Chronic Fatigue, Depression And Anxiety Symptoms in Long COVID Are Strongly Predicted By Neuroimmune And Neuro-Oxidative Pathways Which Are Caused By The Inflammation During Acute Infection Part 1

Abstract: Background: Long-term coronavirus disease 2019 (long COVID) is associated with psychosomatic (chronic fatigue syndrome and somatic symptoms) and affective (depression and anxiety)  symptoms. The severity of the long COVID physio-affective phenome is primarily predicted by increased peak body temperature (BT) and lowered oxygen saturation (SpO2) during the acute infectious phase. This study aims to delineate whether the association of BT and SpO2 during the acute phase and the long COVID physio-affective phenome is mediated by neurotoxicity (NT) resulting from activated immune-inflammatory and oxidative stress pathways. Methods: We recruited 86 patients with long COVID (3–4 months after the acute phase) and 39 healthy controls and assessed serum C-reactive protein (CRP), caspase 1, interleukin (IL) 1β, IL-18, IL-10, myeloperoxidase (MPO),  advanced oxidation protein products (AOPPs), total antioxidant capacity (TAC), and calcium (Ca), as well as peak BT and SpO2 during the acute phase. Results: Cluster analysis revealed that a significant part (34.9%) of long COVID patients (n=30) show a highly elevated NT index as computed based on IL-1β, IL-18, caspase 1, CRP, MPO, and AOPPs. Partial least squares analysis showed that the NT index could explain 61.6% of the variance in the physio-affective phenome of long COVID,  lowered Ca, and peak BT/SpO2 in the acute phase and prior vaccinations with AstraZeneca or Pfizer. The most important predictors of the physio-affective phenome are Ca, CRP, IL-1β, AOPPs, and MPO. Conclusion: The infection–immune–inflammatory core of acute COVID-19 strongly predicts the development of physio-affective symptoms 3–4 months later, and these effects are partly mediated by neuro-immune and neuro-oxidative pathways.

Cistanche can act as an anti-fatigue and stamina enhancer, and experimental studies have shown that the decoction of Cistanche tubulosa could effectively protect the liver hepatocytes and endothelial cells damaged in weight-bearing swimming mice, upregulate the expression of NOS3, and promote hepatic glycogen synthesis, thus exerting anti-fatigue efficacy. Phenylethanoid glycoside-rich Cistanche tubulosa extract could significantly reduce the serum creatine kinase, lactate dehydrogenase, and lactate levels, and increase the hemoglobin (HB) and glucose levels in ICR mice, and this could play an anti-fatigue role by decreasing the muscle damage and delaying the lactic acid enrichment for energy storage in mice. Compound Cistanche Tubulosa Tablets significantly prolonged the weight-bearing swimming time, increased the hepatic glycogen reserve, and decreased the serum urea level after exercise in mice, showing its anti-fatigue effect. The decoction of Cistanchis can improve endurance and accelerate the elimination of fatigue in exercising mice, and can also reduce the elevation of serum creatine kinase after load exercise and keep the ultrastructure of skeletal muscle of mice normal after exercise, which indicates that it has the effects of enhancing physical strength and anti-fatigue. Cistanchis also significantly prolonged the survival time of nitrite-poisoned mice and enhanced the tolerance against hypoxia and fatigue.

Click on Why are We Tired All the Time

【For more info:george.deng@wecistanche.com / WhatApp:8613632399501】

Keywords: long COVID; depression; chronic fatigue syndrome; neuro-immune; inflammation;  oxidative stress; antioxidants

1. Introduction

While the infection rate for severe acute respiratory syndrome coronavirus 2 (SARSCoV-2) has fallen globally [1,2], a new concern has arisen among many post-infection individuals with long COVID symptoms [3]. Several studies have reported a cluster of persistent symptoms extending beyond full recovery from coronavirus disease (COVID-19) [4–6]. These symptoms appear after up to two (infection-related), three (acute post-COVID), six (prolonged post-COVID), or more (chronic post-COVID) months following the acute infectious phase [7]. Many individuals (74-87.4%) with long COVID suffer from a variety of mental and physio-somatic symptoms after recovery from the acute phase [8,9], including chronic fatigue, affective symptoms (low mood and anxiety), cognitive dysfunctions, and sleep disturbances, along with bodily manifestations, such as autonomic symptoms, muscle pain, muscle tension, headache, a flu-like malaise, gastro-intestinal symptoms (GISs),  shortening of breath, persistent cough, and chest pain [5,6,8,10–19]. Available treatments are limited, as there is no clear understanding of the pathophysiological mechanism.

