Relationships between Inflammatory Biomarkers and Fatigue among Patients with Moderate and Severe COVID-19

Background Patients with moderate or severe COVID-19 infection suffer from varying levels of fatigue; however, there is a lack of understanding regarding the effect of inflammation on fatigue; and whether these relationships differ according to the severity of the infection. Aim To assess the relationships between selected inflammatory biomarkers and fatigue levels among hospitalized Jordanian patients with moderate or severe COVID-19 infection. Methods A quantitative cross-sectional design was used. A total of 352 participants were recruited for the study. Data regarding fatigue type and level were collected using the Chalder fatigue scale. Laboratory test results regarding several selected inflammatory biomarkers (e.g., ESR, CRP, IL-6, D-dimer, and others) were collected from patient records. The severity of the COVID-19 infection was determined using the criteria of the Ministry of Health in Jordan based on the results of O2% (oxygen saturation). Results The mean scores of the total fatigue level significantly differed between the two levels of the severity of COVID-19 infection (moderate and severe levels) (t = −3.0, p < 0.05). Similar findings were observed with physiological fatigue (t = −3.50, p < 0.05), and no significant difference was observed in psychological fatigue. Out of the selected inflammatory markers, only neutrophil and lymphocyte count had a significant influence on total fatigue level. Conclusion The level and type of fatigue was affected by the severity of the disease. However, the disease process itself represented by the levels of the inflammatory markers showed little influence on fatigue. The implications such as continuous screening of fatigue, and monitoring of the levels of the inflammatory markers are important to assist in diagnosing and managing COVID-19 patients. Furthermore, the relationship between the inflammatory process and fatigue is complex and requires further exploration.


Introduction
COVID-19 is a viral infection that afects the respiratory system, with symptoms that include tiredness and fatigue. Te World Health Organization (WHO, 2023) in its report that was updated on 23 December, 2022, indicated that the total number of cases had reached 651,918,402 cumulative cases, with a cumulative death count that reached 6,656,601 deaths. Each day, a total of 778,897 cases are discovered worldwide. In the United States, the Center for Disease Control and Prevention [1] reported in its updated report on 30 December, 2022, that the total cumulative number of cases had reached 100,662,056, and the total number of cumulative deaths reached 1, 088,481. Te current reported case fatality rate (CFR) is about 1%. Each day in the U.S. approximately 5668 cases are hospitalized due to COVID-19 infection. Te Jordanian Ministry of Health [2] reported as of August 2022, that the total number of cases had reached 1,731,549 and the total number of deaths reached 14105. A total of 159 cases were admitted on the week of the report (13-19/August, 2022), a total of 4832 new cases were reported, and a total CFR of 0.09 per 100,000 of the overall population was observed.
Dhochak et al. [3] describe the pathophysiological process of COVID-19 infection. Tey stated that the SARS-CoV-2 virus attacks the respiratory epithelial cells by attaching to the angiotensin-converting enzyme-2 (ACE-2) protein using its S-protein. Tis infection causes the activation of an infammatory response that includes various mediators which also activates the hemostatic system due to the endothelial dysfunction. In other words, thromboinfammation and cytokine storm play a signifcant role in the disease process and severity.
Te Jordanian Ministry of Health [4] classifed COVID-19 in terms of its severity of illness into four categories: mild, moderate, severe, and critical. Tese clinical forms which were defned in this study according to the criteria of the Jordanian Ministry of Health are as follows.

Mild Clinical Form.
It is a confrmed case in which the patient sufers from symptoms and signs of upper respiratory tract infections and does not complain of symptoms of shortness of breath, and the percentage of hemoglobin saturated with oxygen is more than 94%.

Moderate Clinical Form.
Te patient sufers from symptoms and signs of lower respiratory tract infections (bronchitis or pneumonia), including shortness of breath, and his hemoglobin saturation with oxygen is more than 94%.

Severe Clinical Form.
Te patient who sufers from pulmonary infections with shortness of breath and whose hemoglobin saturation with oxygen is less than 94% (See Figures 1-3 for 3 for computed tomography (CT) scans for the lungs of patients with severe COVID-19 infection).

