The Impact of Iron Dyshomeostasis and Anaemia on Long-Term Pulmonary Recovery and Persisting Symptom Burden after COVID-19: A Prospective Observational Cohort Study

Coronavirus disease 2019 (COVID-19) is frequently associated with iron dyshomeostasis. The latter is related to acute disease severity and COVID-19 convalescence. We herein describe iron dyshomeostasis at COVID-19 follow-up and its association with long-term pulmonary and symptomatic recovery. The prospective, multicentre, observational cohort study “Development of Interstitial Lung Disease (ILD) in Patients With Severe SARS-CoV-2 Infection (CovILD)” encompasses serial extensive clinical, laboratory, functional and imaging evaluations at 60, 100, 180 and 360 days after COVID-19 onset. We included 108 individuals with mild-to-critical acute COVID-19, whereas 75% presented with severe acute disease. At 60 days post-COVID-19 follow-up, hyperferritinaemia (35% of patients), iron deficiency (24% of the cohort) and anaemia (9% of the patients) were frequently found. Anaemia of inflammation (AI) was the predominant feature at early post-acute follow-up, whereas the anaemia phenotype shifted towards iron deficiency anaemia (IDA) and combinations of IDA and AI until the 360 days follow-up. The prevalence of anaemia significantly decreased over time, but iron dyshomeostasis remained a frequent finding throughout the study. Neither iron dyshomeostasis nor anaemia were related to persisting structural lung impairment, but both were associated with impaired stress resilience at long-term COVID-19 follow-up. To conclude, iron dyshomeostasis and anaemia are frequent findings after COVID-19 and may contribute to its long-term symptomatic outcome.

We herein shed light on the long-term alterations of haematopoiesis and iron metabolism after COVID- 19 and their association with the recovery after acute . For this, we present long-term follow-up data prospectively assessed in the "Development of Interstitial Lung Disease (ILD) in Patients With Severe SARS-CoV-2 Infection (CovILD)" trial [20,44,45].

Patient Characteristics
The CovILD study included 145 COVID-19 patients, who have been comprehensively characterized [20,44,45]. The herein-presented analysis only includes patients who attended all follow-up visits at 60, 100, 180 and 360 days after COVID-19 onset (N = 108). This subcohort encompassed predominately middle-aged (mean age = 56 years (SD ± 14 yrs)) male patients (58%) with severe acute COVID-19 who had mainly been hospitalized (hospitalization rate: 75%, ICU care: 25% of the participants). The mean duration of hospital stay was 14 days (SD ± 16 days). Detailed characteristics of the cohort are depicted in Table 1.  1 The body mass index is the weight (kilograms) divided by the square of the height in meters; 2 due to pre-existing disease or ongoing immunosuppressive treatment; 3 all patients needing non-invasive or invasive ventilation were supplied with oxygen before ICU admission.