The immunopathogenesis of the acute infectious phase of COVID-19 comprises activation of the cytokine network with induction of interferons; interleukin (IL) 1, IL-6, IL-12, IL-18, and tumor necrosis factor (TNF) α signaling; activation of the nucleotide-binding domain, leucine-rich repeat, and pyrin domain-containing protein 3 (NLRP3) inflammasome;  and activation of key antiviral pathways, toll-like-receptor cascades, and NOD-like receptor signaling [20–23]. The NLRP3 is a key component of the innate immune system and an intracellular sensor that is induced by pathogen-associated and damage-associated molecular patterns [24,25]. NLRP3 activation causes increased levels of caspase 1, IL-1β, and IL-18  and cell death or pyroptosis [26,27]. Increased peak body temperature (BT) and lowered oxygen saturation (SpO2) reflect the severity of the immune–inflammatory response during acute COVID and predict critical COVID-19 and increased mortality [28–30].

Recently, we introduced a new concept—namely, the physio-affective phenome of acute and long COVID—to describe the physio-somatic (fatigue and somatic symptoms)  and affective (depression and anxiety) symptoms in both illnesses (see [29,31,32] for reviews). Thus, in acute and long COVID, a validated latent construct could be derived from depression, anxiety, chronic fatigue, and multiple physio-somatic symptoms; therefore, this factor score reflects the severity of the intertwined increases in physio-somatic and affective symptoms [29,31,32].

In acute COVID-19, this physio-affective phenome was largely explained by the cumulative effects of increased pro-inflammatory cytokines, including IL-6 and soluble advanced glycation products (sRAGEs); changes in acute-phase proteins, including C-reactive protein (CRP) and albumin; lowered calcium (Ca); pneumonia as indicated by chest computerized tomography scan abnormalities (CCTAs); and diminished SpO2 [31]. Moreover, elevated peak BT and lowered SpO2 during the acute phase of infection predict the physio-affective phenome of long COVID [29]. In long COVID, the severity of the physio-affective phenome was also predicted by increased oxidative toxicity as indicated by increased levels of malondialdehyde (MDA), protein carbonyls (PCs), myeloperoxidase (MPO), nitric oxide (NO), as well as lowered antioxidant defenses as indicated by lowered zinc and glutathione peroxidase (Gpx) [32]. Moreover, the oxidative toxicity/antioxidant ratio was significantly predicted by increased peak BT and lowered SpO2 [32]. These results indicate that the immune–inflammatory response in acute COVID-19 predicts, at least in part, the physio-affective symptoms of long COVID and that these effects are partly mediated by increased neuro-oxidative toxicity [31,32].

It is important to note that neuro-immune and nitro-oxidative toxicity play key roles in the pathophysiology of chronic fatigue syndrome (CFS), major depression (MDD), and generalized anxiety disorder (GAD) [33–36]. There is now also evidence that upregulation of the NLRP3 inflammasome plays a key role in MDD [37,38], fatigue [39,40], cognitive impairments [41], and anxiety [42,43]. Nevertheless, no data are showing whether the NLRP3 inflammasome and lowered Ca are involved in the physio-affective phenome of long COVID.

Hence, this study was performed to determine (a) the effects of the NLRP3 inflammasome (IL-1β, IL-18, and caspase 1), inflammatory response (CRP), oxidative stress (advanced oxidation protein products (AOPPs), and MPO), total antioxidant capacity (TAC),  and lowered Ca on the physio-affective phenome of long COVID; and (b) whether neuroimmune and neuro-oxidative toxicity is predicted by lowered SpO2 and increased peak BT  during the acute phase. The specific hypotheses were that the physio-affective phenome of long COVID is predicted by lowered SpO2 and increased peak BT and that these effects are,  at least in part, mediated by increased neurotoxicity (NT) due to increased caspase 1, IL-1β, IL-18, CRP, MPO, and AOPP and lowered TAC and Ca levels. The data were analyzed using the new precision nomothetic psychiatry approach [30,44], which involved the construction of a new endophenotype class of long COVID.

