Introduction
Long COVID has been defined by the British health authority (National Institute for
Health and Care Excellence (NICE) as signs and symptoms that develop during or after
infection with COVID-19 and last longer than 4 weeks and cannot be explained by any
other diagnosis. If symptoms remain for more than 12 weeks, they are defined as Post
COVID [1 ]. Symptoms include persistent
fatigue, diffuse myalgia, depressive symptoms, non-restorative sleep, imbalance of
the immune, hematological, pulmonary, cardiovascular, gastrointestinal, hepatic,
renal, skeletomuscular and nervous systems, as well as depression and anxiety [2 ]. The exact molecular mechanisms behind these
symptoms are not understood yet but most likely heterogeneous ([Fig. 1 ]). They might include direct or
indirect consequences of the infection with SARS-CoV-2 [3 ]. Autoimmunity due to targeting of
self-antigens due to impairment in the regulatory T cell response or molecular
mimicry may be another explanation. An additional proposed mechanism for persisting
symptoms is reactivation of latent viruses in the body [4 ]. Also, rheological abnormalities, such as
blood viscosity and red blood cell (RBC) deformations were shown to be caused by
COVID-19 infection [5 ]
[6 ] (Toepfner et al., under review). As reviewed
in [4 ], these different processes are not
mutually exclusive and could exist in combination.
Fig. 1 Potential explanations for fatigue symptoms in Long/Post
COVID.
As symptoms of Long/Post COVID and adrenal insufficiency are partly overlapping this
issue should be taken into account [7 ]. It was
shown that the adrenal gland is a target of SARS-CoV-2, which potentially may
directly or indirectly lead to adrenal insufficiency [8 ]
[9 ]. In
addition, adrenal insufficiency may be caused by glucocorticoid treatment over an
extended period if then suddenly abrubted.
We have previously suggested using extracorporeal apheresis for treatment of Post
COVID (at least 12 weeks after positive PCR) [10 ]. This treatment has received a lot of attention. However, to date,
there are still no controlled randomized trials and it is still discussed whether
this method is valid for treatment of people suffering from long-term symptoms after
an infection with SARS-CoV-2 [11 ]
[12 ]. In the current paper, we will discuss
which experiments were performed to date and relate to our own clinical
experience.
Post-infection chronic fatigue syndrome
Fatigue or muscle weakness are the most commonly reported persistent symptoms in
Long/Post COVID, affecting about half of the patients for at least six
months after the acute disease [2 ].
Long/Post COVID is not linked to the severity of the acute phase of the
disease and often it appears even after mild or moderate initial illness. It is
more often identified in women [13 ]
[14 ]. The symptoms are similar to those seen
in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS)
or those observed after the Russian or Spanish flu [15 ]
[16 ]. The pathology of ME/CFS is not known, but probably it
results from the dysregulation of multiple systems in response to a particular
trigger.
Due to its resemblance with Long/Post COVID, a common etiopathogenesis
has been suggested [17 ]
[18 ]. Unexplainable post-acute infection
syndrome (PAIS) is not unique to SARS-CoV-2 infections, but has been reported
for all kinds of infectious agents, including bacteria, viruses, and parasites
For example, PAIS has been reported for a number of viruses, such as
Epstein-Barr (EBV), Ebola, Dengue, polio, SARS-CoV-1, Chikungunya, West Nile,
Ross River, and enteroviruses [4 ].
Non-viral pathogens known to trigger PAIS are Coxiella burnetti ,
Borrelia , and Giardia lamblia
[4 ]. Due to missing knowledge about the underlying mechanisms, there
is often a poor recognition of these conditions in clinical practice.
Molecular mimicry and autoantibodies
Underlying pathological mechanisms of ME/CFS are largely unknown, but the
presence of autoantibodies, cytokine pattern deviations and the presentation of
cognitive and autonomic nervous system related symptoms provide evidence for
ME/CFS being an immunological disorder with elements of autoimmunity
triggered by the preceding infection [19 ]
. Certain pathogens have the ability to escape from host immune response. Due to
host protein mimicry no antibodies against the pathogen is created [20 ]. Oppositely, similarity between
pathogenic antigens and host proteins may lead to immune cross reactivity,
whereby the reaction of the immune system towards the pathogenic antigens may
harm similar human proteins, essentially causing autoimmune disease [21 ]. Increased levels of autoantibodies
binding to G-protein coupled receptors (GPCRs), such as adrenergic and
muscarinic receptors, were found in ~30% of
ME/CFS-patients [22 ]. Furthermore,
autoantibodies against neuronal proteins, such as serotonin receptors, glial
fibrillary antigen (GFAP) and S100, have been observed [23 ]. In Norway, clinical trials showed that
in a subgroup of ME/CFS patients, prolonged B-cell depletion with
rituximab maintenance infusions was associated with sustained clinical response
[24 ]
[25 ] although later there were problems confirming these results [26 ].
