Introduction

The emergence of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), responsible for the COVID-19 pandemic, has been associated with a wide range of health complications beyond respiratory illness [1, 2]. Neurologic manifestations, including acute cerebrovascular disease (CVD), have been increasingly recognized as significant sequelae of COVID-19 infection [3–5].

Studies have reported a notable increase in the incidence of acute ischemic stroke and intracerebral hemorrhage among COVID-19 patients [6–8]. Given the growing body of evidence suggesting a strong association between COVID-19 and stroke [6], this study aims to investigate the clinical characteristics, neuroimaging findings, and mortality of stroke patients during the pandemic period. By comparing stroke patients with and without COVID-19 infection, as well as a pre-pandemic control group, we aim to answer the following research questions: 1. What proportion of stroke patients had COVID-19 infection? 2. What are the clinical characteristics of stroke in COVID-19 patients? 3. What is the outcome of stroke in COVID-19 patients? 4. What factors may influence mortality in stroke patients with COVID-19 infection?

By addressing these questions, this study will contribute to a better understanding of the impact of COVID-19 on cerebrovascular health and inform clinical practice and public health strategies.

Materials and methods

This retrospective cohort study included all stroke patients admitted to three tertiary care neurology centers in Tehran, Iran (Rasoul-Akram and Firoozgar Hospitals, affiliated with Iran University of Medical Sciences) and Karaj, Iran (Shahid Rejaee Hospital, affiliated with Alborz University of Medical Sciences). Patients were enrolled in two distinct periods: the pre-COVID-19 period which includes patients who came to the hospital one-year period prior to the COVID-19 pandemic and the post-COVID-19 period that is a three-month period following the onset of the COVID-19 pandemic in Iran.

Patients were categorized into three groups: the first was COVID-19 positive (COV-pos) that included patients who tested positive for SARS-CoV-2, the second was COVID-19 negative (COV-neg) composed of patients who tested negative for SARS-CoV-2, and the third was non-COVID-19 (non-COV) with patients from the pre-COVID-19 period.

To select eligible stroke patients, we initially performed a brain CT scan. For cases where the CT scan was inconclusive, we further utilized MRI to definitively diagnose the stroke subtype. Patients were then categorized into two groups: ischemic stroke and hemorrhagic stroke. Ischemic stroke was confirmed by a CT scan without evidence of intracranial hemorrhage, while hemorrhagic stroke was diagnosed by a CT scan demonstrating intracranial hemorrhage.

For all of the individuals, we collected demographic features, past medical history, the characteristics of stroke, laboratory findings, involvement of lungs and the need for mechanical ventilation, and the final outcome of death. The characteristics of stroke included the type of stroke, the circulation and territory involved, the initial National Institutes of Health Stroke Scale (NIHSS), and Modified Rankin Scale (MRS). Mortality rate, need for mechanical ventilation, and the days of hospitalization were also measured.

Results

Hazard ratio for final outcome of death

Table 4 denotes the calculation of HR for each variable and its interaction with COVID-19 infection in stroke patients. After considering the interaction with COVID-19 infection, the greatest HR was observed for posterior cerebral artery (PCA) stroke (HR = 65.099), internal carotid artery stroke (ICA) (HR = 19.102), history of diabetes mellitus (HR = 3.824), arterial ischemic stroke (HR = 2.947), and history of hypertension (HR = 2.194), respectively (P < 0.05) (Figure 1).

Table 4. The Cox proportional hazards model ratio (HR) for each variable and its interaction with having COVID-19 infection in stroke patients

