Robin Macnofsky’s first symptoms, in April 2020, were so mild she didn’t take a test for COVID-19, preferring to save it for someone who really needed it.
A few weeks later her chest tightened and her temperature spiked. These ebbed, but an even stronger wave hit: a high fever and exhaustion that left her bedbound in a days-long “zombie sleep.” Macnofsky tested positive for COVID-19 (likely from remaining viral genetic material). Her symptoms endured.
Previously a champion multi-tasker who could split her concentration in four directions, the then-59-year-old could no longer follow the plot of a TV episode or walk her dog. Her temperature waxed and waned.
A realization dawned: The end was nowhere in sight, and no one could tell her why.
Scientists studying the unfamiliar and unfurling COVID-19 pandemic also began to realize that for Macnofsky and many other COVID-19 patients, a long hospital stay or a short, mild illness were not the only outcomes. For some people, mild symptoms, quickly resolved, were just the beginning.
Researchers worldwide, including many at Fred Hutchinson Cancer Research Center and the University of Washington, are working to understand long COVID-19, the long-lasting effects of COVID-19 infection that can affect adults, teens and children. Hutch and UW investigators are building on deep expertise in immunology and infectious diseases like HIV to figure out what causes long COVID-19, who is at risk, and how to treat it. To do so, they’re tackling challenges that range from basic questions about how best to measure symptoms to uncovering the complex immunological interplay that may drive symptoms.
‘Something is different’
In March 2020, Julie Czartoski, a nurse practitioner working with Fred Hutch virology expert and Joel D. Meyers Endowed Chair holder Dr. Julie McElrath, helped McElrath quickly put together a study, the Seattle COVID Cohort Study, to look at COVID-19 in first responders and those infected with SARS-CoV-2, the virus that causes COVID-19. They wanted to know who was getting it, and how badly. Soon they opened the study to others in the community. Luckily, few participants who contracted the coronavirus had symptoms that warranted hospitalization. Most had relatively mild infections that cleared up quickly.
But that wasn’t the end. For many, symptoms persisted, even worsened. Participants reported lingering fevers, new joint pain, exhaustion and brain fog, among a grab bag of other symptoms. When an unexpected number of young, healthy firefighters reported atrial fibrillation — a rapid, abnormal heart rhythm — Czartoski knew something was up.
“It’s not unheard of in young adults, but to have so many was weird,” she said. “I remember texting Julie McElrath and saying, ‘Something is different, because these people are still sick.’”
Because of their fast action and prompt study enrollment, Czartoski and McElrath were early to recognize a truth that would eventually become clear to other scientists: COVID-19 can cast a long shadow.
Many of the people in McElrath’s study who reported lasting problems had experienced mild, almost cold-like infections with SARS-CoV-2. Very few had been hospitalized.
It takes time to know for sure that a patient is suffering from long COVID-19. A week or two is not long enough to be sure their symptoms won’t clear up soon. Eventually, McElrath and her team found that about 30% of the coronavirus-positive participants in their study had lasting symptoms, or new symptoms attributed to COVID-19, that extended at least 60 days past their initial infection, Czartoski said.
Long COVID-19 dogs some patients, like Macnofsky, much longer, and their symptoms can be severe. After two different week-long hospital stays, in which she underwent terrifying procedures to remove nearly a liter of fluid pressing on her heart and then her lungs, Macnofsky took a leave of absence from her career as a community organizer. By early 2022, she still rations her energy and has yet to regain her prior multi-tasking abilities.
Time defines the syndrome and time challenges those suffering from it and the scientists working to untangle it. Also known as PASC, for post-acute sequelae of SARS-CoV-2 infection, long COVID-19 refers to symptoms that endure long after someone has recovered from their initial coronavirus infection. (The term “sequelae” refers to a disease’s aftereffects.) Symptoms can include fatigue, shortness of breath, brain fog, fever, anxiety and depression. COVID-19 patients who were sick enough to need ICU care may also be dealing with post-ICU syndrome, and it’s not yet clear how their experience will unfold differently from patients who were treated in the ICU for other causes.
For a health problem as mysterious and as complex as long COVID-19, progress requires scientific teamwork.
That’s why, in summer 2021, Dr. Rachel Bender Ignacio reached out to investigators across Fred Hutch and the University of Washington, inviting them to join a long COVID-19 working group to share challenges and solutions, and find collaborators to help investigate specific questions.
As medical director of the Hutch’s COVID-19 Clinical Research Center, or CCRC, Bender Ignacio had a good sense of who at both institutions were treating or studying the syndrome. She was also hearing from CCRC trial participants who had transitioned from acute to long COVID-19 and wanted to know how scientists were addressing it. “I have my ear to the ground,” she said.
Bender Ignacio felt that progress required stronger connections between clinicians and laboratory and translational scientists. Physicians needed a better understanding of the biological mechanisms driving long COVID-19 before they could move proposed treatments into clinical trials, and basic scientists could reveal those mechanisms but needed tissue samples and clinical insights from the people providing patient care.
“Bringing everyone together was the least I could do,” Bender Ignacio said.
The working group she put together is an example of international and multidisciplinary efforts to tackle the challenges that vex investigators studying long COVID-19, including how to best classify and diagnose the syndrome, what’s causing it, and how to treat it. The recently launched National Institutes of Health RECOVER initiative, aimed at understanding PASC, is giving the investigators in the field hope that standards may be forthcoming, said Dr. Eric Chow, a working group member and UW infectious diseases fellow who studies the damage that respiratory viruses can do outside the lungs.
The researchers who joined Bender Ignacio’s collective span disciplines and body systems, including the brain, heart and lungs. They include researchers and clinicians at UW and Harborview Medical Center, which opened one of the nation’s earliest clinics devoted to helping long COVID-19 patients. The working group members bring expertise in long-term complications from viral infections like HIV and influenza, and know how viruses or the immune reaction to them can damage the body. SARS-CoV-2 may be wreaking the most havoc right now, but it’s not the only virus that can upend sufferers’ lives: Many, including Bender Ignacio, have spent their careers studying the long-term effects of HIV.
Now the team is bringing their wide-ranging expertise to bear on the many questions of long-haul COVID-19, hoping to surmount its challenges and help patients.
Learning on the fly
Studying a little-understood problem in a rapidly shifting pandemic is incredibly challenging. At the beginning, scientists had no knowledge base to inform their data collection or study design. Every week or month brought new information that forced them to reassess the data they had already collected — and adjust their data-collection methods to incorporate new understanding.
“It’s like building a boat while you’re sailing it,” said working group and CCRC member Dr. James Andrews, a University of Washington rheumatologist who studies how sepsis, particularly severe cases requiring hospitalization, can lead to long-term disability.
The first challenge was realizing that a problem existed. Many of the long-haulers Czartoski interviewed for the Seattle COVID Cohort Study struggled to find help and even recognition of their symptoms, she said.
“In the beginning, a big part of my job was just listening,” Czartoski recalled.
Study participants wept as they described to her their crushing fatigue and debilitating symptoms, and the struggle to get health care providers to understand that their vague-seeming complaints posed a real problem. Macnofsky, too, found it difficult to get help for her constant fever, headaches, fatigue and brain fog.