Iron Deficiency and Anaemia in Patients Recovering from COVID-19
At post-acute COVID-19 follow-up, the prevalence of iron deficiency without anaemia, combining absolute and functional iron deficiency, ranged from 35% to 16% of the cohort. Anaemia was found in 11.1% to 4.6% of patients, respectively (Figure 1a). Longitudinally, the prevalence of iron deficiency only slightly dropped (27.8% at 60 days vs. 25.0% at 360 days), whereas the rate of anaemia significantly declined (9.3% at 60 days vs. 4.6% at 360 days). Notably, we found a change in the iron deficiency and anaemia phenotype during the observation period. At early post-COVID-19 follow-up, functional iron deficiency was the predominant finding (N = 20, 18.5%), and only a few individuals suffered from absolute iron deficiency (N = 10, 9.3%), whereas at long-term follow-up, absolute iron deficiency became the prominent iron deficiency phenotype (absolute iron deficiency: N = 17, 15.7%, functional iron deficiency: N = 10, 9.3%). In line with the iron deficiency phenotype, AI was the most dominant form of anaemia, especially at early post-acute follow-up, whereas IDA, or combined IDA and AI, or unclassifiable/multifactorial anaemia were more prevalent at the 360 days follow-up ( Figure 1b). Of note, anaemia was predominantly mild, as reflected by the mean haemoglobin levels of anaemic patients of 113 g/L (SD ± 9.4 g/L) at the 60 days post COVID-19 follow-up, and haemoglobin levels gradually increased over the study period ( Table 2). Patterns of serum iron markers were consistent with functional iron deficiency at early post-acute COVID-19 follow-up with a translation to absolute iron deficiency at long-term follow-up, as both serum ferritin and hepcidin concentrations were significantly higher 60 days after acute COVID-19 as compared to the one-year re-evaluation ( Table 2). type during the observation period. At early post-COVID-19 follow-up, functional iron deficiency was the predominant finding (N = 20, 18.5%), and only a few individuals suffered from absolute iron deficiency (N = 10, 9.3%), whereas at long-term follow-up, absolute iron deficiency became the prominent iron deficiency phenotype (absolute iron deficiency: N = 17, 15.7%, functional iron deficiency: N = 10, 9.3%). In line with the iron deficiency phenotype, AI was the most dominant form of anaemia, especially at early postacute follow-up, whereas IDA, or combined IDA and AI, or unclassifiable/multifactorial anaemia were more prevalent at the 360 days follow-up ( Figure 1b). Of note, anaemia was predominantly mild, as reflected by the mean haemoglobin levels of anaemic patients of 113 g/L (SD ± 9.4 g/L) at the 60 days post COVID-19 follow-up, and haemoglobin levels gradually increased over the study period ( Table 2). Patterns of serum iron markers were consistent with functional iron deficiency at early post-acute COVID-19 follow-up with a translation to absolute iron deficiency at long-term follow-up, as both serum ferritin and hepcidin concentrations were significantly higher 60 days after acute COVID-19 as compared to the one-year re-evaluation (Table 2).

Persisting Inflammation and Hyperferritinaemia at Post-COVID-19 Follow-Up
At early post-acute COVID-19 follow-up, a substantial portion of study participants presented with low-grade thrombo-inflammation, as reflected by increased IL6, neopterin, pro-calcitonin, C-reactive protein (CRP) or d-dimer blood concentrations (Table 3). At long-term follow-up, inflammatory parameters significantly dropped and only a small subgroup demonstrated signs of persisting inflammation (Table 3). Accordingly, hyperferritinaemia, increased IL6 concentrations and elevations of d-dimer were frequently found at the 60 days post-COVID follow-up and significantly decreased over time ( Figure 2). Despite this longitudinal decline, a substantial portion of the patients presented with persisting hyperferritinaemia (N = 17, 15.7%) and d-dimer elevations at the 360-day follow-up (N = 23, 21.3%).

Persisting Inflammation and Hyperferritinaemia at Post-COVID-19 Follow-Up
At early post-acute COVID-19 follow-up, a substantial portion of study participants presented with low-grade thrombo-inflammation, as reflected by increased IL6, neopterin, pro-calcitonin, C-reactive protein (CRP) or d-dimer blood concentrations (Table  3). At long-term follow-up, inflammatory parameters significantly dropped and only a small subgroup demonstrated signs of persisting inflammation (Table 3). Accordingly, hyperferritinaemia, increased IL6 concentrations and elevations of d-dimer were frequently found at the 60 days post-COVID follow-up and significantly decreased over time ( Figure 2). Despite this longitudinal decline, a substantial portion of the patients presented with persisting hyperferritinaemia (N = 17, 15.7%) and d-dimer elevations at the 360-day follow-up (N = 23, 21.3%).

Gender-Related Differences in the Prevalence of Iron Deficiency and Anaemia at Post-COVID-19 Follow-Up
Male patients suffered from more severe acute COVID-19 as compared to female study participants ( Figure 3a). Accordingly, the frequencies of hyperferritinaemia, iron deficiency and anaemia were significantly higher in males as compared to females at early post-COVID-19 follow-up (Figure 3b-d). Until the last follow-up, these genderspecific differences extenuated, but males still tended to have a higher prevalence of iron dyshomeostasis and anaemia.

COVID-19 Follow-Up
Male patients suffered from more severe acute COVID-19 as compared to female study participants (Figure 3a). Accordingly, the frequencies of hyperferritinaemia, iron deficiency and anaemia were significantly higher in males as compared to females at early post-COVID-19 follow-up (Figure 3b-d). Until the last follow-up, these gender-specific differences extenuated, but males still tended to have a higher prevalence of iron dyshomeostasis and anaemia.