Critical Clinical Form.
COVID-19 is a condition where a person sufers from severe respiratory depression and requires efective artifcial respiration, and fatigue is one of the main and commonly reported symptoms of COVID-19, which is considered a subjective sensation of psychological or physical exhaustion that reduces the capacity to perform a psychological or physical task because of the depleted resources [5]. Fatigue is a complex phenomenon that is usually associated with COVID-19 infection and cannot be completely explained by a single mechanism. However, fatigue can be attributed to the underlying infammatory condition, which is a common thread of COVID-19 infection [6].
Several infammatory markers have been examined in terms of their relationship with the process of COVID-19 infection. Tese markers included ESR, IL-6, D-dimer, PCT, and CRP [7]. Certain infammatory markers such as ESR, PCT, CRP, and other markers were found to positively correlate with the severity of COVID-19; and CRP was described as a sensitive systemic marker of acute-phase response during infammation and tissue damage, which could be used as an indicator of COVID-19 progress [8].
Tere is a lack of studies that addressed fatigue during COVID-19 infection in relation to the levels of the infammatory markers. So, this study aims to describe the relationship between selected infammatory biomarkers and fatigue among hospitalized patients with moderate or severe COVID-19 infection.
Te following are the research questions: (1) What is the relationship between COVID-19 severity and fatigue among hospitalized patients with COVID-19 infection? (2) What is the relationship between the selected infammatory biomarkers and fatigue among hospitalized patients with COVID-19 infection?

Literature Review
Te current evidence suggested that the infammatory responses play an important role in the progression of COVID-19. In a literature review study conducted by Rostami and Mansouritorghabeh [9], they found that the higher the levels of D-dimer, the worse the prognosis in patients with COVID-19 infection. An increase in the Ddimer levels by three or four times during the early stages of the disease was associated with an elevated level of death. Te study concluded that the D-dimer test is a reliable predictor of identifying thrombosis in COVID-19 patients and disease prognosis [9].   2 International Journal of Infammation Zeng et al. [8] conducted a literature review study using several databases with an aim to investigate the association between several infammatory markers and the severity of COVID-19 disease. Sixteen studies were included in the analysis with around 4000 participants. Te fndings indicated that the patients with severe symptoms had signifcantly higher levels of procalcitonin (PCT), interleukin-6 (IL-6), ESR, and CRP than that of the nonsevere cases of COVID-19. However, these studies were heterogeneous and were conducted in a single country. In addition, these studies were reported to be underpowered indicating limited generalizability of these fndings in exploring the mechanism of the efect of these biomarkers with the severity of COVID-19.
Even though fatigue was reported as a common symptom in COVID-19 infection [6], Poenaru et al. [10] reported a conficting evidence. Poenaru et al. conducted a review study to investigate the etiology of fatigue in patients with COVID-19 infection. Tey stated that the infammation per se might not be the source of fatigue. Te review found several noteworthy similarities between COVID-19 symptoms and chronic fatigue syndrome in terms of symptom patterns, so, they concluded that there was no sufcient evidence to identify COVID-19 as a universal trigger for such symptoms.

Design.
A cross-sectional design was used to assess the COVID-19 patients'fatigue and to determine the relationship with infammatory biomarkers.

Sample.
A convenience sampling strategy was used to recruit participants from the targeted hospitals. Te inclusion criteria were the following: (1) patient who have confrmed diagnosis of moderate or severe COVID-19, (2) equal to or more than 18 years of age, and (3) can speak and read Arabic language. Te exclusion criteria were those diagnosed as critical cases or intubated patients. Te Jordanian Ministry of Health protocol [4] stated that the confrmed case of COVID-19 infection is defned as the case that is laboratory-confrmed by a PCR examination through a positive result to detect the SARS-CoV-2 virus.
Te protocol stated that cases are diagnosed in Jordan only by adopting the polymerase chain reaction (PCR) test. Test samples are taken by using a nasopharyngeal swab, sputum sample, or a pulmonary lysing (BLM) sample. Two or more samples can be taken depending on the availability of the samples and according to the opinion of the attending physician, as required by the patient's condition.
COVID-19 infection in Jordan is diagnosed based on the results of the real-time PCR (RT-PCR) test (COVID-19 MDx RT-PCR COVID-19 detection kits). Tose tests are FDA approved by the Jordanian Food and Drug Administration. Te RT-PCR is an in vitro diagnostic technique that is used to detect SARS-CoV-2 by using a nasopharyngeal, oropharyngeal, mid-turbinate, or an anterior turbinate swap. Only trained staf are allowed to perform this test in predefned test locations. Te reported clinical performance of these RT-PCR kits is 100% positive percent agreement, and 100% negative percent agreement by the manufacturer. Unfortunately, no reports were found regarding the test's sensitivity and specifcity in Jordan. Te estimated sensitivity of the test was reported around 80% and the specifcity was about 99% [11].
Te sample size was estimated using G * Power 3.1 for multiple linear regression. Te following parameters were used to estimate the sample size: alpha � 0.05, beta � 0.80, and efect size � 0.07 (representing small to medium efect size). Te resulting estimation of the sample size was 267 participants. An estimated dropout percentage of 15% was considered. Tus, the minimum fnal estimated sample size was 307 participants (267 + 40), and the fnal recruited sample was 351 participants.