In the beginning, no one knew what information would be important to understand why these patients were suffering and how to help.
A snapshot of data “is one piece of the whole puzzle,” said Hutch statistician Dr. Zoe Moodie, who helps design and analyze HIV vaccine trials and develops statistical methods to analyze immunological data. “Generally, the more pieces the better, and as time goes go on we learn which are the important pieces.”
Czartoski tackled the problem by collecting everything she could: “Sometimes [a symptom] didn’t seem important, then a week later I’d have five people reporting it.”
And sometimes the information that most impacts a patient’s life can seem negligible when committed to paper, she said. One person’s loss of smell or taste may seem like small potatoes compared to others’ chronic exhaustion and continual fevers. But such seemingly small symptoms can make life and some careers difficult: Firefighters smell phantom burning and parents who can no longer smell a dirty diaper. Once-favorite food now repells.
“And [those symptoms] are really tough on quality of life,” Czartoski noted. Depression can set in, straining a person’s long-term relationships, affecting quality of life and for some, the ability to hold down a job.
Initially, she had a limited systematized questionnaire, and took notes longhand while patients noted every symptom they could think of — whether they knew it related to their COVID-19 or not. As time passed, Czartoski and her colleagues were able to spot common symptoms that they added to an ever-expanding checklist. (Then, they had to get the checklist built into the study records system.)
The working group members brainstorm statistical and analytical strategies that could help, and which take into account the fact that not everyone’s data has been collected in the same way at the same time points during the course of their disease.
Even now, their efforts to untangle long COVID-19 are hampered by what they didn’t know six, 12, 18 months ago, Czartoski said.
“Researchers will ask about blood drawn a year and a half ago: Were they taking Tylenol?” she said. “It could change the immune response, but I don’t know!”
Fred Hutch HIV researchers Dr. Julie McElrath (left) and Dr. Rachel Bender Ignacio (right) are parlaying their expertise in viral infections and clinical trials to help patients suffering from long COVID-19. McElrath gathered a cohort of long-haulers who are helping researchers dig into the immune drivers of PASC. As the CCRC’s medical director, Bender Ignacio helps make crucial connections between basic and translational scientists working in the field.
Photos by Robert Hood / Hutch News Service
Finding the right box
Another major challenge that long COVID-19 researchers face is classification. Many studies are producing data on the syndrome, but if symptoms aren’t collected and classified similarly, trying to compare different studies will be like comparing apples and oranges.
Decisions about how to classify symptoms also affect how patients are grouped together and how the data is analyzed. One big concern: Should patients be grouped by symptom, or should symptoms be grouped by the organs they affect?
It’s a bit of a chicken-and-egg issue, but it gets into the problem of what’s behind long COVID-19 to begin with, said Chow, who began treating Macnofsky after her hospital stay. (The two teamed up to tell Macnofsky’s story in a dual first-person essay published in the scientific journal Open Forum Infectious Diseases.)
Even in cases where it’s clear the immune system is at fault, fatigue may not have a single cause. Energy-sucking immune activation could explain one person’s fatigue, but post-infection autoimmunity, in which their own tissues are under attack, could be the reason behind another’s, Andrews said.
Trying to find the biological similarities in data taken from these patients would be like trying to compare pages of text written in different languages: more likely to result in gibberish than to identify a helpful pattern.
And sometimes, symptoms may not even be the result of a person’s coronavirus infection. Part of the problem is the often-vague, widely varying collection of symptoms, many of which long COVID-19 shares with other chronic health problems, such as autoimmune diseases or chronic fatigue syndrome. Autoimmune diseases often strike in young adulthood. For some people, SARS-CoV-2 infection and an autoimmune diagnosis are just two pieces of unrelated bad luck.
“In a longitudinal cohort like this, nothing is ever completely clean,” Czartoski said.
Working group members share questions and strategies. Should they classify symptoms by severity score, or follow the CDC’s recommendations to classify symptoms by outcome measures in different areas? Members often draw on their or other members’ expertise in different disciplines, such as adapting questionnaires used by neurologists to assess cognitive difficulties. Czartoski recommended a severity scale long used by HIV researchers to assess how symptoms impact patients’ daily living.
The team also grapples with the challenges of classifying symptoms that may seem focused on a specific organ system, but are actually emblematic of a body-wide problem. Researchers noted that some simplification needs to occur to make it possible to analyze the reams of data that can be collected.
But sometimes it’s unclear what’s causing someone’s symptom — so researchers can’t classify symptoms by underlying cause. What then?
Members also keep an eye on trends in the wider scientific community to see if they can align with areas of growing consensus, the better to compare their results with other studies.
Who’s at risk for symptoms, and how long will they last?
But sorting out the logistical challenges of classification is just the first step. Long COVID-19 researchers want to understand why symptoms develop and who’s at risk. Why do some symptoms affect some patients but not others? Who will have a mild course, and who will suffer greatly? A deeper understanding, they hope, will shed light on why symptoms linger so long for some people, and how to predict how a patient’s experience will unfold.
UW neurologist Dr. Payal Patel is focusing on the cognitive symptoms of PASC.
“I want to know, what is the cause of the symptoms we see in PASC,” said Patel, who studies the continuing effects of infections of the central nervous system, including HIV. “We know PASC affects different organ systems. I’m trying to get a better clinical understanding of how it affects the brain.”
Without this, it’s difficult to give worried patients a clear picture of what they can expect from long COVID. Patel wants to better understand how long such symptoms last, who’s most at risk, and what’s causing them. Is brain fog caused primarily by immune dysfunction? Or could the clotting problems associated with COVID-19 have damaged the cells lining blood vessels in the brain? Patel and a team of scientists have studies underway to answer these questions.
This type of location-specific question can be very difficult to address, Chow noted. It’s relatively easy to take blood samples and look at general patterns of immune cells or antibodies floating through the blood. But what about problems that are occurring at a hard-to-reach spot, like tiny blood vessels in the brain or lungs?
Some working group members focus their research questions on specific areas of the body. With Patel, Andrews is trying to understand who’s most at risk for cognitive and physical impairments after COVID. Some patients’ fatigue is related to muscle wasting, known medically as sarcopenia, and Andrews wants to know what’s behind that and who’s at risk.
Role of the immune system in long COVID-19
Since it became understood that an overactive immune response (known as a “cytokine storm”) lurks behind some of COVID-19’s most dire complications, scientists have begun digging deeper into how the immune system responds to SARS-CoV-2. Macnofsky herself participated in a Benaroya Research Institute study looking at the immune response to the novel coronavirus.
Bender Ignacio’s working group is drawing on the Hutch’s longstanding expertise in immunology and infectious diseases and looking to the immune system for answers.
“We’re studying what natural infection looks like over time,” said Fred Hutch immunologist Dr. Maria Lemos, who studies immune responses in mucosal tissue like vaginal and nasal surfaces, where we first encounter many viruses. “People could have cold symptoms for nine days, then four moths later they’re diagnosed with a lung condition or a heart condition.”