2. Participants and Methods 

2.1. Participants 

We used the World Health Organization (WHO, Geneva, Switzerland) criteria to diagnose long COVID patients [45]. These criteria were that: (a) daily life activities of the patients should be influenced minimally by two symptoms (fatigue, impairment of memory or concentration, achy muscles, absence of smell or taste senses, affective symptoms, and cognitive impairment); (b) patients should have had confirmed infection with COVID-19; (c) the symptoms should persist beyond the acute phase or should become apparent 2–3 months later; and (d) the symptoms should last for at least two months [45] and be present 3–4 months after recovery. Accordingly, 86 long COVID patients participated in the current study, clustered into two groups: long COVID with low (n = 56) and high (n = 30) NT (as defined below). All Long COVID patients sought professional assistance for post-COVID symptoms and exhibited varying degrees of reduced functioning and productivity, as well as an increased inability to perform daily tasks as effectively as they did before the COVID infection. Additionally, we recruited 39 healthy controls who did not show any clinical signs of infection and showed negative test results in reverse transcription real-time polymerase chain reaction (rRT-PCR) tests when included in the study. All participants were recruited between September and December 2021. The present study involved a combined methodology; namely, investigation of the impact of acute phase COVID-19 on long COVID symptoms (retrospective study design) and comparison of abnormalities in long COVID patients versus healthy controls (case–control design).

In the acute COVID-19 phase, specialized clinicians and virologists diagnosed patients with a positive COVID-19 infection based on: (a) severe symptoms of infection, such as fever, cough, shortness of breath, and the loss of smell and taste senses; (b) positive rRT-PCR test results; and (c) positive immunoglobulin M (IgM) against SARS-CoV-2. All patients were quarantined and treated at several hospitals and specialized centers within Al-Najaf City, Iraq: Al-Sader Medical City of Najaf, Al-Hakeem General Hospital, Al-Zahraa Teaching Hospital for Maternity and Pediatrics, Imam Sajjad Hospital, Hassan Halos AlHatmy Hospital for Transmitted Diseases, the Middle Euphrates Center for Cancer, and Al-Najaf teaching hospital.

All participants with long COVID were free of any signs of acute COVID-19—namely,  dry cough, sore throat, shortness of breath, fever, night sweats, and chills—and, before participating in the current study, patients and controls showed negative rRT-PCR test results. Around 33% of the participants in the control group had minor mental symptoms,  such as low mood, anxiety, and fatigue, resulting from the quarantine period and lack of social activities, which may also affect patients with long COVID. We excluded subjects with a previous major depressive episode, bipolar disorder, dysthymia, GAD, panic disorder,  schizo-affective disorder, schizophrenia, psycho-organic syndrome, substance use disorders (except tobacco use disorder (TUD)), or neurodegenerative and inflammatory diseases,  such as CFS [46]. We also excluded subjects with liver and renal disease, Parkinson’s or Alzheimer’s disease, multiple sclerosis, stroke, or systemic (auto)immune diseases, such as diabetes mellitus, psoriasis, rheumatoid arthritis, inflammatory bowel disease, and scleroderma, as well as pregnant and breastfeeding women.

The present study was designed and performed in line with Iraqi and international ethical and privacy laws, including the World Medical Association’s Declaration of Helsinki, the Belmont Report, the Council for International Organizations of Medical Sciences (CIOMS) Guideline, and the International Conference on Harmonization of Good Clinical Practice. Our institutional review board adheres to the International Guidelines for Human

Research Safety (ICH-GCP). We obtained written consent from all participants, parents,  or any legally responsible person before involvement in our study. The institutional ethics board and the Najaf Health Directorate—Training and Human Development Center approved our research according to their documents with the numbers 8241/2021 and 18378/2021, respectively.