Association of Persisting Inflammation, Iron Dyshomeostasis and Anaemia
Disturbances of iron homeostasis, including hyperferritinaemia and the emergence of anaemia, during acute COVID-19 have been mainly attributed to inflammation. Thus, we evaluated associations between persisting inflammation and iron dyshomeostasis. Still, at least at the systemic level disturbances of iron homeostasis including ferritin elevations were only weakly to moderately associated with thrombo-inflammatory parame-

Association of Persisting Inflammation, Iron Dyshomeostasis and Anaemia
Disturbances of iron homeostasis, including hyperferritinaemia and the emergence of anaemia, during acute COVID-19 have been mainly attributed to inflammation. Thus, we evaluated associations between persisting inflammation and iron dyshomeostasis. Still, at least at the systemic level disturbances of iron homeostasis including ferritin elevations were only weakly to moderately associated with thrombo-inflammatory parameters such as IL6, CRP, procalcitonin, neopterin or d-dimer concentrations (Table 4). At follow-up, both iron dyshomeostasis and systemic inflammation were less frequent, and inflammatory parameters were only moderately associated with serum iron and TSAT. Interestingly, when comparing individuals with and without iron deficiency, iron-deficient patients demonstrated significantly higher inflammatory serum parameters and hepcidin concentrations over the study period. For instance, at the 360 days follow-up, patients with iron deficiency displayed higher CRP (p = 0.024), neopterin (p = 0.001) and hepcidin (p = 0.006) concentrations in the blood as compared to individuals without iron deficiency, whereas patients with absolute iron deficiency tended to have the highest concentrations of these inflammation markers.

Association of Iron Dyshomeostasis and Anaemia with Clinical COVID-19 Severity and Structural Lung Recovery
Iron dyshomeostasis and anaemia were significantly related to the severity of acute COVID-19 at the early post-acute follow-up, as patients with severe COVID-19 showed a disproportionally high prevalence of hyperferritinaemia, iron deficiency and anaemia. Still, at long-term follow-up, this association considerably weakened (Figure 4a). The severity of structural lung abnormalities was significantly higher in patients with iron dyshomeostasis and anaemia at early post-acute follow-up (Figure 4b). Although individuals with iron dyshomeostasis and anaemia demonstrated a higher variability of structural lung involvement at long-term post-COVID-19 follow-up, we did not find a significant difference in the severity of structural lung impairment between patients with or without iron dyshomeostasis or in association with anaemia at long term follow-up (Figure 4b). Conclusively, whereas hyperferritinaemia, iron deficiency and anaemia were related to acute COVID-19 severity at early post-acute follow-up; this relation was not apparent at long-term follow-up.   Iron dyshomeostasis and anaemia are not only relevant findings during acute COVID-19 but may also be of importance for post-COVID-19 recovery [20]. Thus, we investigated the association of iron dysbalance and anaemia with patients' resilience coping, fatigue, quality of life (QoL) and exercise capacity at long-term follow-up. One year after COVID-19 onset, a relevant sub-cohort of patients demonstrated chronic impairments as far as persistence of fatigue, impairment of the QoL and exercise capacity are concerned. In detail, the mean European Quality of Life visual analogue scale (EQ-VAS), which provides patients' QoL self-reports, was 85% (CI 79.7-85.0%), whereas 100% would describe a perfect QoL. The Chalder Fatigue Scale (CFS) assessment, which reflects the severity of tiredness in patients with fatigue, resulted in a median Likert score of 12 points (CI 12.4-15.0 points), whereas 37% exceeded a score of 14-indicating increased fatigue-and two patients demonstrated a CFS score equal or above 28, which is indicative for chronic fatigue syndrome. Accordingly, 8.8% (N = 9) of patients displayed a reduced six-minute walking distance. Notably, resilience was impaired in a substantial portion of the CovILD cohort, as only 45% had high resilience, whereas 26% and 29% demonstrated medium and low resilience, respectively. Interestingly, whereas fatigue, QoL and exercise capacity were not significantly related to iron dyshomeostasis or anaemia, a reduction in resilience was associated with persisting hyperferritinaemia, iron deficiency and anaemia at long-term follow-up ( Figure 5).

