Cognitive impairment and endothelial dysfunction in convalescent COVID‐19 patients undergoing rehabilitation

Abstract

The endothelium acts as the major regulator of vascular homeostasis, and its functions encompass maintenance of vascular integrity, regulation of haemostasis and control of inflammation.1 Endothelial dysfunction is the earliest mechanism of atherosclerosis, thus potentially underlying the association between traditional cardiovascular risk factors (eg hypertension, diabetes, dyslipidaemia) and some neurological disorders.2 However, studies on the relationship between endothelial function and neurocognition are relatively scarce.3 In these studies, the most used measure of endothelial function is flow-mediated dilation (FMD) of the brachial artery, which is widely accepted as a reliable and cost-effective tool and as an independent predictor of cardiovascular events.4, 5 Adopting this method, some authors reported the association between an impaired FMD and poorer cognitive performances, particularly in the attention, executive and memory domains, mainly in patients with cardiac diseases.3 Similar findings were documented in patients with vascular cognitive impairment when considering other clinical or laboratory biomarkers related to endothelial function (eg adhesion molecules, inflammatory mediators, neuroimaging markers).6 Overall, the link between endothelial dysfunction and cognitive impairment is complex and not fully elucidated. In this regard, a number of mechanisms have been called into question, including an altered cerebral perfusion,7 a decreased cerebrovascular reactivity,8 a deficient production of brain-derived neurotrophic factors9 and an inflammatory response in the central nervous system.10 More recently, endothelial dysfunction has been reported in convalescent coronavirus disease 2019 (COVID-19) patients, with a direct correlation between the severity of pulmonary and vascular involvement.4, 5 Moreover, cognitive impairment has been also documented in these patients up to six months after the acute phase.11 To the best of our knowledge, no study investigated the potential relationship between endothelial function, as assessed by FMD, and cognitive efficiency in this clinical setting. This investigation could provide strong support to the possible role of endothelial dysfunction in the pathogenesis of cognitive disorders related to COVID-19. To address this issue, convalescent COVID-19 patients admitted to the Pulmonary Rehabilitation Unit of Istituti Clinici Scientifici Maugeri IRCCS, Institute of Telese Terme, Benevento, Italy, within two months from swab test negativization were consecutively screened for inclusion. Patients were selected according to the following criteria: age between 18 and 65 years; presence of formal education (at least primary); no past or current psychiatric or neurological disease; and no premorbid clinically relevant cardiovascular or respiratory condition potentially impacting cognitive performance (ie hypertension, diabetes, chronic heart failure, chronic obstructive pulmonary disease, asthma). Patients with missing data for the outcome of interest were excluded from the study. Wherever appropriate and applicable, this study was reported following the broad Enhancing the QUAlity and Transparency Of health Research (EQUATOR) guidelines.12 All participants gave their informed consent in written form. The study was approved by the competent Ethics Committee and was performed according to the principles of the 1975 Helsinki Declaration. In all the enrolled patients, we assessed the following cognitive domains by means of standardized tests: spatial and verbal short- and long-term memory (Corsi block-tapping test, Rey's auditory learning test and supra-span learning on Corsi's test, respectively); executive functions (phonemic verbal fluency task and frontal assessment battery); attention (trail making test, TMT); abstract reasoning (Raven's Coloured Progressive Matrices, CPM47); and global cognitive functioning (Montreal Cognitive Assessment, MoCA). The tests were administered at study entry by an experienced neuropsychologist (P.M.). Patients’ scores on each test were evaluated in reference to available Italian age- and education-corrected values. Patients who achieved two or more pathological scores on neuropsychological tests, and those who achieved one pathological score on one test plus one or more borderline scores on other tests were considered to have a reduced cognitive efficiency (RCE). We also assessed functional independence and self-perceived quality of life by means of Barthel scale and EuroQol-5D, respectively. Upon admission, we assessed vascular reactivity by FMD of the brachial artery, as detailed elsewhere.4, 5 In brief, FMD expressed the per cent (%) change of brachial artery diameter compared with the baseline after reactive hyperaemia induced by forearm ischaemia, being calculated as [(maximum post-ischaemic brachial artery diameter – basal brachial artery diameter) / basal brachial artery diameter] × 100. An automatic edge detection software (Cardiovascular Suite®, FMD Studio, QUIPU Srl, Pisa, Italy) registered in