关于COVID-19引起神经系统并发症和神经系统症状的机制的信息每天都在更新,但关于对中枢神经系统产生后遗症的可能性仍有许多不清楚的地方。
虽然一些随机临床试验调查了中重度COVID-19病例的症状、影像学表现和治疗方法,但关于轻症病例的报告仍然很少,而轻症病例占受感染个体的大多数。
与严重急性呼吸综合征和中东呼吸综合征类似,据报道COVID-19直接或间接影响神经系统,患者表现出各种神经精神症状。
新冠病毒入侵中枢神经系统的机制可能包括:
1)新冠病毒从鼻腔经嗅球直接侵入中枢神经系统;
2)感染的单核细胞通过血脑屏障,感染胶质细胞和神经元;
3) SARS-CoV-2与血脑屏障内皮细胞ACE2受体结合,侵入中枢神经系统;
4) SARS-CoV-2通过外周神经系统侵入中枢神经系统。
此外,呼吸衰竭引起的低氧血症、免疫反应异常引起的脑病、血栓形成倾向引起的脑梗死也是新冠病毒影响大脑的机制。一项研究在尸检中,87名患者中有58人的大脑中发现了SARS-CoV-2,这为SARS-CoV-2直接渗透大脑提供了额外的证据。
根据一份总结新冠肺炎住院患者非特异性神经症状的报告,头痛、头晕、厌食、疲劳和嗅觉和味觉受损的比例分别为8%-42%、12%、40%、11%-44%和5%。
关于中枢神经系统疾病,有许多中风并发症的报道。SARS-CoV-2被认为是通过破坏血管内皮、加剧炎症和血栓形成倾向而导致中风的。在中国、欧洲和美国的重症监护患者报告中,有5.9-76.8%的患者有卒中并发症,约80%的患者有脑梗死。
此外,10例患者的尸检报告显示脑实质血管有微出血和梗死,尽管患者没有表现出特定的神经症状。
在需要住院治疗的COVID-19患者中,有多例常见头部MRI/A序列异常报道。在英国125例有神经精神症状的COVID-19患者的头部MRI/ a横断面研究中,77例(62%)患有脑血管疾病,57例(74%)患有脑梗死,9例(12%)患有脑出血,1例(1%)患有中枢神经系统血管炎。
共有39例患者(31%)有精神症状,其中9例患者(23%)有非特异性脑病,7例患者(18%)有脑炎。在法国的一项多中心回顾性观察性头部MRI发现研究中,包括190例至少需要住院供氧的患者,其中37例患者(19.5%)发现异常。
MRI主要表现为单侧FLAIR和DWI信号升高16例(43%),脑白质多发病变11例(30%),脑微出血9例(24%)。Chougar等报道,73例COVID-19患者在住院期间出现神经精神症状并拍摄头部MRI/A图像,发病2 ~ 4周后58.9%出现异常,47.7%出现脑灌注异常,23.3%出现脑梗死,11.3%出现脑微出血,5.5%出现脑白质多发病变。
因此,尽管有几篇关于COVID-19住院患者头部MRI/A结果的报道,但很少有关于轻度疾病、非住院患者的头部MRI/A报告,这些患者占感染个体的大多数,因为头部MRI/A很少在不需要住院且几乎没有神经精神症状的患者中进行。
为了调查后续影响的未来风险,重要的是检查感染后是否存在脑实质异常,即使是轻微病例。在日本,同时进行头部核磁共振成像(MRI)和MRA检查的“Brain Dock”是对健康个体进行头部检查的方法。在这项研究中,我们使用“Brain Dock”来评估轻度COVID-19感染后的MR结果。
我们检测呈阳性的患者数量与当时日本的阳性率相当。所有患者均为轻度感染,未住院治疗,几乎一半患者无症状。
从诊断日期到抗体测试和头部MRI/A的平均时间分别为113.4天和108.9天,这表明评估是在感染一段时间后进行的。
影像学检查结果对比显示,阳性与阴性患者在脑白质病变、脑微出血、脑梗死等严重病例的主要表现上无明显差异。阳性组患者均无嗅球萎缩、嗅黏膜增厚等易受新冠病毒感染损害的鼻部异常征象。
这表明轻症对中枢神经系统的损害相对较轻。在阳性组的患者中,我们能够监测4名患者感染前后的图像。所有患者在感染时均有发热,并观察到头痛(2例)和味觉和嗅觉障碍(3例)等症状。脑实质未见感染相关改变。
此外,为了评估关于进行抗体测试动机的偏见,我们检查了抗体测试结果为阴性的患者与仅进行头部MRI/ a而未进行抗体测试的患者之间的影像学结果是否存在差异。结果显示,两组影像学表现无明显差异。
因此,进行抗体测试的动机对我们的研究结果没有显著影响。
综上所述,轻症患者感染COVID-19后,常见的头部MRI/ a序列可能会有一些特定的变化。
据报道,轻度病例神经精神症状并发症相对较少。Liu等研究发现,轻症患者的平均病毒量比重症患者低约60倍,病毒清除时间较早;因此,直接浸润中枢神经系统的风险相对较小。
我们的结果支持先前基于成像结果的报告。值得注意的是,Douaud等人报道,即使在COVID-19的轻症病例中,脑容量也会减少0.2%-2.0%,特别是在眼窝前额皮质和海马旁回的灰质中,这表明即使是轻症病例也可能导致神经损伤,尽管这些损伤非常小,无法通过常见的MRI/A序列检测到。然而,后遗症的原因-如脑雾和嗅觉异常-可能是非常微妙的脑萎缩,无法通过普通的头部MRI/A序列检测到,这与我们的结果一致。
基于上述,如果轻度COVID-19感染后仍有神经症状,即使常见的头部MRI/A序列没有异常,也建议进行长期随访,包括脑容量测量。
我们研究的局限性包括使用抗体而不是核酸扩增试验来检测感染。使用抗体检测的好处是,可以检测占轻症患者近一半的无症状患者。然而,诸如从感染开始的时间等信息的准确性是缺乏的。此外,由于Brain dock的性质(用于健康个体的成像),在本研究中无法进行对比增强MRI,血流和脑容量评估。此外,有可能在感染的急性期出现异常,但在头部MRI/A上没有观察到,因为成像是在感染后一段时间进行的。
然而,在考虑后续影响的可能性时,有必要评估不可逆的发现。因此,在我们的研究中,分析感染后一段时间常见头部MRI/A序列的异常表现是有意义的。
