Search strategy and databases used

Commonly accessed and broadly utilised electronic databases such as MEDLINE/PubMed, and SCOPUS, were the primary databases used for searching published literature. The keywords used were as follows. “COVID-19/Novel coronavirus and Neuroinvasion”, “Long-COVID and Brain fog/Sleep disturbances/Memory impairment/Neurodegeneration/Alzheimer’s disease/Parkinson’s disease, Long-COVID and Psychological changes/Psychosocial effects/Quality of life/Anxiety/Depression/Altered mental status/Gut dysbiosis”, “Long-COVID and Pathophysiology/Management”. Article titles and abstracts were initially screened, and duplicate records were removed at this stage itself. The remaining articles were assessed for their eligibility to be included in this review.

Inclusion criteria

The inclusion criteria were peer-reviewed original research on long-COVID including clinical, basic research, and related systematic reviews. Resources available in the public domain from major regulatory bodies such as WHO and CDC were also referred to if they contained information relevant to various sections of this review.

Exclusion criteria

Studies excluded from the analysis are articles not officially published in a peer-reviewed journal, conference abstracts and proceedings, corrigendum documents, repeated publications, unrelated articles like acute neurological effects of COVID-19, and non-English articles.

Qualitative synthesis

Full text of each of the articles selected for qualitative synthesis (n = 228) was thoroughly analysed. Four major areas of outcome measure were identified, and they were neurological, psychological, and psychosocial complications of long-COVID and management of neurological complications in long-COVID. Two authors have worked independently to extract data/information from reports deemed to be included in each of the sections of this review. Any conflicts between these authors’ findings were resolved by discussion and consensus in the presence of the third author. Major neurological, psychological, and psychosocial complications seen in long-COVID patients were extracted and summarised alongside suggesting possible underlying pathophysiological mechanisms for these complications and their management.

Neuroinvasion of SARS-CoV-2 and its implications in the development of long-COVID ACE2 expression in the brain

ACE2, a transmembrane glycoprotein, is the key entry receptor for SARS-CoV-2 and plays a vital role in its pathogenesis. Extensive expression of ACE2 has been found on the surface of many tissues, including the respiratory system, cardiovascular system, gastrointestinal tract, kidneys, choroid plexus, testes, placenta, and bladder [7]. In the respiratory system, ACE2 is highly expressed in olfactory, nasal, bronchial epithelial cells, and type II alveolar epithelial cells [8]. In the cardiovascular system, ACE2 is expressed in myocytes, vascular endothelial cells, vascular smooth muscle cells of arteries, and venules [7]. In the brain, ACE2 is expressed in specific areas including the substantia nigra, brain ventricles, middle temporal gyri (excitatory and inhibitory neurons), and the posterior cingulate cortex. High expression of ACE2 has been reported in the posterior hypothalamic area, paraventricular nuclei of the thalamus, lateral hypothalamic area, paraventricular nuclei of the hypothalamus, piriform cortex, amygdala-hippocampal transition area, fastigial nucleus and hippocampal CA2 field [9, 10]. SARS-CoV-2 binds to the ACE2 receptor on the host cell via the receptor-binding domain of its spike protein. TMPRSS2 is the host factor that promotes viral uptake and membrane fusion. Viral uncoating and release of RNA to the cytoplasm leads to immediate translation. The resulting polyprotein is processed into both non-structural and structural proteins. Structural proteins translocate into the endoplasmic reticulum membrane. The new virion is then released from the infected cells by exocytosis (Fig. 1) [11].