A number of studies have suggested that molecular mimicry may also play an
important role in autoimmunity generation in COVID-19 [27 ]
[28 ]. For example, hexapeptides of the SARS-CoV-2 spike glycoprotein
(S) and nucleocapsid protein (N) show notable similarity with three human
proteins, DAB1, AIFM, and SURF1, involved in neuron development and
mitochondrial metabolism [29 ].
Cross-reactivity of several other SARS-CoV-2 amino acid sequences with human
proteins including proteins on the plasma membrane of olfactory neurons and
endothelial cells, and in the cytoplasm of B cells and macrophages have also
been detected [27 ].
Indeed, autoantibodies against various epitopes have been detected in patients
with Long/Post COVID, suggesting that chronic fatigue in these patients
may be caused by autoimmune mechanisms. More studies showed that antibodies
against type I interferons (IFNs) were present in ~10% of
patients with a severe COVID-19 course and correlated with increased CRP and
lower lymphocyte counts [30 ]
[31 ]
[32 ]. Such autoantibodies neutralize the ability of the corresponding
type I IFNs to block SARS-CoV-2 infection explaining why these antibodies are
not observed in asymptomatic patients or those with a mild course [30 ].
As in ME/CFS, antibodies against GPCRs were also observed in
Long/Post COVID patients [33 ]
[34 ]. In a study with 31 patients suffering
from different Long/Post COVID symptoms after recovery from the acute
phase of the disease, autoantibodies against these receptors were found in all
patients [34 ]. This explains why the
aptamer BC007 for neutralization of autoantibodies against GPCRs has been
suggested for treatment of Post COVID symptoms and currently BC007 has been used
in individual patients with Post COVID.
Until now, no correlation between GPCR autoantibody levels and disease severity
could be observed [35 ]. Therefore, the
role of increased β adrenergic and muscarinic cholinergic receptor
autoantibodies in the pathogenesis of ME/CFS and Long/Post COVID
is still uncertain and further research is needed to evaluate the clinical
significance of these findings.
Viral/bacterial reactivation due to COVID-19
Reactivation of cytomegalovirus (CMV) and other herpesviridae, such as EBV and
HHV6, in critically ill patients is common and associated with an increased risk
of secondary infections and mortality [36 ]
[37 ]
[38 ]
[39 ]. Similar observations have been noticed for COVID-19 patients,
where direct pathogenic effects, an unregulated host response and the use of
strong immunosuppressants (tocilizumab) or high doses of steroids may induce
immunosuppression.
In a study from Italy with 431 COVID-19 patients admitted to the intensive care
unit (ICU), CMV blood reactivation was observed in ~20% of the
patients [40 ]. The severity and the
occurrence of secondary bacterial infections were associated with an increased
risk for this CMV reactivation. On the other hand, the CMV reactivation did not
affect the outcome of the patients [40 ].
Conversely, a study from China showed that ~25% of COVID-19
patients had EBV reactivation. The EBV reactivation was associated with age and
female gender and a tendency of a higher mortality rate [41 ]. As compared to patients with COVID-19
who did not receive anti-EBV therapy with ganciclovir, ganciclovir-treated
patients reportedly had improved survival rate [41 ].
In another pilot observational study from Ukraine, 88 COVID-19 patients were
recruited, including 68 subjects with reactivation of herpes viruses and 20
subjects without detectable DNA of herpesviruses. Patients with
Long/Post COVID manifestations presented with reactivation of EBV in
42.6%, HHV6 in 25.0%, and EBV plus HHV6 in 32.4% of the
cases. Compared with controls, patients with herpes virus infections presented
with significantly more symptoms of Long/Post COVID, elevated CRP and
D-dimer, and suppressed cellular immune response [42 ]. These results indicate a potential
involvement of reactivated herpes virus infections, in severe COVID-19 and
formation of Long/Post COVID. Reactivation of hepatitis B virus (HBV) in
COVID-19 patients has also been observed in a smaller number of cases [43 ]
[44 ].
About one-third of the world’s population is thought to be infected with
latent Mycobacterium tuberculosis . Both previous and newly developed
tuberculosis (TB) infections are risk factors for COVID-19 and are associated
with poor outcomes. Clinical evidence suggests that SARS-CoV-2 infection may
predispose patients to TB infection or may lead to reactivation of latent
disease. Similarly, underlying TB disease has been reported to worsen COVID-19
[45 ].
A chronic stimulation of the immune system due to a persistent infection, as
observed in other diseases, has not been observed for SARS-CoV-2 to date [46 ] but cannot be ruled out. In order to
further elucidate the role of persisting infections with SARS-CoV-2 and other
viruses, we recommend a large-scale investigation of possible benefits of
screening and treatment of chronic infections in COVID-19 patients for
prevention of Long/Post COVID.