结论综上所述,我们的研究表明,在重症患者中常见的MRI/A序列中所见的器质性异常在轻症患者中没有出现,这表明轻症患者对中枢神经系统相对没有损害。然而,考虑到即使在轻症病例中也有脑容量微减少的报道,还需要进一步研究COVID-19对神经系统的影响。今后需要对症状进行进一步的调查和随访。

Discussion
The information on the mechanism by which COVID-19 causes neurological complications and neurological symptoms is updated daily, but there are still many unclear points about the possibility of after-effects on the central nervous system. Although several randomized clinical trials have investigated symptoms,imaging findings, and treatment methods in moderate or severe COVID-19 cases, there are still few reports on mild cases, which account for the majority of infected individuals.
Similar to severe acute respiratory syndrome and middle east respiratory syndrome, COVID-19 has been reported to, directly and indirectly, affect the nervous system, with patients exhibiting a variety of neuropsychiatric symptoms [4,22]. The mechanisms by which SARS-CoV-2 invades the central nervous system may include: 1) direct invasion of SARS-CoV-2 from the nasal cavity into the central nervous system via the olfactory bulb; 2) infected monocytes pass through the blood-brain barrier and infects the glial cells and neurons; 3) SARS-CoV-2 binds to the ACE2 receptor on the endothelial cells of the blood-brain barrier and invades the central nervous system; and 4) SARS-CoV-2 invades the central nervous system via the peripheral nervous system [8,9]. Furthermore, hypoxemia due to respiratory failure, encephalopathy due to an abnormal immune response, and cerebral infarction due to a thrombotic tendency are also mechanisms by which SARS-CoV-2 affects the brain. One study found SARS-CoV-2 in the brains of 58 out of 87 patients in autopsy [6], providing additional evidence that SARS-CoV-2 directly infiltrates the brain.
According to a report summarizing the non-specific neurological symptoms of COVID-19 inpatients headache, dizziness, anorexia, fatigue, and impaired smell and taste were reported in 8%-42%, 12%, 40%,11%-44%, and 5%, respectively [23]. Regarding central neurological disorders, there are many reports of complications of a stroke. SARS-CoV-2 is thought to cause a stroke by damaging the vascular endothelium,exacerbating inflammation and thrombotic tendency. Reports on patients requiring intensive care in China,Europe, and the United States have stated that 5.9-76.8% had stroke complications, and approximately 80%had cerebral infarction [24-26]. Additionally, autopsy reports of 10 patients showed microbleeds and infarction in the blood vessels of the cerebral parenchyma, despite patients not presenting specific neurological symptoms [7].