Fig. 1

Schematic diagram showing the replication cycle of coronavirus. SARS-CoV-2 binds to the ACE2 receptor on the host cell via the receptor-binding domain of its spike protein. As depicted, TMPRSS2 is the host factor that promotes viral uptake and membrane fusion. The viral uncoating and release of RNA to the host cell cytoplasm results in immediate translation followed by viral RNA replication

Possible mechanisms of SARS-CoV-2 invasion to the nervous system

Neuroinvasion of SARS-CoV-2 and triggered inflammation has been demonstrated by experimental animal models, organoid models, and autopsy studies [12, 13]. SARS-CoV-2 primarily replicates inside the respiratory tract (Fig. 1), however, this tract is innervated by several cranial nerves and these nerves are suggested to play a role in neuroinvasion (Fig. 2). Various studies have reported evidence of SARS-CoV-2 entry to CNS via the olfactory nerve, trigeminal nerve, or vagus nerve endings utilising retrograde transport mechanisms (Fig. 2). The virus also appears to neuroinvade by entering the blood from the lung, with the resulting viremia subsequently crossing the blood brain barrier (BBB) or the blood cerebrospinal fluid barrier; the choroid plexus [14,15,16] (Fig. 2).

Fig. 2

Pathways through which SARS-CoV-2 enters various cells including the nervous system. (A) Olfactory route via, olfactory receptor and CN-I, (B) Bronchial route via, CN-X, (C) Intestinal route via, CN-X, (D) Hematogenous route via, altered BBB to the brain parenchyma, (E) Hematogenous route via, infected alveolar parenchyma

Olfactory route

Retrograde dissemination of SARS-CoV-2 through the olfactory nerve seems to be one of the possible mechanisms of neuroinvasion. ACE2, TMPRSS2, and neuropilins are highly expressed in the olfactory epithelium, which makes it a key early infection site for CNS invasion [17, 18]. The virus may enter the brain through the olfactory mucosa neuroepithelium to the olfactory bulb, mitral cells, and olfactory nerve and from there it could spread to different areas of the brain [17]. An autopsy study demonstrated highest level of viral load in olfactory mucosa sampled directly beneath the cribriform plate. Additionally, viral RNA was detected in the olfactory bulb, trigeminal ganglion, and medulla oblongata [14]. Another study provided evidence for the involvement of neuropilin 1(NRP1) in SARS-CoV-2 entry into olfactory epithelium, neurons, and blood vessels of the cortex [18]. A non-human primate model study in rhesus monkeys found evidence of viral RNA in nasal mucosa, CSF, the olfactory trigone, and the entorhinal area by qRT-PCR following intranasal inoculation. These facts strongly support SARS-CoV-2 neuroinvasion via the olfactory route [15].

Vagus nerve route

Another possible SARS-CoV-2 spread into the CNS occurs through the vagus nerve of the lungs and gut afferents. SARS-CoV-2 neuroinvasion has been reported from the lungs via the vagus nerve to the autonomic nerve centres in the brain stem [19]. A human study revealed that cough, dizziness, inappropriate sinus tachycardia, and gastrointestinal symptoms were associated with vagus nerve damage [20]. Similarly, the role of the gastrointestinal tract in SARS-CoV-2 neuroinvasion has been proposed. The virus also invades through the gastrointestinal system and reaches the CNS directly through afferent fibres of the intestinal vagus nerve [21].

Hematogenous route

Lung infection of SARS-CoV-2 damages the epithelial barrier and endothelium, which facilitates the dissemination of the virus to circulation and various organ systems including the brain. Studies have shown strong evidence of RNAemia with plasma viremia in COVID-19 patients [22]. These studies have also demonstrated that SARS-CoV-2 viremia is a strong independent marker of COVID-19 disease severity and outcome. Viremia plays a role in broad tissue damage and persistent expression of several SARS-CoV-2 entry factors [22, 23]. An autopsy ultrastructural analysis of tissue from a COVID-19 patient revealed viral-like particles in endothelial cells, pericytes, and active budding across endothelium, strongly supporting hematogenous endothelial neuroinvasion [24].

Blood-brain barrier

ACE2 expression in microvascular endothelial cells of BBB provides an entry point for SARS-CoV-2 to the CNS. The virus may cross the BBB through paracellular, transcellular, or by a trojan horse mechanism. One animal study strongly suggested that the S1 protein can cross the murine BBB through adsorptive transcytosis [25]. Evidence from in vivo and in vitro BBB model studies demonstrated that SARS-CoV-2 crosses the BBB via transcellular paths utilising basement membrane disruption without evidence of tight junction alterations [26].