Other mechanisms responsible for Long/Post COVID
In addition to the potential mechanisms leading to Long/Post COVID
mentioned above, COVID-19 can increase blood viscosity through modulation of
fibrinogen, albumin, lipoproteins, and RBC indices. This may decrease tissue
oxygenation, which can cause cardiovascular and neurological complications in
COVID-19. Increased blood viscosity with or without abnormal RBC function in
COVID-19 may impair tissue oxygenation and thereby foster the development of
cardio-metabolic complications and Long/Post COVID [47 ].
Oxidative stress is defined as an imbalance between elevated levels of cellular
reactive oxygen species (ROS) and low activity and/or levels of
antioxidant defenses [48 ]. ROS production
is an important mechanism for resolving infections, however, excessive ROS
production can result in tissue damage leading to endothelial dysfunction,
increased inflammation, compromised lymphocyte function, and disrupted
neurotransmitter assembly [49 ]
[50 ]
[51 ]. Specifically, oxidative stress can lead to mutations in
mitochondrial DNA, injury to the mitochondrial respiratory chain, activation of
the defense systems in mitochondria and alterations in the membrane permeability
[52 ]. Additionally, oxygen
supplementation used to treat patients with severe COVID-19 can lead to
increased ROS generation in the mitochondria. This damages mitochondrial
complexes and decreases oxidative phosphorylation leading to reduced production
of ATP and elevation in apoptosis rate. Damage of mitochondria by hyperoxia may
reduce antiviral reactions and result in increased tissue damage [52 ]. Recently, excessive inflammation and
oxidative stress have been considered as main factors leading to fibrosis,
thrombosis, autonomic nervous system dysfunction and autoimmunity, which
together result in tissue damage and thus Long/Post COVID [51 ]
[53 ].
Furthermore, in the context of Long/Post COVID, psychological factors
need to be considered. For example, based on data obtained from a large cohort,
a French study suggested that persistent physical symptoms following COVID-19
infection might be associated more with the belief in having been infected with
SARS-CoV-2 than with having laboratory-confirmed COVID-19 infection, which
further emphasizes the heterogeneity of this patient population and the need of
a multidisciplinary approach towards diagnostics and treatment [54 ]. Therefore, non-specific mechanisms
unrelated to SARS-CoV-2 virus infection should also be taken into account [55 ].
Apheresis for treatment of Post COVID
As mentioned above, a number of agents have been proposed to contribute to
Long/Post COVID. Apheresis has been suggested as a way to treat patients
suffering from Post COVID (at least 12 weeks after acute COVID-19). Apheresis is
an extracorporeal method for removal of selected blood components, either
specific cells or specific components of the plasma. The methods for removal of
different pathogenic molecules from plasma were initially developed for removal
of lipids for treatment of severe dyslipidemias and for removal of
autoantibodies. There are several types of apheresis mainly based on three
physical mechanisms: filtration, precipitation and adsorption, whereby lipids,
immunoglobulins, inflammatory agents and further molecules are removed from the
blood. Apheresis can be roughly divided into whole blood and plasma based
methods, in which the cell-rich fraction must first be separated from the
plasma. Whole blood methods are based on the principle of adsorption on either
polyacrylate-coated beads (DALI) or dextran sulfate (Liposorber D). In plasma
methods, protein lowering relies on different principles, such as filtration by
size (MONET, Lipidfiltration, INUSpheresis, and FractioPlas, precipitation of
lipoproteins after pH lowering and the addition of an excess of heparin
(H.E.L.P.), adsorption by antibodies against, for example, apolipoprotein B
(TheraSorb) or against Fc-fragments for removal of autoantibodies [56 ]
[57 ]
[58 ]
[59 ]. While these methods were initially
developed for removal of lipids in severe dyslipidemias, subsequent studies
showed that they have multiple additional beneficial effects due to removal of
high molecular weight proteins and improving blood viscosity, removal of oxLDL
and reducing oxidative stress, removal of cytokines and finally removal of
autoantibodies [60 ]
[61 ]
[62 ]
[63 ]. A recent proteomic
analysis showed that lipoprotein apheresis also removed other proinflammatory
and proatherogenic factors [64 ]. Depending
on the filters used (INUS 30, INUS 50 and TKM58), a reduction and removal of
proatherogenic proteins in different quantities was achieved. This included not
only apolipoproteins, CRP, fibrinogen, and plasminogen (INUS 30, INUS 50) but
also proteins, such as complement factor B (CFAB), protein AMBP, afamin, and the
low affinity immunoglobulin gamma Fc region receptor III-A (FcγRIIIa)
(TKM58) [64 ].
