Brain Hypoxia, Neurocognitive Impairment, And Quality Of Life in People Post‑COVID‑19 Part 1

Abstract

Objective Systemic hypoxia occurs in COVID-19 infection; however, it is unknown if cerebral hypoxia occurs in convalescent individuals. We have evidence from other conditions associated with central nervous system inflammation that hypoxia may occur in the brain. If so, hypoxia could reduce the quality of life and brain function. This study was undertaken to assess if brain hypoxia occurs in individuals after recovery from acute COVID-19 infection and if this hypoxia is associated with neurocognitive impairment and reduced quality of life.

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 Solution of Mentally Exhausted

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

Methods Using frequency-domain near-infrared spectroscopy (fdNIRS), we measured cerebral tissue oxygen saturation  (St O2) (a measure of hypoxia) in participants who had contracted COVID-19 at least 8 weeks before the study visit and healthy controls. We also conducted neuropsychological assessments and health-related quality of life assessments, fatigue, and depression.

Results Fifty-six percent of the post-COVID-19 participants self-reported having persistent symptoms (from a list of 18), with the most reported symptom being fatigue and brain fog. There was a gradation in the decrease of oxyhemoglobin between controls, and normoxic and hypoxic post-COVID-19 groups (31.7±8.3 μM, 27.8±7.0 μM, and 21.1±7.2 μM, respectively, p=0.028, p=0.005, and p=0.081). We detected that 24% of convalescent individuals' post-COVID-19 infection had reduced St O2 in the brain and that this relates to reduced neurological function and quality of life.

Interpretation We believe that the hypoxia reported here will have health consequences for these individuals, and this is reflected in the correlation of hypoxia with greater symptomology. With the fingers technology, combined with neuropsychological assessment, we may be able to identify individuals at risk of hypoxia-related symptomology and target individuals that are likely to respond to treatments aimed at improving cerebral oxygenation.

Keywords Brain hypoxia · Cerebral hypoxia · Post-COVID-19 condition · Cerebral tissue Oxygen saturation · Frequency domain near-infrared spectroscopy

Introduction

Coronavirus disease 2019 (COVID-19) is an acute viral illness caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Initially, it was thought to largely impact the respiratory system. It is now recognized that COVID-19 can severely impact other organ systems,   including the brain, heart, kidneys, liver, skeletal muscle, and skin [1, 2]. About 34% of people receive a neurological or psychiatric diagnosis within 6 months of COVID-19 infection [3]. Persistent symptoms after the apparent elimination of the SARS-CoV-2 have been reported [4–6].  This is termed long COVID, long-haul COVID, or post-acute COVID-19 syndrome (PACS), where after recovery from the acute phase, the individual still feels symptoms [7]. One in five people aged 18–34 years with no chronic medical conditions reported that they have not returned to their baseline health post-acute COVID-19 [5]. Persistent neurological symptoms after acute COVID-19 have also been reported in individuals who had mild disease, the majority who were never hospitalized during their acute COVID-19 illness, were healthy and active before infection, and are less than 50 years old [8]. Furthermore, in patients with comorbidities who had recovered from acute COVID-19, 87% reported persistent symptoms over  60 days post-recovery from acute illness [6].

It is well known that systemic inflammation can induce neuroinflammation and cellular changes, which can impair cognitive function [9, 10], and cognitive impairment has been reported in COVID-19 patients [11]. Additionally, neurological complications in COVID-19 survivors are widely reported, including mild confusion, myalgias, headaches, encephalopathy, dizziness, and loss or changes in taste and smell [3, 12]. It has been known since early in the pandemic that systemic hypoxia is a key feature of COVID-19 infection [13].

However, it is unknown if there is hypoxia in the brain, and if there is, whether this occurs with normal levels of systemic blood oxygenation. We have previously proposed that inflammatory responses within the brain can result in hypoxia and that this hypoxia can worsen inflammation,   thereby creating a hypoxia-inflammation cycle [14]. We have also detected hypoxia in people with multiple sclerosis and primary biliary cholangitis, both conditions appearing to cause inflammation in the brain [15, 16].

We aimed to determine if brain hypoxia exists in individuals post-COVID-19 and if there were associations with neurocognitive impairment and quality of life. We can detect this hypoxia with a measure of cerebral tissue oxygen saturation (St O2) using frequency-domain NIRS (fdNIRS) [15].  This method also provides a measure of light scattering which may relate to changes in mitochondria [17]. In a study investigating NIRS parameters in individuals with acute mountain sickness, there was an increase in light scattering without changes in absorption, and this was indicative of hypoxia-induced cerebral edema [18]. Therefore, our measure of hypoxia may also be related to vasogenic, cellular, osmotic, or interstitial brain edema. We hypothesized that a proportion of convalescent individuals post-COVID-19 will have cortical hypoxia, which will be associated with increased symptomology, and it will occur even with normal arterial oxygen saturation (SaO2).