Variables	Main variable			Interaction with COVID-19 infection
	Beta	HR	P value	Beta	HR	P value
Age	0.012	1.012	0.115	0.009	1.009	0.006
Gender	–0.070	0.932	0.780	0.407	1.502	0.006
Hx of hypertension	–0.007	0.993	0.978	0.786	2.194	0.006
Hx of atrial fibrillation	–0.042	0.959	0.916	0.649	1.913	0.357
Hx of cardiac valvular disease	0.988	2.687	0.006	–0.652	0.521	0.233
Hx of ischemic heart disease	–0.251	0.778	0.486	1.080	2.945	0.055
Hx of diabetes melitus	–0.714	0.490	0.037	1.341	3.824	0.004
Hx of any comorbidity	–0.039	0.962	0.892	0.672	1.958	0.010
Number of previous comorbidities	0.006	1.006	0.946	0.281	1.324	0.012
Hx of smoking	–0.842	0.431	0.069	0.446	1.563	0.685
Hx of alcohol consumption	–1.374	0.253	0.185	–12.484	0.000004	0.990
Ischemic stroke (arterial)	–0.567	0.567	0.031	1.081	2.947	< 0.001
Venous stroke	–1.346	0.260	> 0.999	0	1	> 0.999
Intracerebral hemorrhage	0.294	1.342	0.258	–0.017	0.983	0.965
Subarachnoid hemorrhage	0.257	1.293	0.618	0.138	1.147	0.903
Anterior circulation	–0.100	0.904	0.724	0.445	1.560	0.090
Posterior circulation	–0.293	0.746	0.442	1.220	3.388	0.033
Mixed circulation	–0.739	0.478	0.463	2.087	8.064	0.071
ACA stroke	–13.999	8e–07	> 0.999	0	1	> 0.999
MCA stroke	–0.321	0.725	0.231	0.600	1.822	0.117
PCA stroke	–1.142	0.319	0.258	4.176	65.099	0.004
ICA stroke	–0.920	0.399	0.362	2.950	19.102	0.008
Initial NIHSS	0.107	1.113	< 0.001	0.009	1.009	0.575
MRS before stroke	0.234	1.264	0.162	0.148	1.16	0.457
MRS during admission	0.795	2.214	< 0.001	0.009	1.009	0.884
Need for mechanical ventilation	3.653	38.585	< 0.001	0	1	0.999

Display full size

The first column presents the HR for the variable alone, and the second column displays the interaction of that variable with COVID-19, indicating how much COVID-19 has increased the hazard rate due to that variable. Hx: history; ACA: anterior cerebral artery; MCA: middle cerebral artery; PCA: posterior cerebral artery; ICA: internal carotid artery; NIHSS: National Institutes of Health Stroke Scale; MRS: Modified Rankin Scale

Display full size

Figure 1. This hazard ratio chart shows the interaction between COVID and main characteristics of stroke factors. The red line represents a hazard ratio of 1. Hx: history; ACA: anterior cerebral artery; MCA: middle cerebral artery; PCA: posterior cerebral artery; NIHSS: National Institutes of Health Stroke Scale; MRS: Modified Rankin Scale

Discussion

In the present study, we retrospectively investigate the characteristics and outcomes of patients infected with SARS-CoV-2 and stroke and compare them with stroke patients in non-COVID era, to provide novel insights into the spectrum of this disease. Of 124 individuals with stroke who were admitted to three centers, 59 patients had laboratory and/or radiographically-confirmed SARS-CoV-2.

The pathophysiological mechanisms associated with the risk of acute ischemic stroke (AIS) appear to be a complex interplay of multiple factors. A primary consideration is the presence of underlying conventional risk factors. These factors may predispose individuals to either incidental stroke or increase their susceptibility to stroke triggered by the infectious process.

Several potential mechanisms have been proposed. The inflammatory and cytokine response associated with COVID-19 may destabilize atherosclerotic plaques, leading to thromboembolism. Additionally, matrix metalloproteinases (MMPs) are enzymes involved in various physiological and pathological processes, making them fewer specific biomarkers for COVID-19. Other factors, such as chronic diseases or comorbidities, can influence MMP levels in the context of SARS-CoV-2 infection and accordingly increase stroke risk. Moreover, impaired cardiac function, resulting from myocardial dysfunction or arrhythmias, can increase the risk of cardioembolic stroke. The downregulation of ACE-2 receptors, which may reduce the activity of the renin-angiotensin system, could contribute to vascular dysfunction and increased thrombotic risk. Furthermore, emerging evidence suggests a prothrombotic state in COVID-19 patients, which may further elevate stroke risk [1, 9–16].