Instruments
3.3.1. Fatigue. Te Chalder fatigue scale was used to measure fatigue. Tis scale is composed of 11 items that talk about sensations and functionality. Each one of these items is measured using a 4-point Likert-type scale. Te responses range from asymptomatic (no symptoms) to the maximum/ highest symptomology. Responses on the far left receive a score of 0, and as the case becomes more symptomatic, the responses increase to 1, 2, or 3. Te global fatigue score for each case was calculated by the summation of the scores, so that the global score can range from a minimum of 0 to a maximum of 33. In addition, the Chalder fatigue scale was used to determine the fatigue type, where items from 1 to 7 are used to measure the physical fatigue, and items from 8 to 11 measures the psychological fatigue. Tis scale can be used to score the values bimodally, i.e., if fatigue exists or not. In this case, items are scored as 0 or 1. A score of 4 or more using this bimodal measurement indicates the existence of fatigue (fatigue caseness). Te questionnaire was translated into Arabic. Te translation process was based on the recommendations by Sousa and Rojjanasrirat [12].

Severity of COVID-19
Infection. Te severity of COVID-19 infection was measured as an ordinal variable to determine whether the patient is considered to have a moderate or severe COVID-19 infection. Te protocol was proposed by the diagnostic and treatment protocol for patients with the emerging coronavirus (COVID-19) issued by the Ministry of Health in Jordan and approved by the Jordanian National Committee for Epidemic Control; and it was used to describe if the case is considered to be moderate or severe COVID-19 infection as previously mentioned.

Infammatory Biomarkers.
Te results of the infammatory biomarkers for the corresponding participants were gathered from the hospital records. Te aforementioned protocol by the MOH described the required tests for any and all admitted/hospitalized patients with COVID-19 infection; and this protocol indicated that the following tests International Journal of Infammation should be performed for the hospitalized COVID-19 patients on a daily basis.
Te following are the infammatory markers tested according to the management protocol:

Settings.
Tree of the hospitals that were allocated by the Jordanian government as hospitals to receive patients with COVID-19 infection were selected for data collection. Only certain hospitals were selected by the government to admit COVID-19 patients. Other hospitals were instructed to transfer any COVID-19 patients to those previously identifed as a precaution to control disease transmission.

Data Collection
Procedure. IRB approval was received on April 20, 2022. Data collection was performed between May 1st, 2022, and June 4th, 2022. After acquiring the IRB and the approval from the selected hospitals, the investigator introduced the topic to the head nurse and staf nurses to provide clarifcation regarding the study and the questionnaire. Te investigator distributed the questionnaire to the potential participants who matched the inclusion criteria. Tose who accepted to participate signed an informed consent and completed the questionnaire. Results of the infammatory markers were collected from the corresponding participant's records by the investigator.

Data Analysis
Procedures. Data analysis was performed using the SPSS program. Te variables that were tested in this study were reported using the descriptive statistics such as average, SD, frequency, range, and percentage. Pearson R, multiple regression analysis, chi-square test, and t-tests were used to answer the research questions. TheP value for this study was determined at the level of 0.05.

Ethical Consideration.
Te study method was approved by the ethical research committee and by the IRB of the Jordan University of Science and Technology (IRB number 656-2021; Jordan University of Science and Technology). A formal informed consent was obtained from the participants who decided to participate in the study. Te data collection sheets and the questionnaires were coded so that the confdentiality of the participants is protected. Te participation was completely and thoroughly voluntary, and the participants were assured that their answer sheets and their data will remain confdential. Te participants were informed that they have the right to withdraw from the study at anytime during the conduction of the study, and that their withdrawal will be without any penalty. Te participants were informed about the estimated time needed to complete the participation and to answer the questionnaire; and they received the contact information during the data collection. No harm or risk was imposed on the participants.  Table 1 for demographic characteristics).