We report that there was brain hypoxia in approximately 24% of individuals who were at least 8 weeks post-COVID-19 infection, despite normal arterial saturation and no signs of fever. Moreover, we show hypoxia was associated with poor neuropsychological assessment, depression,   fatigue, and reduced health-related quality of life.

Materials and methods

Subjects

Healthy controls aged 18–65 years (n=17) who were nonsmokers (nicotine or marijuana), with no recent systemic infection, and no history of cardiovascular/vascular disease or neuropsychological disease were recruited. We recruited  34 participants who had contracted COVID-19 at least  8 weeks before the study visit from the general population.  Exclusion criteria included smokers (nicotine or marijuana),   history of cardiovascular/vascular disease, and other systemic inflammatory diseases such as inflammatory bowel syndrome, asthma, autoimmune diseases, celiac disease, glomerulonephritis, and hepatitis. Post-COVID-19 participants were screened for lingering symptoms. Participant demographics are summarized in Table 1. All participants provided written informed consent before the commencement of their participation. Informed consent was obtained from participants in Fig. 1 for the publication of identifying images in an online open-access publication.

Data collection was initiated no sooner than 20 min after participants entered the laboratory. This time was used to obtain consent. This calm 20-min period will help minimize physiological changes that may occur from previous activities. We also asked about physical activities over the previous 6 h.

NIRS measurement

fingers measurements were taken on the frontal cortex using a quantification system called the ISS (OxiplexTS  Frequency Domain Near-Infrared Spectrometer model  96,208, ISS Inc., Champaign, IL USA) (Fig. 1). The principle behind this commercially available equipment and its application is described in detail elsewhere [18–20].

Briefly, the fdNIRS probe consists of one fiber-optic detector and eight fiber-optic sources, with a source-to-detector separation of 2.0–3.5 cm. Source fibers emitted NIR light at 690 and 824 nm. Emitted light had an amplitude modulation frequency of 110 MHz, and the light was emitted by one source at a time according to a continuous cycle wherein the eight sources alternated between being switched on and of.  The estimation of the tissue absorption coefficients at multiple wavelengths enables the oxy- and deoxyhemoglobin (HbO, and HHb) concentration to be calculated using the Beer–Lambert law. The microvascular cortical oxygenation  (St O2) is calculated using the formula:

StO2 = [HbO]∕([HbO] + [HHb]).     (1)

Before data collection, the fdNIRS system was warmed up for at least 30 min and the system was calibrated using a phantom calibration block with known absorption and scattering coefficients. During data acquisition, participants were asked to sit quietly and upright in a chair. The probe was placed symmetrically on both the right and left side of the participant’s forehead and data was collected for about  1 min on each side and averaged (Fig. 1). Data were collected at a rate of 2 Hz, giving a total of 120 data points per subject. The fdNIRS quantifies the absolute values for HbO and HHb. This enables microvascular tissue oxyhemoglobin saturation (St O2) to be calculated, which serves as an indicator of the oxygenation status of the brain. The absolute level of absorption and scattering coefficients (µa and µs, respectively) at 690 and 824 nm were determined from the measured intensity (AC or DC) and phase shift by the ISS using the theory of photon migration [21]. Details of the mathematical equation and assumptions are discussed by  Hammer et al. [21].

Systemic oxygen saturation and heart rate were measured in the finger using a pulse oximetry device (Nonin Medical,  Inc. Minneapolis, MN USA Model 9500 Oximeter).

Tympanic temperature measurement was taken using a tympanic thermometer (Braun Thermoscan IRT 6520  ExacTemp).

Neuropsychological assessments

A Neuropsychological test battery was conducted on all participants. This included quality control using a test of memory malingering (TOMM), symbol digit modality test  (SDMT) oral to test visual information processing speed, control of word association test (COWAT) to test language and verbal fluency, and paced auditory serial addition test  (PASAT) to test attention, concentration, auditory information processing speed, and working memory.

For the COWAT, participants were asked to list as many words as they could in 1 min that began with the letters of the alphabet F, A, and then S, except for proper names or words with different endings. The score for each trial was the sum of correct responses, excluding repeats and rule breaks. The primary outcome measure for the COWAT was the sum of the correct responses for the FAS trial. This score was converted into a z-score using normative data from a healthy control population [22] to account for age and level of education-related effects. The calculation for z-score is demonstrated by the equation:

In a fourth trial for COWAT, participants were asked to list as many animals as they could that began with any letter of the alphabet. The total score was converted to a z-score using normative data with no correction for education, and the z-score was calculated by the equation:

The SDMT is a timed 90-s test, where participants used a reference key to match numbers (1–9) with nine randomized geometric shapes [23]. The total number of correctly recorded matched pairs was tallied to give an overall score, which was converted into a z-score.