To understand why, she and others on McElrath’s team are mapping the immune response to SARS-CoV-2 infection as it unfolds over months. With collaborators at Emory University in Atlanta, they’re charting the rise and fall of antibodies against the virus and how different immune-cell populations grow, shrink and morph over time.
By describing how these responses differ between people who did and did not develop long COVID-19, the researchers hope to identify key biomarkers, like specific inflammatory proteins, that help predict which patients will have persistent problems. Such biological predictors could help doctors intervene early, either to help connect patients with the right services to help them deal with symptoms, or (once scientists crack this problem) stave it off entirely.
McElrath’s team, with collaborators at Emory University and the Seattle-based Allen Institute for Immunology, has revealed some tantalizing immunological patterns, Lemos said, which the group posted on the preprint server biorxiv.org. The immune trajectory in many long-haulers looks startlingly unlike that seen in people who recover quickly and permanently.
“The alarm system of the immune system doesn’t get turned on as quickly in these people — but surprisingly it seems to remain on for way longer,” Lemos said. They’re currently putting together a paper for peer review at a scientific journal.
On top of this project, Moodie is working with investigators the Allen Institute to identify immunological signatures, including cellular features like proteins and gene expression, that distinguish long COVID-19 from acute COVID-19. Some patients — including Macnofsky — report improvement of symptoms after vaccination for COVID-19, and Moodie and her collaborators want to understand how vaccination may help their bodies resolve their chronic, damaging immune responses.
(Whether prior COVID-19 vaccination helps protect against developing long COVID-19 is still being explored. A recent study in The Lancet suggests that vaccinated people are less likely to have long-lasting symptoms after SARS-CoV-2 infection.)
Treatments for long-haulers?
Macnofsky said she’s recently been helped by six months of weekly pulmonary rehab sessions. It’s possible that early access to rehabilitative therapies could help prevent or alleviate severe long COVID-19 in others.
It is taking time for scientists at the Hutch and elsewhere to get a clear enough picture of what’s driving long COVID-19 to open clinical studies to address patients’ problems, but a few have begun.
The Seattle arm of a multi-center, Phase 2 trial of a drug called RSLV 132, is administered through the CCRC and headed by Andrews. (People interested in participating can contact the CCRC.) Developed by biotech company Resolve Therapeutics, the drug has already shown promise against fatigue in people with autoimmune diseases like Sjögren syndrome, and scientists hope long COVID-19 patients will also benefit.
“One of the exciting things about this study is that it’s taking a targeted therapy approach to treating symptoms,” Andrews said. “It’s targeting the mechanism behind fatigue — chronic inflammation — to see if it helps.”
In addition to testing whether RSLV 132 outperforms a placebo when it comes to alleviating fatigue in long COVID-19 patients, researchers will collect tissue samples that they’ll study to get a better picture of how it might be working (if it does), he said. If it turns out that some patients respond and some don’t, such samples could also help investigators figure out why.
And, if there’s a similar biological process underpinning the fatigue seen in long COVID and other diseases and syndromes, the study could benefit a wide array of patients, Andrews said.
That’s a hope that other members of the working group also harbor. One of the big questions, said Chow, is whether long COVID-19 patients are suffering from the same immunologic problems as patients with chronic fatigue syndrome, autoimmune diseases, or other virus-associated chronic damage. Or is there something unique about the biology behind long COVID-19?
Bender Ignacio sees potential for the close ties between the CCRC and the long COVID-19 working group to help fast-track promising treatments or treatment strategies that emerge from group members’ projects.
Working toward a more certain future
One thing Chow hopes come from the studies is more predictability. Right now, clinicians struggle to determine whose symptoms will last, and whose will resolve. A better understanding of long COVID-19 subgroups will help clinicians guide patients toward the best therapies for them. Improving doctors’ ability to diagnose and clinically characterize long COVID-19 will also help improve insurance reimbursement, he said. Right now, he stressed the need to recognize and validate what patients with long COVID-19 face.
Macnofsky, who at one point couldn’t take a phone call from a friend without falling back into a zombie sleep, feels fortunate. She joined Facebook groups organized by long COVID-19 sufferers, where many reported not just horrible symptoms but job loss and crushing debt. Macnofsky’s leave of absence from her career was fully supported — emotionally and financially — by her husband. She’s recovered enough now to step back into some job duties, though not at her previous level. But she knows other patients who continue to suffer, with no idea when — or if — their symptoms will ever improve.
On top of everything else, her uncertain future is one of the biggest challenges Macnofsky faces Luckily, she said, her “keep calm and carry on” attitude (and compassionate family) are helping her ride her waves of symptoms.
“There’s a mental health component to being so ill,” Czartoski said. She still encourages patients to treat themselves gently and take it one day at a time. While most will improve with time, a few will worsen — and she still can’t tell someone which patient they’ll be. The answer will only come with more data.
“I’m still collecting everything I can,” Czartoski said.
Given the continued spread of coronavirus 2, the early predictors of coronavirus disease 19 (COVID-19) associated mortality might improve patients’ outcomes. Increased levels of circulating neurofilament light chain (NfL), a biomarker of neuronal injury, have been observed in severe COVID-19 patients. We investigated whether NfL provides non-redundant clinical value to previously identified predictors of COVID-19 mortality.
We measured serum or plasma NfL concentrations in a blinded fashion in 3 cohorts totaling 338 COVID-19 patients.
In cohort 1, we found significantly elevated NfL levels only in critically ill COVID-19 patients. Longitudinal cohort 2 data showed that NfL is elevated late in the course of the disease, following the two other prognostic markers of COVID-19: decrease in absolute lymphocyte count (ALC) and increase in lactate dehydrogenase (LDH). Significant correlations between ALC and LDH abnormalities and subsequent rise of NfL implicate that the multi-organ failure is the most likely cause of neuronal injury in severe COVID-19 patients. The addition of NfL to age and gender in cohort 1 significantly improved the accuracy of mortality prediction and these improvements were validated in cohorts 2 and 3.
A substantial increase in serum/plasma NfL reproducibly enhanced COVID-19 mortality prediction. Combined with other prognostic markers, such as ALC and LDH that are routinely measured in ICU patients, NfL measurements might be useful to identify the patients at a high risk of COVID-19-associated mortality, who might still benefit from escalated care.
Since early 2020, the coronavirus disease 19 (COVID-19) pandemic has exhausted medical systems worldwide. Even after the development of safe and effective vaccines, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to spread.1 A reliable early predictor of COVID-19 associated mortality would help prioritize use of medical resources and maximize patient survival.