The rationale for using apheresis to treat Post COVID was recently addressed in a
statement by the German Society of Nephrology
(https://www.dgfn.eu/stellungnahmen-details/stellungnahme-der-deutschen-gesellschaft-fuer-nephrologie-zu-lipidapherese-bei-long-oder-post-covid.html,
accessed on the 25th of June 2022). While the German Society of Nephrology saw
no justification for lipid removal in Post COVID, it was concluded that there is
a rationale for autoantibody removal, for example, by immunoadsorption, and
therefore clinical studies to investigate the efficiency of this therapeutic
approach are urgently needed [15 ]
[16 ].
Currently, numerous patients with ME/CFS symptoms due to Post COVID are
treated with apheresis. However, until now, no controlled randomized trials have
been performed and there are no publications on this subject. Nevertheless,
several centers in Germany have started performing apheresis therapy using size
filtration, H.E.L.P., or immunoadsorption approaches with variable results. In
some Post COVID patients, short-term improvements were observed but often for
just a few weeks. Because of different filters used at different centers and no
defined patient groups, it is usually impossible to compare results. Thus, these
observations suggest that apheresis may have benefits in certain patients but in
many other patients, apheresis alone may not be sufficient. Therefore, it has
been proposed that apheresis should be combined with steroid treatment in order
to decrease the amount of autoantibodies produced. This is in accordance with
our knowledge about other antibody-mediated neurological diseases, for example,
multiple sclerosis or chronic inflammatory demyelinating polyneuropathy, where
apheresis therapy alone cannot be successful, but needs accompanying
immunosuppressive/immunomodulatory therapy to inhibit further production
of the pathogenic compounds. Therefore, in the above-mentioned diseases, drug
therapy is always the primary option and apheresis therapy is used only in acute
deterioration [65 ].
In Cham in Germany, a cohort of patients with ME/CFS was treated with a
filtration-based therapeutic apheresis approach, specifically INUSpheresis,
which is known to remove autoantibodies, inflammatory cytokines, oxidated LDLs,
environmental toxins and large molecules, contributing to plasma viscosity. In
order to prevent further production of autoantibodies after their removal, the
treatment protocol included application of prednisolone between the apheresis
treatment sessions. To increase the antioxidant and anti-inflammatory effects,
the protocol also included high doses of vitamin C. Included were 1111 patients
(2009–2022) with ME/CFS (148 following COVID-19, 963 other
infections (e. g., Lyme disease, toxoplasmosis, EBV, or chlamydia),
environmental factors (e. g., organic solvents) or unknown cause).
However, no placebo controls were included. Following this protocol, 56%
of the patients reported to be without symptoms or substantially improved
following 2nd INUSpheresis (TKM58), 64% were without symptoms or
substantially better following 3rd INUSpheresis (TKM58 or INUS 30) and
additional therapy (prednisolone or vitamin C), and 74% were without
symptoms or significantly better 6 months after INUSpheresis with follow-up
therapy. Eleven percent of the ME/CFS patients experienced a moderate
improvement and 15% did not encounter an improvement. The promising
results of this empirical treatment with INUSpheresis in a heterogeneous
population of patients with ME/CFS emphasize the need for a
placebo-controlled study using therapeutic apheresis methods in Post COVID
patients in order to generate scientific evidence to justify introduction of
this treatment approach into clinical practice.
Conclusion
Despite promising experiences with several forms of apheresis in the treatment of
Post COVID, either alone or in combination with other therapies, confirmatory data
on its efficacy from large well-designed interventional studies is still lacking. A
randomized sham-controlled trial is therefore needed and should include a defined
patient group with Post COVID (at least 12 weeks after positive PCR) with fatigue
and other symptoms. Optimally, this should be a 3-arm trial, where the patients
should be randomized for filtration, immunoadsorption, or a sham procedure. All
patients should undergo structured evaluation of fatigue severity using a
questionnaire and a visual analogue scale of tiredness, as well as a detailed
clinical evaluation of the other symptoms before and after the treatment. Also,
blood samples should be collected before and after the treatment in order to measure
blood count, routine biochemical parameters, rheological parameters, markers of
oxidative stress, immunoglobulins A, M and G, autoantibodies against α and
β adrenergic, muscarinic cholinergic receptors, ACE2, MASR, AT1R, ETAR,
ETBR, PAR1, bradykinin receptor and CXCR3 in the plasma and
Real-Time-Deformability-Cytometry of patients’ blood should optimally be
performed.
Based on previous research, it might be postulated that in patients with Post COVID,
extracorporeal apheresis may lead to a significant reduction of autoantibodies while
maintaining an adequate immune response against SARS-CoV-2 and other pathogens.
Moreover, this quantitative decrease in autoantibodies might translate into
alleviation of Post COVID related symptoms and clinical outcomes.