For the PASAT, a recorded series of 61 numbers (1–9)   was played aloud at the rate of one number every 3 s. The participants were asked to add each spoken number to the number that was presented previously. Before the testing trial, participants completed up to three practice trials that consisted of only 11 numbers. Participants only proceeded to the test once they demonstrated a sufficient understanding of the task. The score for the PASAT was the sum of correct responses, with a maximum score of 60 [24]. This was converted into a z-score using normative data from a healthy control population (retrieved from the PASAT manual) to account for the level of education-related effects.

Health‑related quality of life assessment, fatigue, and depression measured in post‑COVID‑19 participants only

Questionnaires included the health-related quality of life (HRQoL) assessment using a 36-item instrument for adults, the RAND 36-Item Short-Form Health Survey  (SF-36) [25], Functional Assessment of Chronic Illness Therapy-Fatigue Scale (FACIT-F) to assess fatigue [26] and Beck Depression Inventory second edition (BDI-II).

In COVID-19 participants, HRQoL was measured using the 36-Item Short Form Survey (SF-36). Participants were asked to score their quality of life compared to what it was before contracting the COVID-19 infection and at the time of the visit. Anxiety and depression were measured using the Beck Depression Inventory (BDI-II) [27].

The SF-36 measures eight health concepts (physical functioning, role limitations due to physical health problems, role limitations due to personal or emotional problems, energy/fatigue, emotional wellbeing, social functioning, bodily pain, and general health perceptions) using multi-questions, 35 in total. It also includes a single question that indicates a perceived change in health. Participants’ response to each question is recorded so that each is scored from 0 to 100%, with higher scores indicating a more favorable health state. This questionnaire is a generic HRQoL tool that is useful for comparing general and specific populations and the relative burden of a health condition, in this case, COVID-19 infection [28].

The FACIT-F (version 4) is a 13-item self-report questionnaire that measures the severity and impact of an individual’s level of fatigue during their usual daily activities over the past week. The level of fatigue is measured on a   four-point Likert scale (4=not at all fatigued to 0=very   much fatigued) [29]. The subscale scores are calculated by first reversing negatively stated items (subtracting the response from‘4’) and then summing the raw (0–4) scores. A total score is then derived by summing subscale scores. Participants’ fatigue subscale score ranges from 0 to 52, where a lower score indicates more severe fatigue and a cut point suggesting clinically relevant fatigue set at < 34 [30, 31]. Although there is no gold standard for the measurement of fatigue, FACIT-F has been applied in conditions like cancer, HIV, lupus, rheumatoid arthritis, psoriatic arthritis, anemia, COPD, Parkinson's disease, and post-stroke [32–38], and is valid and reliable [36, 39, 40]. We do not claim to validate the use of the FACIT-F (version 4) to “diagnose” fatigue in individuals with post-COVID-19 conditions; however, we use a score of<34 as a crude indication of clinically relevant fatigue.

The BDI-II is a 21-item self-report questionnaire that assesses the extent of common depressive symptoms occurring throughout the past 2 weeks. This questionnaire uses a scale, from 0 to 3, and responses from all items are summed to give a total score from 0 to 63, with a higher score indicating greater levels of depression [27].

Results

We recruited 17 healthy controls and 34 individuals who have had COVID-19 and were at least 8 weeks post-diagnosis of SARS-CoV-2 infection (Table 1). We had subjects sit for 20 min to help standardize for exercise. In addition, we recorded exercise over the last 24 h. Four post-COVID-19   participants in the normoxic group and two in the hypoxic group reported a period of exercise before the study visit.  None of these six participants’ values was outside two standard deviations from their respective means. Thus, we suggest previous activity did not impact our results.

Of the 34 individuals who have had COVID-19, 19 self-reported as having persistent symptoms, defined as having at least two symptoms that suggest long COVID. The most commonly reported symptoms were fatigue and brain fog.  There were no differences between healthy controls and post-COVID-19 participants for age, SaO2% (% arterial blood oxygen saturation), heart rate (HR) (BPM), and tympanic temperature (°C) (Table 1).

Comparison between controls and all post‑COVID‑19 participants

We compared all people post-COVID-19 with the controls (Fig. 2 and Table 1).