Neurofilaments are essential cytoskeleton proteins of the central and peripheral axons exclusive to the nervous system. Compared to neurofilament heavy and medium chains (NfH and NfM, 200 and 150 kDa), neurofilament light chain (NfL, 68 kDa) has a lower molecular weight and easily diffuses from parenchyma to cerebrospinal fluid (CSF) and blood.2–4 Recent developments of ultrasensitive assays, such as single molecule array (SIMOA), allow reproducible measurement of low NfL concentrations in serum or plasma.5, 6 Consequently, blood NfL became a key noninvasive biomarker of acute neuronal injury in diverse neuropathological conditions,7 including sepsis-associated encephalopathy.8
Although previous studies have demonstrated an association between COVID-19 morbidity and central nervous system (CNS) damage,9–13 several questions still remain unanswered: (1) Does a single measurement of NfL provide meaningful prognostic information at individual patient level?; (2) Is there a relationship between NfL and previously described COVID-19-associated mortality biomarkers14 of prognostic value, such as absolute lymphocyte count (ALC), C-reactive protein (CRP), and lactate dehydrogenase (LDH)?; and (3) Does NfL improve COVID-19 mortality prediction by demographic markers such as age and gender?
Materials and Methods
Research subjects and cohorts
Serum or plasma samples from COVID-19 patients admitted at ASST Spedali Civili (Brescia, Italy) were obtained through the Laboratory of Clinical Immunology and Microbiology (LCIM), National Institute of Allergy and Infectious Diseases (NIAID), under Institutional Review Board (IRB)-approved protocols (Comitato Etico Provinciale: NP 4000 – Studio CORONAlab and NP 4408 – Studio CORONAlab and ClinicalTrials.gov: NCT04582903). Blood samples from all patients were taken between 6:00 and 11:00 AM; samples were collected in S-Monovette® serum and S-Monovette® lithium heparin (Catalog # 04.1934.001 and 04.1936; Sarstedt, Numbrecht, Germany) tubes for isolation of serum and plasma, respectively. Tubes were centrifuged at 2000g for 10 min at 20°C, samples were collected, aliquoted, and stored at −80°C; frozen samples were shipped on dry ice.
SARS-CoV-2 infection was confirmed using the nasopharyngeal swab – polymerase chain reaction test. COVID-19 disease severity was determined as per Diagnosis and Treatment Protocol for Novel Coronavirus Pneumonia guidelines, released by the National Health Commission & State Administration of Traditional Chinese Medicine.15
Serum and plasma samples from healthy controls (HC) and multiple sclerosis (MS) subjects were collected at Neuroimmunological Diseases Section (NDS), NIAID after informed consent under IRB approved protocol (ClinicalTrials.gov: NCT00794352). The NfL levels measured in HC and MS subgroups were previously reported16 and are used in the current study only as a positive control of neuronal injury; the measurements of other COVID-19 prognostic biomarkers in these control samples were not reported previously.
Three hundred seventy-eight serum or plasma samples were collected from 338 COVID-19 patients grouped into three independent cohorts (Fig. 1, Table 1, and Data S1). In cohort 1, 30 cross-sectional serum samples were collected from COVID-19 patients with three levels of disease severity. In cohort 2, 60 longitudinal plasma samples were collected from 20 critically ill COVID-19 patients (T1, T2, and T3: collected averagely at 5- to 10-day intervals, within 30 days of hospitalization). Cohort 3 consisted of 288 cross-sectional plasma samples collected from critically ill COVID-19 patients where a large proportion of the subjects eventually died (39.2%).
Table 1. Demographic details of the three cohorts.
Positive Control (MS)
Cohort 1 (N = 168) (serum)
Cohort 2 (N = 38) (plasma)
Cohort 3 (N = 288) (plasma)
Age (unpaired t-test), gender, and comorbidities (Chi-square test) were compared across disease diagnosis/severity subgroups. COVID-19, coronavirus disease 19; HC, healthy controls; MS, multiple sclerosis.
*p < 0.05 versus HC and #p < 0.05 versus COVID-19, critical – survived.
NfL single molecular array (Simoa™) assay
Frozen serum and plasma samples were thawed on ice and were used immediately; repeated freezing and thawing of the samples was avoided. NfL concentrations in samples were measured using the Simoa™ assay (Catalog # 103186; Quanterix, Billerica, MA, USA). Samples were diluted 1:4 and randomly distributed on 96-well plates. Quality control (QC) samples provided with the kit had concentrations within the pre-defined range and the coefficient of variance (CV) across the plates was <10%: (1) cohort 1: 1 plate, no CV; (2) cohort 2: 3 plates, control 1 = 8.9% and control 2 = 3.7%; (3) cohort 3: 5 plates, control 1 = 8.6% and control 2 = 7.6%. All samples were analyzed blindly under alpha-numeric codes. The diagnostic codes were broken only after QC-verified NfL concentrations were reported to the database manager.
To determine the effect of sample age on NfL degradation and assay performance, within each cohort and disease diagnosis/severity subgroups, we analyzed the correlations between sample age (date of sample analysis – date of sample collection, in days; Data S1) and NfL concentrations (pg/mL) using linear regression models. We did not observe statistically significant (p < 0.05) correlations (data not shown).
Adjustment for the effect of healthy aging
As serum/plasma NfL levels increase with physiological aging,17 the measured NfL concentrations were adjusted for the effect of healthy aging as described previously.16 Following age versus serum- or plasma-NfL equations from HC cohorts were used: ln(serum NfL) = 0.0177 × Age + 0.9696 and ln(plasma NfL) = 0.0158 × Age + 1.247. The age-adjusted NfL concentrations represent residuals from the above-stated linear regression models.
NfL levels were compared across disease diagnosis and severity subgroups using either Kruskal–Wallis ANOVA or Welch’s t-test. Correlations between NfL and systemic markers of COVID-19 morbidity were evaluated using Spearman analysis and linear regression model.
Prediction models of COVID-19-associated mortality were developed in R Studio Version 1.1.463 (R version 4.0.2) using logistic regression (glm function of the “stat” package).18 Optimal cutoff for the predictive models was calculated using the optimalCutoff function of the “InformationValue” package (https://cran.r-project.org/web/packages/InformationValue/index.html). The area under the receiver operating characteristic curve (AUROC) was calculated using the roc function of the “pROC” package.19 AUROC measures model’s ability to discriminate between positive cases versus negative cases, AUROC >0.7 is considered a good performing model.
NfL levels increase with COVID-19 severity and mortality
Although increased blood NfL levels have been reported in patients with severe COVID-19,9–13 previous studies had insufficient numbers of subjects who died from the disease to assess whether NfL can predict COVID-19 mortality.
To fill this knowledge gap, we measured NfL levels in 30 COVID-19 patients with three levels of severity: (1) moderate severity (n = 10); (2) critical condition but survived (n = 10); and (3) critical condition but died (n = 10). Positive and negative control subgroups consisted of (1) patients with acute COVID-19-like symptoms admitted in critical health conditions who tested negative for SARS-CoV-2 infection (n = 10); (2) HC (n = 58); (3) MS patients with acute focal CNS inflammation measured as contrast-enhancing lesions on brain MRI (active MS, n = 35); and (4) MS patients without evidence of acute focal CNS inflammation (non-active, n = 35).
After diagnostic codes were unblinded, we found elevated levels of NfL in COVID-19 patients compared to HC (Fig. 2A). NfL levels in COVID-19 patients increased with disease severity, but only cohorts of critically ill COVID-19 and MS patients reached statistical significance compared to HC.