Discussion
A disequilibrium of iron homeostasis is a significant laboratory finding of CO 19. Most prominently, hyperferritinaemia and hypoferremia are frequently observed ing acute COVID-19 and have been described as prognostic markers for COVID-19 come [7,9,[25][26][27][28]30,31,46,47]. Accordingly, we previously reported an association betw the resolution of structural pulmonary impairment and the persistence of hyperferrit mia and inflammation at early post-acute COVID-19 follow-up [20,44]. Herein, we light on the long-term recovery after COVID-19 focusing on iron dyshomeostasis anaemia post-COVID-19. Of interest, whereas anaemia was a frequent finding at post-acute follow-up, we discovered a significant change in the anaemia phenotype frequency at follow-up. Most notably, anaemia at the 60 days follow-up mainly dem strated features of AI, suggesting that COVID-19-related inflammation followed by cidin-induced iron sequestration and functional iron deficiency is the major trigge anaemia during acute and early post-acute COVID-19. This theory is supported b herein-presented observation of a high frequency of hyperferritinaemia and incre hepcidin concentrations at the 60 days post-COVID-19 follow-up as well as previo

Discussion
A disequilibrium of iron homeostasis is a significant laboratory finding of COVID-19. Most prominently, hyperferritinaemia and hypoferremia are frequently observed during acute COVID-19 and have been described as prognostic markers for COVID-19 outcome [7,9,[25][26][27][28]30,31,46,47]. Accordingly, we previously reported an association between the resolution of structural pulmonary impairment and the persistence of hyperferritinaemia and inflammation at early post-acute COVID-19 follow-up [20,44]. Herein, we shed light on the long-term recovery after COVID-19 focusing on iron dyshomeostasis and anaemia post-COVID-19. Of interest, whereas anaemia was a frequent finding at early post-acute follow-up, we discovered a significant change in the anaemia phenotype and frequency at follow-up. Most notably, anaemia at the 60 days follow-up mainly demonstrated features of AI, suggesting that COVID-19-related inflammation followed by hepcidin-induced iron sequestration and functional iron deficiency is the major trigger for anaemia during acute and early post-acute COVID-19. This theory is supported by the herein-presented observation of a high frequency of hyperferritinaemia and increased hepcidin concentrations at the 60 days post-COVID-19 follow-up as well as previously published data describing a high prevalence of AI in COVID-19 patients [10,18,48]. At long-term follow-up, the prevalence of anaemia significantly decreased. This observation is explained by the resolution of inflammation during COVID-19 convalescence, as reflected by decreasing inflammatory markers such as IL6, procalcitonin and CRP, as well as a restoration of iron homeostasis, mirrored by a drop of hepcidin and ferritin levels. Still, a substantial portion of patients demonstrated long-term impairment of iron handling resulting in the emergence of IDA and combined forms of anaemia (IDA + AI). Although this observation urges further mechanistic evaluation, a prolonged disturbance of iron absorption following COVID-19 might be a possible explanation for the development of absolute iron deficiency as well as IDA at long-term follow-up.
The observation of differential phenotypes of anaemia at post-COVID-19 follow-up warrants a thorough characterization of the iron status, as treatment of anaemia depends on the precise assessment of the anaemia phenotype [49]. For instance, IDA without significant signs of persisting inflammation may be successfully treated with iron supplementation, whereas iron substitution is less effective for patients with AI and systemic inflammation at COVID-19 follow-up [17].
Interestingly, the prevalence of ID and anaemia was higher in males as compared to females recovering from COVID-19. This is in contrast to the prevalence of iron deficiency and anaemia in the general population, where both conditions are far more frequent in females, suggesting that iron dyshomeostasis and anaemia are related to COVID-19 and disease severity [50,51]. Accordingly, iron deficiency and anaemia at follow-up were related to acute COVID-19 severity, and males tended to have higher markers of systemic inflammation during follow-up (data not shown).