Although there have been many reports of young strokes in COVID-19 patients [11, 17, 18], in our study, the mean age of patients with concomitant stroke and COVID-19 was 70.90 ± 14.3, which is significantly higher than mean age of stroke patient in pre-COVID period (P value < 0.001). Moreover, as opposed to a systematic review by Stefania Nannoni et al. [19], comparison with non-infected patients with stroke showed no special differences in mean age (70.90 ± 14.3 vs. 69.03 ± 16.0) [19]. We observed a female predominance in patients with COVID-19 infection suffering a stroke which is unlike previous studies that showed equivalent results [20] or a higher number of male involvements [18, 21, 22].

There was a decrease in stroke admissions after the onset of COVID-19 pandemic. The same decrease was reported in other studies conducted in other countries all over the world [23–25]. In addition, detecting signs of ischemic stroke in critically-ill COVID-19 patients who are often intubated and strongly sedated, is especially difficult, masking their clinical features [26]. It also could be due to the fear of becoming infected, that patients with asymptomatic and mild to moderate strokes may be reluctant to seek hospital care [23, 27]. Reduction in cardiology services for coronary artery disease admissions during the COVID-19 pandemic was also noticed [28]. Furthermore, a decrease was observed in the utilization of acute stroke treatments, such as mechanical thrombectomy and recombinant tissue plasminogen activator (rtPA) [29]. Accordingly, this finding is probably not limited to stroke, nor to the geographic region where the present study conducted. Moreover, we observed that mean NIHSS score in COVID-19 period was higher than the same period a year before. Thus, the findings of our study about the age and NIHSS of stroke patients signifies the hypothesis that in the COVID era, all of the patients with any cause including stroke, did not come to hospital because of fear of infecting to COVID virus [30]. Albeit, when the symptoms of stroke were markedly disabling, they came to the hospital for treatment. A recent review by Ménard et al. [31] highlighted the importance of early intervention, patient education, and timely medical care for stroke patients, even during the COVID-19 pandemic. However, interestingly, we observed that the NIHSS was higher in COV-pos patients compared with COV-neg patients. In a study that compared the stroke characteristics of COV-pos patients with COV-neg patients and historical controls, consistent findings were reported. They have demonstrated that patients with concomitant COVID-19 suffered more severe strokes, with a higher NIHSS score [26]. This finding may signify the effect of concomitant COVID-infection on the signs of stroke. The study of Shahjouei et al. [8] revealed that 24.4% of patients with stroke and COVID-19 infection had no identifiable risk factor for vascular events. In addition, they found that 37.8% of patients were initially admitted with COVID-19 infection in the hospital and were asymptomatic for stroke at the onset of admission.

Observational studies support the role of modifying lifestyle-related risk factors (e.g., smoking, alcohol use, physical activity, diet) in stroke prevention [32]. The observation of lower rate of smoking and alcohol consumption among COV-pos patients in our study (P value < 0.001) signifies the role of COVID-19 as the cause of stroke. Furthermore, the number of previous comorbidities was the same in both COVID-era and a year before. In both periods, the most prevalent comorbidities were hypertension and diabetes mellitus, followed by cardiovascular diseases. The same prevalence of risk factors for stroke was reported in a meta-analysis of 8 studies from China [33, 34].

Similar to what previous studies have shown, the most common type of stroke in our study was ischemic stroke as well [19, 35]. We did not observe any differences in types of ischemic stroke, whether it is territorial, lacunar, hemodynamic, or dissection in comparison to the data from a year before COVID-19 pandemic. Although many studies have categorized ischemic stroke subtypes to cryptogenic, cardioembolic, large vessel occlusion, and small vessel occlusion showing the higher prevalence of cryptogenic causes [26, 36], we have used another system. In our study, the most common type of ischemic stroke in COV-pos patients was territorial one, including 88.1% of cases, followed by hemodynamic strokes accounting for 7.1% of cases, and lacunar strokes in 4.8% of cases. We observed that posterior circulation strokes were as common as anterior and mixed circulation strokes between three groups. This is in contrast with the recently described phenotype in two other case series, which shows a high incidence of posterior territory predisposition, and multi-territory involvement [6, 37].