Next, we compared cohort differences in other blood biomarkers of COVID-19 morbidity: ALC, CRP, and LDH (Fig. 2B–D). Like NfL, decreased ALC and increased LDH correlated with COVID-19 severity; statistically significant differences in ALC and LDH were observed only in critically ill COVID-19 patients compared to HC. Interestingly, although non-COVID-19 acute respiratory illness control had levels of COVID-19 prognostic biomarkers (i.e., NfL, ALC, and LDH) comparable to HC, they had the highest levels of the prototypical acute phase reactant, CRP.
We conclude that NfL, LDH, and ALC abnormalities increase with COVID-19 severity, are associated with COVID-19 mortality, and can differentiate COVID-19 from other acute respiratory conditions that lead to ICU admission.
In COVID-19 patients NfL rises later during hospitalization, trailing transient abnormalities in ALC and LDH by 5–20 days
The earlier a biomarker can identify patients at risk for COVID-19 mortality, the greater its clinical value. Because none of the previous studies addressed the dynamics of NfL rise in COVID-19 and compared it to the dynamics of other prognostic biomarkers, we addressed this knowledge gap in the longitudinal cohort 2.
We measured NfL in 60 samples collected from 20 critically ill COVID-19 patients within 30 days of hospitalization, at three timepoints (T1, T2, and T3) taken at approximately 5- to 10-day intervals. We observed statistically significant, progressive increases (T1 vs. T2 and T3) in NfL levels only in patients who later died (Fig. 3A and Appendix S1).
When plotting measurements against the number of days before death, we observed a progressive increase in NfL approaching death, while no such increases occurred in subjects who eventually survived (Figs. 3B, 4 and Fig. S1). Consistent with prior reports that NfL levels remain elevated for weeks (up to 3 months) following acute CNS injury,20 increased NfL in COVID-19 patients did not return to normal within the observation period. In contrast, ALC, LDH, and CRP demonstrated large day-to-day fluctuations (Fig. 4) and were also frequently elevated in surviving patients (Fig. S1 and Appendix S1).
To assess if transient abnormalities in LDH, CRP, and ALC levels precede increases in NfL, we investigated correlations between these systemic markers measured at initial timepoints (T1 and T2), with NfL measured later (i.e., T1 vs. T2, T1 vs. T3, and T2 vs. T3). Only three of these comparisons reached statistical significance (Fig. 3C), with the strongest relationship observed between LDH measured at first time point (T1) and NfL measured at last time point (T3), which explains almost 60% of variance (R2 = 0.598, p = 0.0001). Consistent with the lack of association of CRP measurements with COVID-19 severity, CRP elevations did not predict subsequent rise in NfL.
We conclude that critically ill COVID-19 patients experience earlier abnormalities in ALC and LDH measurements, which are strongly associated with later elevation in NfL levels.
NfL measured later during hospitalization enhances mortality prediction of age- and gender-based classifier
As all the above-described observations supported the clinical value of NfL to predict COVID-19 mortality, we sought to quantify this predictive value on an individual patient level and compare it to demographic prognostic markers such as age, gender, and comorbidities.
In cohort 1, used as a training cohort, we predicted COVID-19 mortality using measured NfL as a continuous variable (Fig. 5A, left panel). Single, cross-sectional NfL measurements could not reliably predict death, reaching an area under receive operator characteristic curve (AUROC) of only 0.61 with a 95% confidence interval ([CI]: 0.33–0.89) crossing the value of random guessing (i.e., AUROC 0.5). The optimal cut-off from NfL to predict mortality from cohort 1 ROC curve was 124 pg/mL.
As shown in Table 1, cohorts 1 and 2 were not matched for demographic predictors of COVID-19 mortality: in both cohorts, patients who survived were generally younger, with a higher proportion of females and a lower proportion of subjects with comorbidities. Therefore, it should not be surprising that NfL measurements alone, ignoring these important demographic variables, had low predictive power. Instead, we built a prognostic classifier that integrated NfL (dichotomized based on optimal cut-off 124 pg/mL) with age and gender and compared it to the model(s) without NfL. We also tested a more complex classifier consisting of dichotomized NfL, age, gender, and comorbidities, but observed weaker independent validation of this model compared to a model without comorbidities (Fig. S2). For the sake of space and clarity, we will present data only on the strongest model.
Adding dichotomized NfL enhanced the predictive value of age and gender in cohort 1 from AUROC 0.80 (CI: 0.58–1.00) to 0.85 (0.66–1.00) and p-value from 0.023 to 0.0068 (Fig. 5A and Fig. S3A).
Next, we sought to assess the performance of the leading mortality predictor in cohort 2, which did not contribute to model generation (Fig. 5B and Fig. S3B). The addition of dichotomized NfL to the age and gender at first longitudinal timepoint (T1) did not improve the predictive value of the model, consistent with the observation that at the early timepoint the NfL values were indistinguishable between patients who survived and those who died. In contrast, NfL significantly improved the predictive power of the combined classifier at later timepoints (T2 and T3; T2: AUROC from 0.76 (CI: 0.53–1.00) to 0.89 (CI: 0.74–1.00) and p-value from 0.06 to 0.0048; T3: AUROC from 0.76 (0.53–1.00) to 0.96 (0.87–1.00) and p-value from 0.06 to 0.00094).
We conclude that NfL measurement provides additive COVID-19 mortality predictive value to the traditional demographic prognostic factors, provided that NfL is measured in critically ill patients later in the disease.
Finally, we were able to assess the non-redundant prognostic value of NfL in a unique large cohort of patients with high COVID-19 mortality risk (i.e., elderly patients with high proportion of males with comorbidities; Fig. 5C). As expected, out of these 288 critically ill COVID-19 patients, a large proportion (n = 113; 39.2%) eventually died.
Although surviving and dying cohorts were matched for age, gender, and comorbidities as univariate predictors (Table 1), the combined age plus gender model correctly predicted marginally higher mortality in the cohort of subjects who eventually died (10% vs. 93%; p = 0.047). NfL levels differentiated survivors from non-survivors with much stronger statistical significance (p = 4.1e-08). Adding dichotomized NfL to demographic data improved the accuracy of mortality prediction compared to demographic data alone. Specifically, the AUROC increased from 0.57 (CI: 0.50–0.64) to 0.62 (CI: 0.55–0.69) and p-value improved from 0.047 to 0.00063 (Fig. 5C and Fig. S3C). Nevertheless, the sensitivity (71.4%) and specificity (40.7%) of this predictor remained weak in this unique cohort.
