Characteristics and risk factors for hospitalization and mortality among persons with COVID-19 in Atlanta metropolitan area

The epidemiology and predictors of outcomes among confirmed COVID-19 cases in a large community healthcare system in South Florida

Predictors at admission of mechanical ventilation and death in an observational cohort of adults hospitalized with COVID-19

Comorbidities in SARS-CoV-2 patients: a systematic review and meta-analysis

Diagnosis and treatment of adults with community-acquired pneumonia. An official clinical practice guideline of the

Methods for constructing and assessing propensity scores

An introduction to Propensity Score methods for reducing the effects of confounding in observational studies

Patient characteristics and predictors of mortality in 470 adults admitted to a district general hospital in England with Covid-19

Pre-existing cerebrovascular disease and poor outcomes of COVID-19 hospitalized patients: a meta-analysis

Case fatality rate of SARS-CoV-2 infection at regional level and across different phases of the epidemic in Italy

Cognitive and neuropsychiatric manifestations of COVID-19 and effects on elderly individuals with dementia

Chronic neurological disorders and related comorbidities: Role of age-associated physiological changes

Outcomes of dementia: Systematic review and meta-analysis of hospital administrative database studies

A systematic review and meta-analysis examining pneumonia-associated mortality in dementia

Evidence of the COVID-19 Virus Targeting the CNS: Tissue Distribution, Host-Virus Interaction, and Proposed Neurotropic Mechanisms

Severe acute respiratory syndrome coronavirus infection causes neuronal death in the absence of encephalitis in mice transgenic for human ACE2

Incidence and case fatality rate of COVID-19 in active epilepsy

Clinical characteristics of 30 COVID-19 patients with epilepsy: A retrospective study in Wuhan

The data that support the findings of this study are available from the corresponding author, upon reasonable request.J o u r n a l P r e -p r o o f