COVID-19 Severity and Fatigue.
Te relationship between the severity of COVID-19 infection and fatigue was tested using Pearson R and showed that the severity of the infection had a signifcant positive correlation with the total fatigue score and physiological fatigue (r � 0.16, p < 0.05; r � 0.18, p < 0.05), respectively, but had no signifcant correlation with the psychological fatigue. Further testing was performed using an independent sample T-test to assess the diferences in fatigue level between the moderate and the severe COVID-19 infection groups. Both the total fatigue level and physiologic fatigue level were entered as test variables. Te results supported the initial correlation tests and showed that there was a significant diference between the mean scores of fatigue in the two severity groups (moderate and severe) (t � −3.0, p < 0.05); and showed the presence of signifcant diference in the mean scores of physiological fatigue among these groups (t � −3.5, p < 0.05). In these two tests of diference, the mean score of fatigue was higher amongst the severe COVID-19 infection group than that of the moderate infection group. In addition, there were no signifcant differences between the mean scores of psychological fatigue and the mentioned severity groups. Additional testing was performed to further explore this relationship. A chi-square test was conducted between fatigue as bimodally (binary) and the severity of COVID-19 infection. Te chi-square test showed that the severity of COVID-19 infection was independent from whether the participant had fatigue or not (chi-square � 1.24; p > 0.05).

Infammatory Markers and Fatigue.
To test the relationships between the selected infammatory markers and level of fatigue, a multiple linear regression analysis was done. In this analysis, age, body temperature, and O 2 saturation were entered with the previously identifed infammatory markers. Te regression analysis showed that the overall model was signifcant (F � 6.3, p < 0.01) and the overall model accounted for about 20% of the variability of the fatigue scores (R square � 0.20; adjusted R square � 0.174).
However, only age, neutrophil count, and lymphocyte count were found to have a signifcant infuence on the fatigue level (See Table 2). Age had a positive relationship with the total fatigue level (B � 0.151; p < 0.05) indicating that a difference of 10 years of age would increase the level of fatigue by approximately 1.5 points. Meanwhile, neutrophil and lymphocyte count had a negative (inverse) relationship with fatigue (B � 0.045; B � 0.082; p < 0.05 respectively), indicating that an increase in 100 units of neutrophil count would be associated with a decrease in the fatigue level by about 4.5 points, whereas 100 increments in lymphocyte would be associated with a decreased fatigue level by 5 points. Additional regression analyses were performed to assess if the relationships between the infammatory markers and the type of fatigue (physical and psychological) were any diferent. Te results remained the same and showed that only age, neutrophils, and lymphocytes had a signifcant infuence on the level of physiological and psychological fatigue.
Further analysis was performed to examine the relationships between the infammatory markers and fatigue. Binary logistic regression tests were performed to assess the ability of the level of infammation (infammatory markers) and to predict the fatigued and nonfatigued cases. In these tests, fatigue caseness were entered as binary independent factors and checked against the previously mentioned dependent factors. First, a multivariate binary logistic regression analysis was performed. Te results of this test showed that the overall model was signifcant and was a better ft than the model with no predictors (model chisquare � 50.4, p < 0.05; Hosmer and Lemeshow test chisquare � 12.5, p > 0.05). In this model, only age and lymphocytes were signifcant predictors of fatigue caseness. Te model proposes that an increase in age and a decrease in the lymphocyte count are associated with an increased likelihood of fatigue (See Table 3). Te proposed model showed that the number of cases that were correctly predicted based on the observed value was 71.3%. However, the model had a high percentage of correctly predicting cases with fatigue (194 cases were correctly predicted to have fatigue versus 28 cases predicted to have fatigue but observed to not have fatigue), whereas 43.8% were correctly predicted by the model (57 correctly predicted vs. 73 not correctly predicted by the model to not have fatigue), indicating better model sensitivity compared to its specifcity.
In the second analysis, a univariate binary logistic regression was conducted to assess if the results of the frst test remained the same, or whether independent variables/factors' efects on fatigue difered (See Table 4 for the univariate logistic regression). Results showed that the fndings remained about the same where only age and lymphocyte count signifcantly predicted fatigue caseness.