Long-COVID and its neurological complications Brain fog

‘Foggy brain state’ is the accepted term for a condition in which the cognitive functioning of an individual is not as sharp as usual. First formally described in the late 1980s and 1990s through reports of patients suffering from debilitating fatigue and significant cognitive dysfunction following the onset of an infection, a similar ‘brain fog’ has now been reported in many patients diagnosed with long-COVID [27, 28]. Initially this was termed chronic fatigue syndrome, however it lacked consensus among experts regarding its terminology, definitions, and importantly the criteria for diagnosis. Long-COVID is now undergoing a similar process in terms of definitions and diagnostic criteria, while also contending with the difficulties associated with including ‘brain fog’ as a symptom. Despite these difficulties a growing body of evidence suggests that in chronic fatigue syndrome there is central nervous system, autonomic nervous system, and immune system involvement, and there may be similar mechanisms at work in the long-COVID of today [29,30,31].

Although often lacking diagnostic clarity, brain fog is usually associated with lack of intellectual clarity, chronic mental fatigue, confusion, alongside short and long-term memory loss. Other symptoms include disability, dizziness, anxiety, reduced concentration, and impaired cognitive and mental abilities [32]. Brain fog has been reported in many conditions, such as hypothyroidism [33], menopause [34], coeliac disease [35], and now it has been closely associated with COVID-19 infection [36]. Diagnosis of brain fog often involves the patient reporting subjective experiences through symptom questionnaires and researchers or clinicians comparing that to objective performance-based tests used to evaluate patients with other cognitive abnormalities. Unlike the respiratory symptoms, which often become life-threatening, the neurological manifestations of COVID-19 are less noticeable to both patient and physician. Nonetheless, these manifestations are associated with significant disability, reductions in quality of life, and even mortality in COVID-19 patients [32]. Studies demonstrate that nearly 30% of COVID-19 patients experience symptoms such as the discussed ‘brain-fog’ [37] while also experiencing more concrete symptoms including delirium, stroke, and more noticeable cognitive impairments along with physical changes in the brain.

The neurological basis for the occurrence of brain-fog during long-COVID appears to be multifactorial. One potential factor involved is activities of the microglia, which are often implicated in neuroinflammation and neurodegenerative diseases [38,39,40]. This may occur as microglia express Toll-like receptors (TLRs) [41] which may get activated through damage-associated patterns (DAMPs). This process was recently implicated in the progressing of COVID-19 infection [42]. Another possibility is the hypothalmic-pituitary-adrenal axis (HPA axis) which controls the emotional state of an individual. This is typically activated by stress and has been seen to be activated by COVID-19 [43, 44]. It is interesting to note that microglia also express receptors for CRH and get activated when the stress level of the individual is elevated and this interaction has been implicated in the symptoms of COVID-19 [45]. Microglia may interact with mast cells in the brain resulting in their activation, which eventually leads to neuroinflammation [46,47,48]. Activation of mast cells and microglia in the hypothalamus leading to cognitive dysfunction is often seen in patients with mast cell activation syndrome, providing an example of how such an interaction may lead to the symptoms described here [49,50,51]. Psychological stress through CRH receptors directly activates the microglia [52, 53] and has pro-inflammatory effects leading to increased vascular permeability and disruption of the blood-brain barrier (BBB) [54, 55] possibly through the release of IL-6 [56], and CRH [57] which then further worsens the brain inflammation implied by the prior mechanisms by permitting the entry of even more viral particles, cytokines, and other toxic substances. It is now clear that damage to blood vessels and inflammation plays a role in the neurological symptoms described earlier. Additionally, no evidence of direct infection of the brain has been seen in COVID-19 [58], suggesting that the symptoms involved have a predominantly cytokine-mediated pathway. The microglia-activated neuroinflammation could explain both increased neurologic [59] and frequent psychiatric [60] disorders in patients with COVID-19, an increase which is also seen in patients with long-COVID syndrome [61].