There are different ways of defining hypoxia. A starting point is to test whether the post-COVID-19 population has lower St O2 than controls. Using Welch’s t-test, we show that the post-COVID-19 population is significantly different from healthy controls (lower, p=0.037). The mean values between groups are 63.1± 3.4% and 60.1± 6.9%  (mean±SD) for the controls and COVID-19 groups, respectively. The respective coefficients of variation are 5.4% and  11.5%. We note that the coefficient of variation is higher in the post-COVID-19 group. If we take a conservative view that hypoxia is defined as 2xSD below the control mean,   the overall mean±SD of controls is 63.1±3.4%, then anything below 56.3% would be hypoxic. There were eight post-COVID-19 participants (34 or 24%) who were hypoxic,   while none of the controls could be classified as hypoxic.

Although nearly missing the criterion for statistical significance (p=0.053), total hemoglobin in post-COVID-19 participants was lower compared with healthy controls  (43.4±10.2 μM vs. 49.8±11.1 μM, mean±S.D). The scattering coefficient (µs) at 690 and 824 nm and the absorption coefficient (µa) at 824 nm were significantly lower in post-COVID-19 participants compared with healthy controls,   whereas µa at 690 nm was not significantly different between groups.

Neuropsychological assessments all show significant impairment in the post-COVID-19 participants compared with healthy controls: symbol digit modality test (SDMT) oral (−0.62±1.20 vs. 0.98±1.08 p<0.001); control of word association (COWAT) for FAS (− 0.61 ± 0.90 vs.  − 0.02 ± 0.62 p = 0.010); and animals (− 0.26 ± 1.02 vs.  0.29±0.51 p=0.016) and paced auditory serial addition test  (PASAT) (−0.42±0.84 vs. 0.33±0.68 p=0.002).

Comparison between controls and post‑COVID‑19 participants grouped as normoxic or hypoxic

We divided the post-COVID-19 participants into hypoxic or normoxic groups (Table 2). A one-way ANOVA, when data was normally distributed, or the Kruskal–Wallis test was carried out between healthy control, and normoxic and hypoxic groups. Hypoxic participants were measured on average 7 months (range 3–15) after infection, and normoxic participants 8 months (range 2–19) after infection.  There were significant differences between normoxic and hypoxic post-COVID-19 groups for fdNIRS parameters  St O2 (p<0.001), oxyhemoglobin (p=0.007), and µs at 690 and 824 nm (p=0.020 and p=0.031, respectively). In post hoc analysis, there was no significant difference in St O2 between normoxic post-COVID-19 participants and healthy controls; however, as expected, hypoxic post-COVID-19 participants had lower St O2 compared with healthy controls and normoxic post-COVID-19 participants. HBO was significantly lower in hypoxic post-COVID-19 participants compared with healthy controls, and there were no significant differences between normoxic post-COVID-19 participants vs. healthy controls and normoxic vs. hypoxic post-COVID-19 participants (Table 2).

It may be that we should use age as a covariate. Although nearly missing the criterion for statistical significance  (p = 0.054), hypoxic post-COVID-19 participants had higher age, compared with the normoxic group and healthy controls (Table 2). When we did a univariate analysis with St O2 as the dependent value, group as the fixed factor, and age as a cofactor, the group was close to being significantly different (p=0.052), while age was significant (p<0.001).

When age is defined by groups of 10 years (e.g., 20–29,  30–40, 40–50, and 50–63 years), there are eight independent groups (4 controls and 4 post-COVID-19 groups).  A univariate general linear model (GLM) with Bonferroni post hoc tests indicated that St O2 for the control age group 20–30 years (group 1) was higher than those of postCOVID-19 age groups 40–50 and 50–63 years, and that of the post-COVID-19 group 20–30 years was higher than those of post-COVID-19 age groups 40–50 and 50–63 years. If we group by age, we can identify hypoxic individuals by calculating how many are 2xSD below the control mean. The St O2 threshold for hypoxia in the age groups is as follows: 20–30 (57.7%), 30–40 (55.1%), 40–50 (56.0%), and  50–63 years (57.1%). In these four age groups, the number of COVID-19 subjects that were hypoxic was 1, 1, 4, and 3, respectively, or 9 in total (26%). There were no controls that would be classified as being hypoxic.

In summary, if we do a simple comparison of means, the controls, and post-COVID-19 groups' St O2 values were different and very close to being different with a univariate analysis of variance with age as a cofactor. A clearer picture emerges if we look at how many individuals in the different groups are defined as hypoxic by being greater than 2xSD below the control mean. When we adjust for age or not, the number is 26% or 24%, respectively. These data indicate that approximately ¼ of people post-COVID-19 have significant hypoxia in the brain. Given these results, we will use the conservative cut-off for hypoxia (56.3%) for all further analyses.

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