There are multiple reports of abnormal findings on common head MRI/A sequences in COVID-19 patients who required hospitalization [13-15]. In a cross-sectional study of head MRI/A findings of 125 COVID-19 in patients with neuropsychiatric symptoms across the UK, 77(62%) had a cerebrovascular disorder, 57(74%)had cerebral infarction, nine (12%) had a cerebral hemorrhage, and one (1%) patient had central nervous system vasculitis [27]. A total of 39 patients (31%) had psychiatric symptoms, of whom nine patients (23%)had nonspecific encephalopathy and seven patients (18%) had encephalitis. In a French multi-center retrospective observational study of head MRI findings which included 190 patients who required hospitalization for at least oxygen administration, abnormalities were found in 37 patients (19.5%) [13]. The main MRI findings were increased signals in unilateral FLAIR and DWI in 16(43%), multiple cerebral white matter lesions in 11(30%), and cerebral microbleeds in nine (24%) patients. Chougar et al. reported that among 73 COVID-19 patients who had neuropsychiatric symptoms during hospitalization and head MRI/A images taken, 58.9% had abnormal findings two to four weeks after onset, 47.7% had abnormalities of cerebral perfusion, 23.3% had cerebral infarction, 11.3% had cerebral microbleeds, and 5.5% had multiple cerebral white matter lesions [15]. Thus, despite several reports on head MRI/A findings of hospitalized COVID-19 patients, there are few reports of head MRI/A for mildly ill, non-hospitalized patients-who account for the majority of infected individuals because head MRI/A is rarely performed in patients who do not require hospitalization and have few neuropsychiatric symptoms.
To investigate the future risk of after-effects, it is important to examine the presence or absence of brain parenchymal abnormalities after infection even in mild cases [11]. In Japan, "Brain Dock," in which both head MRI and MRA are conducted, is performed as a head screening for healthy individuals. We used "Brain Dock" to evaluate MR findings after mild COVID-19 infection in this study. XThe number of our patients who tested positive was comparable to the positivity rate in Japan at that time. All patients had a mild infection and were not hospitalized, and almost half were asymptomatic. The average duration from the date of diagnosis to the antibody test and head MRI/A were 113.4 and 108.9 days, respectively, indicating that the evaluation was performed after some time had passed since being infected.
A comparison of imaging findings showed no significant difference in findings (e.g., cerebral white matter lesions, cerebral microbleeds, and cerebral infarction, which are major findings in severe cases) between individuals who tested positive and negative. No patient in the positive group showed abnormal nasal findings such as olfactory bulb atrophy or thickening of the olfactory mucosa, which are easily damaged by COVID-19 infection. This suggests that there is relatively no damage to the central nervous system in mild than severe cases. Among the patients in the positive group, we were able to monitor the images of four patients before and after being infected. All patients had fever at the time of infection, and symptoms such as headache (two patients) and impairments in the senses of taste and smell (three patients) were observed.No infection-associated changes in brain parenchymal findings were observed.
Furthermore, to evaluate biases regarding the motivation for taking an antibody test, we examined whether there was a difference in imaging findings between those who had a negative antibody test result and who had only head MRI/A without undergoing an antibody test. The results showed that there was no significant difference in the imaging findings. Hence, the motivation for taking an antibody test did not significantly affect the results of our study. Considering the above findings, there may be a few specific changes in common head MRI/A sequences after infection with COVID-19 in mild cases.
Reportedly, there are relatively few complications of neuropsychiatric symptoms in mild cases [28,29]. Liu et al. found that the average amount of virus in COVID-19 mild cases is approximately 60 times lower than in severe cases, and virus clearance is observed at an earlier stage; thus, the risk of direct infiltration into the central nervous system is relatively small [29]. Our results support previous reports based on imaging findings. Notably, Douaud et al. reported brain volume reductions of 0.2%-2.0% even in mild cases of COVID-19, particularly in the gray matter of the orbitofrontal cortex and parahippocampal gyrus,suggesting that even mild cases may cause neurological damage, although these are very minimal and cannot be detected by common MRI/A sequences [30]. However, the cause of sequelae- such as brain fog and olfactory abnormalities-may be very subtle brain atrophy that cannot be detected by common head MRI/A sequences, concurring with our results. Based on the above, if neurological symptoms remain after mild COVID-19 infection, long-term follow-up-including brain volume measurement-is recommended even if there are no abnormalities in common head MRI/A sequences.
The limitations of our study include the use of antibodies instead of nuclear acid amplification tests to detect infections. The advantage of using an antibody test is that asymptomatic cases, which account for nearly half of the mild cases, can be examined. However, the accuracy of information such as the elapsed time from infection-is lacking. Additionally, due to the nature of the Brain Dock-which is for the imaging of healthy individuals-contrast-enhanced MRI, blood flow and brain volume evaluation could not be performed in this study. Further, it is possible that abnormal findings could have occurred during the acute phase of infection but were not observed on head MRI/A as imaging was performed sometime after infection.
However, it is essential to evaluate the irreversible findings in considering the possibility of after-effects.
Thus, the analysis of abnormal findings on common head MRI/A sequences sometime after infection, as in our study, is meaningful.
Conclusions
In conclusion, our study has shown that organic abnormalities that are seen on common MRI/A sequences in severe cases do not appear in mild cases, indicating relatively no damage to the central nervous system in mild patients. However, given that microreductions in brain volume have been reported even in mild cases,further studies regarding the aftereffects of COVID-19 on the nervous system are needed. Additional investigation and follow-up of symptoms will be necessary in the future.