The LDH, CRP, and lymphopenia were previously associated with COVID-19 mortality, especially in Chinese patients where a tree-based classifier (XGBoost) that included all three of these biomarkers achieved greater than 90% accuracy in predicting death in the independent validation cohort.14 Unfortunately, this model failed to validate in a cohort of Dutch (Caucasian) patients.21 As we did observe in univariate analyses prognostic value of LDH and ALC in our Caucasian (i.e., Italian) cohort, we construed a model that included age, gender, and these three biomarkers using modeling strategy analogous to our best validated NfL mortality predictor (Fig. 5: column 4 and Fig. S3: column 3). This model outperformed the winning NfL model in the training cohort (cohort 1), achieving AUROC of 0.91 (CI: 0.79–1.00) and p = 0.0011. However, consistent with our univariate observation, this last classifier outperformed the winning NfL model only in the earliest timepoint (T1) of independent longitudinal cohort 2 (T1: AUROC = 0.88 [CI: 0.69–1.00] and p = 0.0056). Its performance was inferior to the winning NfL model in cohort 2 for later timepoints (T2: AUROC = 0.80 [CI: 0.59–1.00] and p = 0.031; T3: AUROC = 0.80 [0.59–1.00] and p = 0.031). Finally, the model containing LDH, ALC, and CRP completely failed to validate in cohort 3 (AUROC = 0.60 [CI: 0.45–0.75] and p = 0.19 [Fig. 5 and Fig. S3]), although one must note that we did not have these laboratory values for all subjects in cohort 3.
An increase in serum or CSF NfL has been previously associated with increased ICU mortality due to sepsis-associated encephalopathy.8 This study expands these data to COVID-19 ICU admissions: First, we validated reports linking high serum/plasma NfL levels to COVID-19 severity.9–13, 22, 23 Our longitudinal measurements demonstrated that rise in NfL generally occurs during hospitalizations of critically ill patients and trails other transient laboratory abnormalities such as decreased ALC and increased LDH by 5–20 days. The degree of LDH increase is a strong determinant of the subsequent magnitude of NfL rise, suggesting that COVID-19-associated CNS injury is secondary to damage of other critical organs, such as liver, kidneys, and lungs. This conclusion aligns with pathology studies ruling out strong primary infiltration of CNS tissue by the SARS-CoV-2 or by the immune system; those studies instead attribute COVID-19-associated CNS damage to processes such as hypoxia or intravascular coagulation.24
Compared to previous studies of NfL in COVID-19,9–13, 22, 23 we studied a cohort of patients in which a high proportion eventually died (133/338 = 39.3%). This allowed us to unequivocally link high serum/plasma NfL levels with COVID-19 mortality, something that remained ambiguous in the previous studies.
We constructed a model that combined demographic predictors of COVID-19 mortality with NfL measurement and validated its greater predictive accuracy. Nevertheless, the accuracy of this classifier varied between the cohorts, depending on the timing of NfL measurement (i.e., later measurements enhanced predictive power) and underlying premorbid risk. Indeed, comparing model performance among our three cohorts, it appeared that NfL has greater predictive value in younger (cohorts 1 and 2) versus older (cohort 3) subjects. This is perhaps not surprising as younger patients with fewer comorbidities have a higher likelihood of withstanding multi-organ failure and therefore CNS injury may become a key determinant of their survival. In contrast, elderly subjects with high premorbid risk and greater vulnerability of CNS tissue to sepsis-associated injury rapidly succumb to multi-organ failure before CNS injury manifests by high NfL concentrations.25
Although speculative at the moment, integrating all our observations, we recommend that NfL should be measured longitudinally and integrated with existing prognostic markers to optimize care. For example, a screening NfL measurement at the beginning of hospitalization, expected to be normal in most patients, might identify a few subjects with either neurological comorbidity or with advanced stage of COVID-19 who require care focused on preventing further CNS injury. After an initial negative NfL test, critically ill COVID-19 patients might be best monitored by standard laboratory tests such as LDH and ALC. Identified spikes should prompt more aggressive management that includes longitudinal NfL monitoring approximately every 5 days. Any increase in NfL should be considered a poor prognostic indicator necessitating escalation therapies, including neuroprotective strategies. Stabilization of NfL levels indicates that escalation therapy worked, while further increases signify continuous neuro-axonal injury that must be stopped to limit mortality.
While the COVID-19 pandemic demonstrated the prognostic value of NfL in critically ill patients with SARS-CoV-2 infection, noninvasive, ultrasensitive measurement of NfL could be used to monitor neuronal injury in all comatose, or heavily sedated critically ill patients regardless of SARS-CoV-2 infection status. Ultra-sensitive assays will hopefully become broadly adopted by clinical laboratories and might include in the future other CNS-derived analytes for enhanced accuracy of noninvasive monitoring of CNS tissue.
Severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) causes a wide spectrum of clinical manifestations, with progression to multiorgan failure in the most severe cases. Several biomarkers can be altered in coronavirus disease 2019 (COVID-19), and they can be associated with diagnosis, prognosis, and outcomes. The most used biomarkers in COVID-19 include several proinflammatory cytokines, neuron-specific enolase (NSE), lactate dehydrogenase (LDH), aspartate transaminase (AST), neutrophil count, neutrophils-to-lymphocytes ratio, troponins, creatine kinase (MB), myoglobin, D-dimer, brain natriuretic peptide (BNP), and its N-terminal pro-hormone (NT-proBNP). Some of these biomarkers can be readily used to predict disease severity, hospitalization, intensive care unit (ICU) admission, and mortality, while others, such as metabolomic and proteomic analysis, have not yet translated to clinical practice. This narrative review aims to identify laboratory biomarkers that have shown significant diagnostic and prognostic value for risk stratification in COVID-19 and discuss the possible clinical application of novel analytic strategies, like metabolomics and proteomics. Future research should focus on identifying a limited but essential number of laboratory biomarkers to easily predict prognosis and outcome in severe COVID-19.
Severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) causes a wide spectrum of clinical manifestations, from mild respiratory symptoms to pneumonia and, in more severe cases, multiple organ failure (1). The mechanisms underlying multisystem involvement may include an unbalanced immune response that facilitates the progression of coronavirus disease-2019 (COVID-19). This hypothesis has been confirmed by laboratory biomarker alterations, showing greater potential for abnormal immune response, mainly an increase in neutrophil counts and a substantial reduction in lymphocyte counts, thus altering the neutrophil-to-lymphocyte ratio. Such an abnormal immune response is driven by an increased serum concentration of many pro-inflammatory mediators. These include interleukin (IL)-1β, IL-2, IL-6, IL-8, interferon (IFN)-γ-induced protein 10, granulocyte colony-stimulating factor, monocyte chemoattractant protein 1, macrophage inflammatory protein-1α, and tumor necrosis factor-α, among others (2–5). Nevertheless, the inflammatory cytokine storm in patients with COVID-19 is less injurious than that observed in patients with sepsis or acute respiratory distress syndrome (ARDS) but without COVID-19 (6), thus raising questions regarding the mechanisms underlying multiorgan involvement in COVID-19.
Several biomarkers other than cytokines have been found altered in COVID-19, and are associated with diagnosis, prognosis and outcomes (7). Some of these biomarkers can be easily used to predict disease severity, hospitalization, intensive care unit (ICU) admission, and mortality, while others, like metabolomic and proteomic analysis, are still of purely investigational concern and difficult to translate into clinical practice, despite their prognostic potential (8–10).
The aim of this narrative review is to identify laboratory biomarkers that have shown significant diagnostic and prognostic value for risk stratification in COVID-19 and to discuss the possible clinical application of novel analytic strategies, such as metabolomics and proteomics.