2.2. Clinical Measurements

Three to four months after recovery from acute COVID-19, a semi-structured interview was conducted by a senior psychiatrist to obtain the socio-demographic and clinical characteristics of all participants. The psychiatrist assessed several symptom domains,  including chronic fatigue and fibromyalgia symptoms utilizing the Fibro Fatigue (FF)  scale [47], the severity of depression using the Hamilton Depression Rating Scale (HAMD) [48], and the Beck Depression Inventory-II (BDI-II) [49], and severity of anxiety symptoms using the Hamilton Anxiety Rating Scale (HAMA) [50]. Moreover, we utilized those rating scale items to derive subdomains for the major symptoms. We made two subdomains with the HAMD; the first was pure depressive symptoms (pure HAMD), which was the sum of sad mood, feelings of guilt, suicidal thoughts, and loss of interest; and the second was physio-somatic HAMD (physiosom HAMD), which was the sum of somatic anxiety, gastrointestinal (GIS) anxiety, genitourinary anxiety, and hypochondriasis. Likewise, two subdomains of the HAMA were computed; namely, pure anxiety symptoms (pure HAMA), the sum of anxious mood, tension, fears, anxiety, and anxious behavior during the interview, and physio-somatic HAMA symptoms (physiosom HAMA), the sum of somatic sensory, cardiovascular, genitourinary, and autonomic symptoms, as well as GIS. Furthermore, after omitting items reflecting cognitive and affective symptoms in the FF  scale, a pure physio-somatic FF (pure FF) score was computed as the sum of muscular pain, muscle tension, fatigue, autonomous symptoms, GIS, headache, and flu-like malaise. We also computed the sum of all pure depressive BDI-II (pure BDI) symptoms—thus excluding physio-somatic symptoms—including sadness, discouragement about the future,  feeling like a failure, dissatisfaction, feeling guilty, feeling punished, self-disappointment, self-criticism, suicidal ideation, crying, loss of interest, difficulties with decisions, and work inhibition. In our previous studies, the physio-affective phenome was defined [29,31] as the first factor extracted from the pure FF and BDI scores and the pure and physiosom HAMA  and HAMD scores. The DSM-5 criteria were used to diagnose TUD. Weight in kilograms was divided by height in meters and squared to calculate the body mass index (BMI).

We used the patients' records to determine the lowest SpO2 and peak BT values, which were obtained during hospitalization for the acute infectious phase. The assessments were undertaken by a well-trained paramedical professional who employed an electronic oximeter manufactured by Shenzhen Jumper Medical Equipment Co. Ltd. and a sublingual digital thermometer with a beep sound. By subtracting the z-transformed SpO2 values from the z-transformed peak BT values, we generated a new index involving both lowered SpO2  and increased peak BT (dubbed the TO2 index). Additionally, we recorded the different types of vaccines received by the patients; namely, AstraZeneca (Cambridge, UK), Pfizer (New York, NY, USA), and Sinopharm (Beijing, China).

2.3. Assays

Five milliliters of venous blood were sampled using disposable syringes at 7:30–9:00 am after an overnight fast. The blood was then transferred directly to serum tubes. We avoided any hemolyzed, lipemic, and icteric blood samples. All tubes were centrifuged at 3000 rpm after 10 min incubation at room temperature. Then, we produced three aliquots of serum,  which were stored at −70 ◦C in Eppendorf tubes until they were thawed for assaying. We employed ELISA kits provided by Nanjing Pars Biochem Co., Ltd. (Nanjing, China) to assess serum levels of IL-1β, IL-18, IL-10, caspase 1, MPO, TAC, and AOPPs (albumin ratio). Total serum Ca was measured spectrophotometrically using ready-to-use kits obtained from Agappe Diagnostics Ltd., Cham, Switzerland. We used a z unit-based composite score to determine a new NT index by computing the z transformation of IL-1β (z IL-1β) +  z IL-18 + z caspase 1 + z MPO + z AOPP + z CRP.

2.4. Statistical Analysis

In the present study, IBM SPSS software, version 28, was used to carry out the statistical analyses. We analyzed variance (ANOVA) to delineate the differences in continuous variables among the study groups and an analysis of contingency tables to examine associations between nominal variables. Pearson product–moment correlation coefficients were used to analyze the relationship between two scale variables. Multivariate and univariate general linear models (GLM) were used to examine the association between clinical and biomarker data and the diagnostic categories while allowing or controlling for age, TUD, sex, BMI, and education. Estimated marginal mean (SE) values were computed and multiple group mean differences were assessed using Fisher’s protected (the omnibus test is significant) least significant difference (LSD). The ability of biomarkers and clinical variables to predict the physio-affective symptoms was determined through multiple regression analysis. We utilized an automated stepwise approach with a p-to-enter of 0.05 and a p-to-remove of 0.06. For each of the explanatory variables, we computed the standardized beta-coefficients, t-statistics, and exact p-value, along with F-statistics and the total variance explained (R2 ). Furthermore, we used the variance inflation factor and tolerance to examine multicollinearity. The heteroskedasticity was checked by employing the White and modified Breusch–Pagan tests. We used cluster analysis (two-step) and followed the precision nomothetic approach [30] to construct endophenotype classes of patients with long COVID based on a combination of SpO2, peak BT, and the neurotoxic biomarkers in a z unit-based composite score (z IL-1β + z IL-18 + z caspase 1 + z MPO + z AOPP + z CRP) dubbed the NT index. The cluster solution was considered adequate when the silhouette measure of cohesion and separation was > 0.5. Canonical correlation analysis was employed to investigate the correlations between two sets of variables; namely,  physio-affective symptoms 3–4 months after acute COVID infection as the dependent variables and biomarkers as explanatory variables. The variance explained by the canonical variables of both sets was computed, as well as the variance in the canonical dependent variable set explained by the independent canonical variable set. We accepted the canonical components when the explained variance of both sets was > 0.50 and when all canonical loadings were >0.5.