Discussion
Tis study aimed at examining the relationships between various infammatory markers and fatigue in patients with COVID-19. Also, this study examined whether the severity of COVID-19 infection infuences fatigue, and if the severity plays a role in the relationships between the level of the tested infammatory markers and fatigue.
Studying fatigue during and post COVID-19 infection has been recognized as an important aspect of the research study due to the prevalence of this infection, the prevalence of fatigue in COVID-19 infection, and the long-term efect of fatigue on the quality of life, thus, fatigue ought to be screened and continued to be monitored in patients with COVID-19 infection [13] especially for that substantial proportion of people who got COVID-19 infection and continue to sufer from the ongoing fatigue or post viral fatigue [14].
Tis study showed that fatigue is a prominent symptom and reported a high prevalence of fatigue among patients with moderate and severe COVID-19 infection. Not many studies were found to address the prevalence of fatigue in hospitalized patients with moderate and severe COVID-19 infection. A study was found and reported results regarding the prevalence of fatigue that were congruent with our study. International Journal of Infammation   Shendy [15] reported about 64% prevalence of fatigue among their study sample (nonhospitalized mild and moderate cases of COVID-19 infection). Whereas the prevalence of fatigue in this study sample was about 63%. Rudrof et al. [16] addressed fatigue in patients with COVID-19 infection and indicated that fatigue is one of the main symptoms of COVID-19 that negatively afect the patients and causes deterioration in their quality of life through decreasing patient's physical and psychological performance; and these deteriorations are attributed to the pathological changes caused by the disease process.
Our study found diferent results from that of Rudrof et al. [16] in that no direct relationship was found between most of the infammatory markers and fatigue. Tis fnding was substantiated in this study when this relationship remained the same even when assessed between infammation and any of the two types of fatigue: the physical and psychological fatigue. Tese fndings were congruent with commentary by Azzolino and Cesari [6] who stated that fatigue cannot be explained by a unique pathogenic mechanism. Such fndings indicate that the relationship between infammation and fatigue in patients with COVID-19 infection is complex and might be indirected or mediated by other factors. Te results of the logistic regression further support this notion especially since lymphocyte count was the only infammatory marker that contributed to the prediction of the presence of fatigue. Moreover, further investigation is required to describe the pathological processes that explain the rationale behind the inverse efect that lymphocyte count has on fatigue. Similar fndings regarding the inverse relationship between lymphocyte and fatigue were reported by Illg et al. [17]; however, contradictory to our study this inverse relationship was mediated by the severity of COVID-19 infection.
Many studies in the literature also addressed the relationship between fatigue and the severity of COVID-19 infection such as Islam et al. [18] who reported in their review study that the severity of the infection as represented by the severity of the cytokine storm can infuence the development of health problems such as fatigue. In addition, Poenaru et al. [10] in their review study reported that with regard to fatigue, multiple explanations regarding what infuences fatigue exist, some suggested that COVID-19 patients experience dysregulations in their immune and neurological systems, and dysregulations in their metabolic pathways. However, these fndings are not consistent with the other studies. Te fndings of our study were congruent with the reports of Islam et al. and Poenaru et al. where a signifcant relationship was found between the severity of COVID-19 infection and the total fatigue level. However, results showed that physiological fatigue was infuenced by the severity of the infection and not psychological fatigue. Not only that, but also the severity of the infection played no role in the existence of fatigue, but the severity of the infection infuenced its levels. Tese fndings demonstrate the complexity of these relationships and the need to continue addressing fatigue among those with COVID-19 infection.

Conclusions
Fatigue in COVID-19 infection is prominent and is afected by many factors. Fatigue can be afected by the severity of the disease and the process of infammation. However, many other factors may have a substantial infuence on fatigue, and on the relationships between disease severity and process (infammatory markers) on fatigue. Te type of fatigue (physical, psychological) difered based on the severity of the disease. Te severity of COVID-19 infection, disease process, levels of infammatory markers, and fatigue are associated with each other in a complex relationship, and other factors may play a role in these efects. Tese fndings necessitate the need to further test these relationships.

Limitations
Te fndings of the current study should be considered within the context of certain limitations. Te participants were recruited using a nonprobability sampling technique (convenience sample). In addition, the authors collected the data from the participants at a single data point. Along the same line, fatigue was only measured during participants' hospitalization without performing a follow-up data collection to measure the level of fatigue after discharge. Tese limitations could limit the generalizability of the fndings. Terefore, the authors recommend conducting future longitudinal studies with data collection from randomly selected participants to overcome these limitations.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request. Data are in the form of a SPSS fle that the author has stored on his personal PC. Questionnaires were disposed of according to the institutional policy and regulation.

Additional Points
What does this study add? (i) Te evidence regarding the efect of the disease process in COVID-19 infection and the development of fatigue is still indistinct. (ii) Te level and the type of fatigue were afected by the severity of the disease. (iii) Many other factors may have a substantial infuence on fatigue, and on the relationships between disease severity and process (infammatory markers) on fatigue.

Disclosure
Tis research was conducted as part of the employment of the authors. All authors were afliated to Jordan University of Science and Technology at the time of research conducting.