SARS-CoV-2 is an enveloped, single-stranded ribonucleic acid (ssRNA) virus. The SARS-CoV-2 genome is composed of two polypeptides encoded between two open-reading frames that are processed by viral proteases to produce nonstructural proteins (11). These proteins are involved in viral replication and suppression of host innate immune defense. On the other hand, structural proteins of SARS-CoV-2 include the spike (S), envelope (E), and nucleocapsid (N) protein, as well as the membrane (M) glycoprotein. The S protein is a transmembrane glycoprotein that is located on the viral surface and cleaved by host-cell proteases. After anchoring the S protein, SARS-CoV-2 enters host cells via angiotensin receptor-2 (ACE2), thus activating transmembrane serine protease 2 (TMPRSS2), cathepsin B and L. The E protein is a glycoprotein involved in virion maturation and pathogenesis, while the M protein is involved in viral assembly and delineates the shape of the viral envelope; finally, the N protein binds directly to viral RNA (11). The pathogenic mechanisms of SARS-CoV-2 include 1) direct epithelial damage, 2) dysregulated immune response, 3) ACE2 dysregulation and downregulation of the renin-angiotensin- aldosterone system (RAAS), 4) direct endothelial damage, and, possibly, 5) tissue fibrosis (11). Hence, patients with severe COVID-19 are at high risk of multiple organ involvement and, ultimately, death. Indeed, the virus has been identified in multiple tissues, including endothelial, liver, kidney, pulmonary, and neuronal cells, suggesting direct invasion as possible pathological mechanism underlying systemic effects (1). Therefore, laboratory biomarkers of organ damage play a key role in the diagnosis, prediction, and prognosis of patients at high risk of multiorgan involvement, and their use should be implemented in clinical practice (1). Table 1 summarizes the most investigated biomarkers in COVID-19, while Figure 1 depicts possible multiorgan involvement in COVID-19. In the following section, we will describe individual organ systems and how they can be affected by severe COVID-19, associated laboratory and clinical biomarkers of damage, severity, and outcome, and their potential utility for patient management.
COVID-19 multiple organ dysfunction. This figure shows the potential for multiorgan involvement in COVID-19. Respiratory (AIP, acute interstitial pneumonia; ARDS, acute respiratory distress syndrome; DAD, diffuse alveolar damage), renal, cardiovascular, coagulative/hemostatic, liver, gastrointestinal, metabolic/endocrine, and cerebral functions and systems, as well as their possible alterations, are presented.
Biomarkers reflecting multiple organ involvement and/or pharmacological effects have been widely examined in critically ill patients. Some of these biomarkers are also used to monitor dysfunction in distinct organs at the same time, due to their redundancy or non-specificity. However, the most appropriate biomarkers to be studied in critically ill patients with COVID-19 have yet to be defined. Figure 2 depicts a proposed algorithm for critical care management which includes the investigation of biomarkers in severe COVID-19 patients at ICU admission.
Proposed algorithm for the management of patients with COVID-19 at ICU admission. This figure shows a potential algorithm for initial patient management at ICU admission, including the most useful biomarkers to be used in the COVID-19 critical care setting. Neurological system: sequential transcranial doppler (TCD) and/or optic nerve sheath diameter (ONSD) in sedated patients for whom conventional neurological evaluation is impossible. Cardiovascular system: electrocardiogram and echocardiography, as well as continuous monitoring of mean arterial pressure (MAP) and heart rate (HR), are suggested on ICU admission. Respiratory system: computed tomography (CT) scan is the gold standard; if not feasible, chest X-ray, CT angiography, and/or lung ultrasound should be performed. Lactate dehydrogenase (LDH), C-reactive protein (CRP), neuron specific enolase (NSE), neurofilament light polypeptide (NfL), glial fibrillary acidic protein (GFAP), thyrotropic stimulating hormone (TSH), NGAL, aspartate transaminase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (γGT), interleukin-6 (IL-6). BNP, brain natriuretic peptide; UN, urea nitrogen; NT-proBNP, N-terminal pro-hormone.
The lungs are usually the organs affected primarily by SARS-CoV-2, due to their large and highly vascularized surface area (11). The pathogenesis of COVID-19 in the lung includes an initial phase of local inflammation, endothelial cell damage, and antifibrinolytic activation in the upper and lower respiratory tracts, followed by repair mechanisms that can elicit the restoration of normal pulmonary architecture. Inflammation is followed by platelet recruitment with degranulation, clot formation, altered vessel permeability, and accumulation of leukocytes in the injury site, leading to the recruitment of other inflammatory cells with the involvement of specific cytokines (i.e., IL-4, IL-13, transforming growth factor-β) that are also responsible for pro-fibrotic activity (12).
SARS-CoV-2 lung infection causes a wide variety of clinical manifestations and symptoms, from asymptomatic, mild, and moderate disease to severe COVID-19. Severe and critical illness accounts for up to 14% and 5% of cases, respectively, with the ARDS occurring in 10-20% of patients; multiorgan failure and death may supervene (13, 14). Various phenotypes have been identified by computed tomography (CT) (15, 16), including phenotype L or 1, which is characterized by low compliance, altered ventilation and perfusion, and shunting with focal hypo/hyper-perfused ground-glass opacities; and phenotype H or 2, which is identified by an inhomogeneous distribution of atelectasis with a patchy ARDS-like pattern (17, 18). Progressive evolution of COVID-19 (19) may lead to phenotype F, caused by mechanical stretch of lung epithelial cells and pathological fibro-proliferation and remodeling of the extracellular matrix, with increased expression of pro-fibrotic markers, as is mainly typical of severe forms of lung disease (20).
Although not specific to pulmonary disease, several biomarkers of different stages of lung involvement in COVID-19 have been identified and have been associated with pulmonary and systemic hyperinflammation and fibrotic damage (12). In the early disease course, neuron-specific enolase (NSE) can be used to differentiate patients who are going to develop dyspnea (21). On admission, higher lymphocyte and platelet counts and lower ferritin, D-dimer, lactate dehydrogenase (LDH), and aspartate transaminase (AST) have all been associated with lower risk of mortality in COVID-19 patients who ultimately required intubation and mechanical ventilation (22). Surfactant protein-D, angiopoietin-2, triggering receptor expressed on myeloid cell (TREM)-1, and TREM-2 levels were found to be higher in mild/moderate and severe/critical COVID-19 pneumonia than in asymptomatic and uncomplicated cases. Moreover, these biomarkers correlated well with clinical severity (23, 24). In severe COVID-19 cases, total thiol, ferritin, and LDH were identified as prognostic biomarkers for ARDS development (25). At extubation, COVID-19 survivors had higher platelet counts and neutrophil-to-lymphocyte ratios and lower C-reactive protein (CRP), D-dimer, ferritin, LDH, and AST (22).
Infection and Systemic Inflammatory Response
Following SARS-CoV-2 invasion of the host cells, the virus replicates at the infection site, thus triggering activation of the innate and adaptive immune responses (26). Neutrophils are rapidly recruited to infection foci, while innate cells recognize the virus and secrete multiple cytokines. Antigen-presenting cells recognize viral antigens which are carried to the local lymph nodes, while activating the T-helper cell response, which is also responsible for stimulating B cells to secrete antibodies (27). The systemic immune-inflammatory response is activated; if left unchecked, this may progress to multiorgan illness (28).