Partial least squares (PLS) analysis was used to study the causative relationships between SARS-CoV-2 infection, peak BT, and lowest SpO2 during the acute phase of the disease and the physio-affective phenome of long COVID, whereby the effects of the input variables were partly mediated by NT and other biomarkers. All input variables were inputted as single indicators, and the output variable was a latent vector extracted from the values for the pure and physiosom HAMA and HAMD and pure FF and BDI (the physio affective phenome). Complete PLS analysis was conducted only when the outer and inner models met the following prespecified quality criteria: (a) all loadings on the extracted latent vector were >0.6 at p < 0.001; (b) the output latent vector showed high construct and convergence validity, as indicated by rho A > 0.8, Cronbach’s alpha > 0.7, composite reliability > 0.7, and average variance extracted (AVE) > 0.5; (c) blindfolding demonstrated that the construct’s cross-validated redundancy was sufficient; (d) confirmatory tetrad analysis (CTA) demonstrated that the latent vector extracted from the rating scale scores was not misspecified as a reflective model; (e) the model’s prediction performance as measured by PLS Predict was satisfactory; and (f) the model fit was <0.08 in terms of standardized root squared residual (SRMR) values. If all model quality data conformed with the prespecified criteria, we conducted a complete PLS-SEM pathway analysis with 5000 bootstrap samples and calculated the path coefficients (with p-values), as well as specific and total indirect (mediated) effects and total effects. The primary statistical analyses were the results of multiple regression analyses, particularly those conducted with PLS-SEM.

3. Results 

3.1. Socio-Demographic Data 

We employed a two-step cluster analysis to classify the patients with long COVID  into two groups using the peak BT, SpO2, and the NT index (IL-1β + IL-18 + caspase 1+ MPO + AOPP + CRP) to develop a new biomarker-derived endophenotype class within the long COVID patient group. The diagnosis of long COVID and the peak BT, lowered SpO2, and NT composite score were entered into the cluster analysis as the nominal variable and the continuous variables, respectively. According to the silhouette measure of cohesion and separation of 0.53, the quality of the clusters was adequate. Three clusters were derived; namely, the healthy control sample (n = 39); patients with high peak BT, lowered SpO2, and an increased NT index (dubbed high TO-NT—T for temperature, O for SpO2, and NT for neurotoxicity) (n = 30); and patients (n = 56) with less pronounced changes in these biomarkers (dubbed low TO-NT). Therefore, long COVID patients were classified based on the combination of two acute COVID-19 phase markers and blood biomarkers 3–4 months after clinical recovery.

Socio-demographic and clinical data for the three groups are presented in Table 1. SpO2, peak BT, the TO2 index, and the NT index were significantly different between the high and low TO-NT clusters. No significant changes were detected between these groups in terms of age, sex, BMI, marital state, smoking status, residency status, vaccination state, or education.

3.2. Differences in Psychiatric Rating Scales between Study Groups

Table 2 shows the measurements of the rating scales; namely, the total and subdomain scores. There were significant differences among the three study groups in total and pure HAMD, HAMA, BDI, and FF and physiosom HAMD and HAMA scores. Table 2 shows that the scores of all scales and subscales, except pure HAMA, increased from controls to the low TO-NT group to the high TO-NT group. The pure HAMA score was not significantly different between the two patient groups but was higher in the latter than in the controls.

3.3. Differences in Biomarkers between Study Groups

The measurements of the biomarkers in both classes of patients with long COVID  versus healthy controls are displayed in Table 3. CRP and AOPP were significantly different between the three study groups. Caspase 1 was significantly higher in the high TO-NT class than in the two other groups. There was a trend towards higher IL-1β levels in the high TO-NT group than in controls and significantly higher IL-10 in both long COVID  groups than in controls. Total Ca was lower in patients than in controls.

【For more info:george.deng@wecistanche.com / WhatApp:8613632399501】