Both iron deficiency and anaemia are well-known contributors to patients' morbidity [17,52,53]. For instance, exercise capacity is impaired in individuals with iron deficiency, even without the development of anaemia, which may be partly linked to reduced mitochondrial functionality [54][55][56][57]. However, inflammation and iron loading can trigger radical formation, thereby resulting in mitochondrial dysfunction as well [58,59]. Nonetheless, iron deficiency impacts mental health, fatigue and quality of life [60]. Fatigue and reduced exercise capacity are frequently observed in patients recovering from COVID-19 and thus are often described in individuals diagnosed with post-acute sequelae of COVID (PASC) or long-COVID [47,61]. Notably, these diagnostic terms still suffer from blurry definitions and uncertain clinical value. Objective measures often fail to describe somatic causes of the symptom burden of these individuals and the overall understanding of the multifactorial causes of PASC is still rudimentary [47]. Given the high frequency of iron dyshomeostasis during acute COVID-19 and at post-acute COVID-19 follow-up, we hypothesized that alterations in iron metabolism and haematopoiesis might contribute to the persisting symptom burden post-COVID-19. As fatigue and impaired exercise capacity are among the most frequent persisting symptoms at COVID-19 follow-up, we focused on these outcomes in the CovILD cohort and assessed the impact of iron dyshomeostasis and anaemia on these outcomes [62]. PASC are reported at a high prevalence and are related to acute COVID-19 severity [47,61]. The CovILD trial mainly included severe COVID-19 patients; thus, we expected a high rate of PASC connected with a significant impairment of QoL, exercise capacity and a substantial rate of chronic fatigue [45]. Still, at the 360 days post-COVID-19 follow-up, the observed impairment of QoL and exercise capacity was relatively mild, and high CFS scores were only found in a few individuals. Although fatigue and exercise impairment are considered to be major symptoms of PASC, our hereinpresented evaluation does not reflect a severe impairment at long-term post-COVID-19 follow-up. This finding is striking and may be explained by two considerations. First, the interpretation of the clinical significance of PASC is still related to various ambiguities, as the definition of PASC includes a variety of different symptoms and typically does not describe the clinical significance of the assessed symptoms. Secondly, the prevalence of PASC may decrease over time, and PASC may be reversible in most patients.
Iron deficiency was frequently found in the study participants and persisted in 25% of the patients until the 360 days post-COVID-19 follow-up. Additionally, 5% of the study participants suffered from anaemia at the long-term follow-up. Interestingly, neither iron deficiency nor anaemia significantly impaired exercise capacity. The latter may be explained by the fact that hardly any patient developed severe anaemia or iron deficiency, and the predominantly mild anaemia and iron deficiency may not significantly impact exercise capacity in a low-intensity exercise test, such as the herein applied six-minute walking test. Accordingly, QoL and fatigue were not significantly influenced by mild anaemia. Still, when we evaluated patients' stress coping capacities, iron deficiency and even more pronounced anaemia were associated with reduced resilience. This finding is of high interest, as it points toward an impact of iron deficiency and anaemia on convalescence in post-COVID-19 patients who frequently report neuropsychiatric symptoms, including sleeping disorders and impairment of neurocognition [63,64]. Accordingly, iron deficiency and anaemia are risk factors for developing a psychiatric disorder, and psychiatric disorders are generally associated with lower levels of resilience compared to mentally healthy controls [65,66]. Mechanistically, it has been suggested that iron deficiency induces poor myelination or disruptions in neurotransmitter levels or mitochondrial activity [67,68].
Finally, we must acknowledge that the herein-presented study has some limitations. First, the observational study design reports associations rather than causality, and mechanistic studies are needed to provide further insight into involved signalling pathways and a potential cause-effect relationship. Second, the prevalence of iron deficiency and anaemia in the general population is high, and according to the study design, we cannot assess the prevalence of anaemia or iron deficiency in the CovILD cohort before COVID-19 onset. Third, treatment effects such as iron supplementation were not investigated in this study. Fourth, the role of chronic low-grade inflammation and its interaction with iron metabolism and haematopoiesis cannot be fully evaluated with the presented study design, as low-grade inflammation may not be assessed by the analysis of inflammatory biomarkers at the systemic level. Thus, at this point, we cannot rule out or confirm if persisting iron dyshomeostasis is functionally related to chronic low-grade inflammation at long-term follow-up.