Our study also included patients with CVD caused by hemorrhagic stroke. It was the second most prevalent type of stroke. We have not observed any differences in its prevalence in COVID-era and non-COVID-era (P value = 0.867). Interestingly, there was a substantially higher proportion of lobar intracerebral hemorrhage (ICH) in a year before COVID-19 pandemic, probably highlighting greater referral approaches before this pandemic (18.75% vs. 50%). Besides, there was a significant difference in the incidence of subarachnoid hemorrhage (SAH) between COV-pos and COV-neg patients (17.6% vs. 6.3%), but this difference was not apparent in non-COV patients (15.4%).

Furthermore, we observed a high rate of functional dependency during the hospitalization period for COV-pos patients (MRS = 3.81 ± 1.5), compared to MRS of our COV-neg patients and prior to the pandemic (P value < 0.001). This could be alluded to extreme illness and general weakness secondary to COVID-19 disease. Moreover, mortality and morbidity rates both were higher in COV-pos patients. In line with our results, a study by Hernández-Fernández et al. [37] illustrated that functional prognosis during the hospital period was unfavorable in 73.9% of their patients (17/23 MRS 4–6). Our findings revealed that, more than half of COV-pos patients needed mechanical ventilation, but it was 29.2% in COV-neg patients and 17.4% a year before pandemic, highlighting the prominent role of COVID infection in respiratory failure and related complications.

We have observed that the duration of hospitalization was less in non-infected stroke patients (8.53 ± 8.6) compared to a year before (10.3 ± 9.2). Having in mind the higher mean NIHSS score of both COV-pos and COV-neg patients in comparison with a year before, this could be due to fear of infection and higher rate of self-discharges. It has been said that patients with stroke had a 65% proportion of critical rate of COVID-19 in mainland China, which is higher than the average 5% critical rate [38], which can explain longer length of hospital-stay in the patients with stroke and COVID-19 infection (12.61 ± 11.3) than the patients without COVID-19 [39]. The overall mortality rate among COV-pos patients was 49.2%, which was much higher than COV-neg patients (24.2%) and non-COV-era (17.3%). Although Shahjouei et al. [8] did not compare the mortality of stroke in COV-pos and COV-neg patients, their study reported an in-hospital mortality rate of 131 out of 432 patients (30.3%) with stroke and COVID-19; among these patients, 323 (74.8%) had acute ischemic stroke, 91 (21.1%) had intracranial hemorrhage, and 18 (4.2%) had cerebral venous or sinus thrombosis. The rate of mortality at ICH and SAH patients was significantly higher than AIS patients (62.5% and 55% vs. 27.6%, respectively) [8]. In this regard, our study revealed that HR of mortality for ICH and SAH was not statistically different, with and without the effect of COVID-19 infection (P value > 0.05). Interestingly, the HR for ischemic stroke is 0.567 which means that ischemic stroke had less mortality outcome than other types of strokes. But when we consider the effect of COVID-19 infection, the HR reaches 2.947 (P value < 0.001). In other words, these findings suggest that COVID-19 infection has a significant effect on mortality of patients with ischemic stroke.

Furthermore, the results of our study revealed that COVID-19 infection has the most effect on poor outcome of patients with PCA and ICA strokes, and those having the history of diabetes mellitus and of hypertension. The markedly significant HR of PCA infarction (HR = 65) and ICA infarction (HR = 19), draws our attention to the notion that these types of strokes should carefully be managed in the context of COVID-19 infection. In a large multicentric study [8] in 135 tertiary centers of stroke worldwide, large vessel occlusion was observed in 44.5% of AIS plus COVID-19 infection, however, mechanical thrombectomy was performed only in 7.4% of patients. Even, intravenous thrombolysis was administered in 13.6% of patients. These low rates may reflect the delay to reach hospital in COVID-era or delay in diagnosis of stroke in the context of COVID-19 symptoms.

One final point is the effect of comorbidities on the mortality of stroke in COVID-19 patients. In line with other studies, our study showed that having the history of hypertension, diabetes mellitus, and at least one comorbidity contributes to unfavorable outcomes and higher mortality rates.

Nevertheless, we need to highlight that, this study includes a short time interval which may not adequately reflect the impact of a pandemic on the healthcare stroke systems, and this can lead to underestimations due to insufficient observation time.

In this study, we conducted some interesting data by analyzing the effect of different variables when COVID-19 exists relative to when COVID-19 is absent. These data showed that COVID-19 infection changed markedly the characteristics and the outcome of stroke.