Sleep disturbances

Altered sleep patterns are observed to be one of the most commonly reported neurological symptoms of long-COVID [62]. Huang et al. [63] reported that among the 1733 patients with long-COVID, 26% experienced sleep disturbances. In another study, investigators evaluated 251 COVID-19 survivors 1 month after hospital discharge, and among them 41.8% reported to have experienced insomnia at some point. Although this insomnia was improved among 25.5% of the initial population when re-evaluated 3 months post-discharge. The remaining patients however still experienced sleep disturbances marking a substantial portion of the initial study population [64]. Disturbances in sleep due to COVID-19 are of great concern because of bidirectional associations between mental health problems and sleep disturbance. This is likely a contributing factor for the mental health complications earlier described as related to COVID-19 [65]. Scarpelli et al. [66], argue that patients who have reported sleep disturbances, nightmares, and lucid dreaming could be suffering from long-COVID and that sleep disturbance are either directly a symptom or may be a reflection of the stress of the life-altering pandemic. In another study, 402 COVID-19 patients were assessed for insomnia 1 month after the hospital treatment and among them, 40% reported sleep disturbances [67]. Another follow-up study included 94 COVID-19 survivors who experienced COVID-19-related pneumonia with respiratory failure and when they tested for insomnia in 4th month of their discharge, 31% of them reported that they had insomnia [68]. In a similar follow-up study, 478 COVID-19 patients were evaluated for insomnia 4 months post-discharge and 53.6% of them reported they had insomnia [69].

Another observational follow-up study conducted on 797 COVID-19 survivors revealed that at 6 months of post-discharge, 4.9% of these patients have reported sleep disturbance [70]. However, another 6-month follow-up study that evaluated 796 patients found that at the time of assessment, 23.2% of these patients had neuropsychiatric symptoms such as a sleep disorder [71]. A study, conducted on 165 patients 6 months after their hospitalisation due to COVID-19 revealed that 31.5% of the patients had sleep disorders as one of the common neuropsychiatric symptoms [72]. Poor sleep quality was seen in 34.5% of 1142 COVID-19 patients who were evaluated for sleep quality at 7 months after their hospital discharge [73]. Another retrospective study reported that among the 23,6379 COVID-19 patients included in the study, 5.42% reported insomnia 6 months after the initial COVID-19 infection [74]. A prospective longitudinal study that assessed 1077 COVID-19 patients for sleep quality at 5.9 months after their hospital discharge found that 41.8% had worsened sleep quality along with other neurological symptoms [75]. A systematic review and meta-analysis that included 29 studies involving 13,935 patients revealed that the prevalence of sleep disturbance was 46%, poor sleep quality was 56%, the prevalence of insomnia was 38% and the prevalence of excessive daytime sleepiness was 14% [76]. Another systematic review (of 33 reports involving 282,711 participants with long-COVID) aimed to summarise the main psychiatric manifestations of long-COVID revealed that sleep disturbances, depression, post-traumatic symptoms (PTS), anxiety, and cognitive impairment were the common psychiatric manifestations in long-COVID patients after 4 weeks from COVID-19 infection recovery [77].

Memory impairment

A case series that addresses cognition in post-COVID-19 patients demonstrated evidence for neurocognitive deficits involving encoding and verbal fluency after severe COVID-19 infection [78]. Another case study involving a 62-year-old female without any psychiatric history developed psychiatric symptoms during COVID-19 infection and treatment, emphasising the correlation between the COVID-19 infection and confused state, short and long-term memory deficits, and delirium in patients as comorbidity, especially in elderly [79]. A case report from 3 patients identified neuropsychological symptoms such as inattention, executive function, and memory difficulties as potential post-infective complications [80]. Godoy-González et al. [