Patients and Study Design
The CovILD study is a prospective multi-centre observational cohort trial which included 145 COVID-19 patients (ClinicalTrials.gov number, NCT04416100). The study longitudinally assessed post-COVID-19 recovery 60, 100, 180 and 360 days after disease onset. Medical history, acute disease severity and symptom burden were assessed, and each visit included a clinical examination, a standardized questionnaire evaluating COVID-19-related symptoms, performance testing (e.g., six-minute walking test (SMWT)), structural lung evaluation using CT without contrast agent, lung function testing and the acquisition of blood for laboratory analyses. Inclusion criteria were an age of 18 years or older, COVID-19 diagnosis established by typical clinical symptoms for COVID-19, a positive reverse transcription-polymerase chain reaction (RT-PCR) SARS-CoV-2 result obtained from a nasopharyngeal or oropharyngeal swab and the ability to perform repetitive follow-up visits at the primary study centre at the Medical University of Innsbruck. Participants were recruited at three participating study centres, namely the St. Vinzenz hospital in Zams, the post-COVID rehabilitation Center of Münster and the Medical University of Innsbruck. Patients with mild (out-patient treatment), moderate (in-hospital treatment without oxygen supply or ventilation), and severe (in-hospital treatment with the need for oxygen supplementation or intensive care unit (ICU) treatment) acute COVID-19 were included. Informed written consent was obtained from all participants, and the study was approved by the local ethics committee at the Medical University of Innsbruck (EK Nr: 1103/2020).

Laboratory Assessment
Blood samples were collected at each follow-up and obtained via routine peripheral vein puncture. Laboratory evaluation was performed with standardized ISO-certified procedures as previously described [44,45]. Blood gas analysis was performed via punctuation of the hyperperfused earlobe following Finalgon application.

Structural Lung Evaluation with CT
A low-dose (100 kVp tube potential) CT scan of the chest without the use of an iodine contrast agent was performed at each follow-up. We used a 128-slice multidetector CT hardware with a 38.4 × 0.6 mm collimation and spiral pitch factor of 1.1 (SOMATOM Definition Flash, Siemens Healthineers, Erlangen, Germany).
Structural lung evaluation was performed by three radiologists experienced in lung CT diagnostics. The analysis included pattern description (ground-glass opacities (GGO), consolidations, bronchiectasis and reticulations as defined by the glossary of terms of the Fleischner Society) and severity grading [76]. The severity grading was independently analysed by three radiologists. The latter used the following standardized scoring system: Each lung lobe was graded according to the presence of pulmonary abnormalities: 0-none; 1-minimal (subtle GGO, very few findings); 2-low (several GGO, subtle reticulation); 3-moderate (multiple GGO, reticulation, small consolidation); 4-marked (extensive GGO, consolidation, reticulation with distortion) and 5-massive (massive findings, parenchymal destructions), and a total lung score was obtained via the addition of the scores for each lobe, resulting in a score ranging from 0 to 25 points (i.e., a maximum score of 25 if each lobe was scored 5 points).

Assessment of Exercise Capacity, Resilience, Fatigue and Quality of Life
Exercise capacity was evaluated with the SMWT. Additionally, we used standardized questionnaires to assess resilient coping (BRCS), fatigue (CFS) and quality of life (European Quality of Life 5 Dimensions 5 Level Version and the EQ-VAS) [77][78][79].

Statistical Analysis
Statistical analyses were performed with IBM SPSS Statistics version 27.0.1.0 (IBM, Chicago, IL, USA). Descriptive data analysis included tests for homoscedasticity and data distribution (Levene test, Kolmogorov-Smirnov test, Shapiro-Wilk test and density blot/histogram analysis). Group comparisons of continuous data were assessed with the Mann-Whitney-U test and Kruskal-Wallis test for continuous data and Fisher's exact test or Chi-Square test for binary and categorical data. Multiple testing was adjusted by the Sidak formula, as appropriate. Correlation analysis was performed with the Spearman rank test.

Conclusions
Alterations of iron homeostasis and anaemia are frequent findings in acute COVID-19 but also at post-acute COVID-19 follow-up. During COVID-19 convalescence, systemic thrombo-inflammation, hyperferritinaemia and the prevalence of anaemia gradually decline; still, a significant subgroup of the post-COVID population displays persisting iron deficiency or even anaemia, which may contribute to persisting symptom burden in these individuals. A link between iron dyshomeostasis and SARS-CoV-2 driven inflammation is evident and may at least partly contribute to this finding, but the interconnection of COVID-19 pathobiology and iron dyshomeostasis is likely far more complex and urges further evaluation.  . BI had no role in the design, analysis or interpretation of the results. BI was given the opportunity to review the manuscript for medical and scientific accuracy as it relates to BI substances, as well as intellectual property considerations.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Ethics Committee of the Medical University Innsbruck (EK Nr: 1103/2020). The CovILD study is registered at ClinicalTrials.gov (NCT04416100).
Informed Consent Statement: Written informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available in the article.