Europe as a medical device (MED 31116) was used for a real-time evaluation. An expert operator (P.A.) assessed parameters of vascular reactivity. Lower FMD values indicated worse vascular response. Among 137 convalescent COVID-19 patients screened for eligibility, 73 (53.3%) were ineligible for protocol adherence issues. A total of 4 (6.2%) out of the 64 eligible patients dropped out before completion of the project requirements, while 5 (7.8%) were not considered due to incomplete neuropsychological evaluation or unsuccessful FMD measurement. Thus, 55 convalescent COVID-19 patients (83.6% men, mean age 60.1 years) were considered for the final analysis. As shown in Table 1, nearly half (41.8%) of the enrolled patients had RCE. The most impaired scores belonged to the domains of executive functions [impaired scores on phonemic verbal fluency in 14/55 patients (25.5%)], verbal long-term memory [impaired scores on Rey's Auditory Learning test in 13 of 55 patients (23.6%)] and spatial long-term memory [impaired scores on supra-span learning on Corsi's test in 12 of 55 patients (21.8%)]. No patient scored below the normal range on the measure of global cognitive functioning, assessed by MoCA. RCE patients exhibited a worse pulmonary function than non-RCE, as expressed by a significantly lower forced vital capacity (FVC) and diffusing lung capacity for carbon monoxide (DLCO), and by a trend towards a lower forced expiratory volume in 1 s (FEV1). Moreover, they had a lower self-perceived quality of life, as assessed by the EuroQol-5D (Table 1). FMD values were significantly lower in RCE patients as compared to non-RCE (2.25% ±1.94 vs. 3.90% ±2.40, p = 0.006; Figure 1), being negatively correlated with the number of impaired scores on neuropsychological tests (rho = −0.364, p = 0.020; Figure 2). The differences in FMD (F = 9.386, df = 1, p = 0.007), DLCO (F = 4.469, df = 1, p = 0.043) and EuroQol-5D (F = 5.408, df = 1, p = 0.032) were confirmed in multivariate analyses after adjusting for age, gender and body mass index. Moreover, in line with previous evidence,4, 5 FMD values were significantly and positively correlated with measures of pulmonary function, such as FEV1 (rho = 0.385, p = 0.019), FVC (rho = 0.365, p = 0.026) and DLCO (rho = 0.402, p = 0.025). Overall, this pilot study represents the first attempt to explore the association between endothelial function and neurocognition in convalescent COVID-19 patients. In general, our findings confirmed that, after a severe-to-critical form of COVID-19, a relevant cognitive impairment may be present in a remarkable percentage of cases (higher than 40%), with a severe reduction in vascular reactivity and a residual impairment of pulmonary function. Most important, in our sample of convalescent COVID-19 patients without premorbid known cardiovascular or pulmonary conditions, we found an association between endothelial dysfunction and cognitive impairment, confirmed in multivariate analyses. This suggests that a virus-induced alteration of the endothelium may play a key pathogenic role in the systemic manifestations of COVID-19, including pulmonary, vascular and neuropsychological manifestations. Since endothelial dysfunction may be responsible for the cerebral hypoperfusion that alters the neural metabolism,13 it is possible to argue that a dysfunctional endothelium in the central nervous system may contribute to the neurological disturbances and cognitive impairment in COVID-19 patients. In keeping with this, some studies employing 18F-fluorodeoxyglucose positron-emission tomography/computed tomography (18F-FDG-PET/CT) described an altered brain metabolic pattern in COVID-19.14 Indeed, such patients had a consistent pattern of hypometabolism in a widespread cerebral network, including the frontal cortex, the anterior cingulate, the insula and the caudate nucleus, up to 6 months after disease onset.14 Overall, in the frame of a comprehensive care approach to COVID-19, our preliminary findings confirm that endothelial dysfunction could be regarded as an additional therapeutic target in this clinical setting, with a number of both pharmacological (eg antioxidants, anti-inflammatory therapies) and non-pharmacological strategies (eg rehabilitation, exercise, diet) having potential clinical applications in the post-acute phase.15 Moreover, this study highlights the importance of assessing FMD and other clinical or laboratory markers of endothelial function during convalescence, in order to timely predict the long-term outcomes of COVID-19, including cognitive decline. Further prospective studies on a larger sample size should be specifically designed to confirm and investigate the potential relationship between vascular homeostasis and cognitive performance in COVID-19 patients. None. The authors declare no conflict of interest. PM and PA conceived and designed the study, performed statistical analysis, interpreted results and drafted the manuscript. MM, PM and AL acquired clinical data and drafted the manuscript. LT and PA supervised the project and performed critical revisions. All Authors read and approved the final version of the manuscript.

Topics & Concepts