Patients with severe COVID-19 are highly susceptible to superimposed bacterial, fungal, and viral infections, including ventilator-associated pneumonia and bloodstream infection, among others (29, 30). As for systemic biomarkers of infection, procalcitonin is a predictor of disease severity (31), and can be useful to guide antimicrobial stewardship (32, 33). Another study found an association between procalcitonin and mortality in COVID-19 patients more than 75 years old (34). Neutrophil count was also predictive of clinical outcome in hospitalized COVID-19 patients (35), while the neutrophil-to-lymphocyte ratio was strongly associated with severity and mortality in COVID-19 (36). Additionally, total lymphocyte count, cluster differentiation (CD)3+, CD4+, CD8+, CD25+, CD127– T cells, and natural killer (NK) cells were found to be depressed in severe COVID-19 (37), whereas C-reactive protein, erythrocyte sedimentation rate, and IL-6 – common markers of inflammation – were elevated (38).
SARS-CoV-2 can directly trigger endothelial dysfunction, causing a status known as COVID-19-associated coagulopathy. After viral entry into the cells, increased vascular permeability and tissue factor expression in subendothelial cells, with activation of platelets and leukocytes, may trigger the coagulation cascade. Endothelial damage and a generalized inflammatory state are drivers of thrombosis, which can contribute to cardiovascular manifestations (39).
Cardiovascular manifestations of COVID-19 are frequently reported (2, 40). Acute heart failure and exacerbation of chronic heart failure are reported in up to 20-30% of hospitalized patients, and carry high mortality rates, especially in patients with severe comorbidities (41–43). Acute coronary syndrome has been reported in a high proportion of patients, probably because of plaque rupture, coronary spasm, or microthrombi triggered by systemic inflammation and cytokine storm (44). In general, the mechanisms underlying cardiovascular manifestations include increased cardiac workload, hypoxemia, hypervolemia, myocardial injury, arrhythmias, myocarditis, stress-induced cardiomyopathy, acute kidney injury, and, as noted above, systemic inflammatory response with the release of several cytokines and chemokines (45). Triggering mechanisms may be attributed to an imbalance between heightened cardiac workload and reduced oxygen supply secondary to systemic conditions, with possible type-2 myocardial infarction (46).
Cardiac biomarkers (47), electrocardiography (ECG), and transthoracic echocardiography (TTE) play a pivotal role in risk stratification and early detection of cardiovascular complications, as well as to guide treatment (48, 49). Recent evidence confirmed that cardiac biomarkers, including natriuretic peptides (NPs) and troponins, may reflect cardiovascular involvement and inflammation in COVID-19, and are strongly associated with poor prognosis and mortality (41, 50–53). In some cases, troponin elevation in COVID-19 has been associated with ECG changes (54), ICU admission, and in-hospital death (55, 56). However, despite the confirmed prognostic impact of troponins, routine testing is still a matter of debate, because of several other variables that have been associated with outcome and prognosis (48). Additionally, pre-existing cardiac disease and/or acute stress injury may justify mild elevations in cardiac troponins, while myocarditis, Takotsubo syndrome, type 2 myocardial infarction triggered by severe respiratory failure, systemic hypoxemia, or shock are mostly associated with more marked increase in troponins (44, 57, 58). Other cardiac and non-cardiac biomarkers are common findings in COVID-19-associated cardiovascular disease, including creatine kinase (CK)-MB, myoglobin, D-dimer, brain natriuretic peptide (BNP) and its N-terminal pro-hormone (NT-proBNP), and neutrophil-to-lymphocyte ratio (55, 59–61). Myoglobin seems to offer higher prognostic accuracy than other cardiac-specific biomarkers (troponins and CK-MB) in COVID-19 (62). Moreover, mid-regional pro-adrenomedullin (MR-proADM) levels were found to be associated with endothelial dysfunction and mortality in COVID-19, potentially making it an optimal biomarker for the prediction of survival in this patient population (63). Nevertheless, only limited evidence exists so far to define any of these biomarkers as an independent predictor of prognosis in COVID-19 (48, 64).
Coagulation and Hemostasis
Coagulation derangement is a well-known systemic effect of COVID-19 that can originate from direct or indirect viral impact on the endothelium, or from immunothrombosis (65). COVID-19 can cause alterations in the coagulation cascade, with imbalance of the regulatory mechanisms of coagulation and fibrinolysis, altered platelet function, and a hyperinflammatory response (11, 65). In this context, D-dimer has been identified among the first altered coagulation biomarkers in COVID-19, and is predictive of mortality on admission (66). Similarly, plasma fibrinogen appears to be associated with hyperinflammation and disease severity in COVID-19 (67). A coagulopathy signature diagnostic of COVID-19 has been identified, including elevated levels of soluble vascular cell adhesion molecule (sVCAM)-1 (68), von Willebrand Factor (vWF), thrombomodulin, soluble tumor necrosis factor (TNF) receptor I (sTNFRI), heparan sulfate, C5b9 complement, plasminogen activator inhibitor (PAI)-1, and alpha-2 antiplasmin, among others. Some of these markers, such as sVCAM-1, vWF, sTNFRI, and heparan sulfate, were also associated with disease severity (69). Fibrinogen, thrombin peak, vWF, and ADAMTS13 at admission and elevated vWF : Ag to ADAMTS13 activity ratio were associated with severity and higher risk of death (70, 71). Endothelial dysfunction seems to be persistent after resolution of COVID-19, and directly associated with the severity of pulmonary impairment (72).
Sphingolipid metabolism regulates the inflammation and immune response through the conversion of sphingosine to sphingosine 1-phosphate, increasing the release of lymphocytes into the blood, with subsequent systemic inflammation and release of cytokines and chemokines in COVID-19 (73). Like lipid metabolism, fat-soluble vitamins such as vitamin D have been implicated in suppressing the cytokine storm and enhancing the immune response (74). Investigating lipid metabolism and its biomarkers could thus be of diagnostic and prognostic value in COVID-19.
Metabolic comorbidities including obesity, diabetes, cardiovascular, and hypertension have been associated with poor prognosis in COVID-19 (75). A certain degree of metabolic dysregulation has been found in COVID-19, possibly due to immune-triggered inflammation and hypercoagulability, as well as microbial changes in host physiology (10, 76). Indeed, COVID-19 patients with lower levels of high-density lipoprotein (HDL) cholesterol are more susceptible to hospitalization, while low-density lipoprotein (LDL) cholesterol was associated with higher inflammation (77). Critically ill patients with COVID-19 showed significantly lower levels of vitamin A than non-critical ones, and this was associated with higher inflammation (78). Vitamin A levels below 0.2 mg/L were significantly associated with the developments of ARDS and higher mortality (78). Vitamin D, a well-known regulator of phosphate and calcium metabolism with immunomodulatory functions, seems to not influence mortality or hospital length of stay in COVID-19 (79, 80). Finally, thyroid hormones showed marked association with disease severity and mortality, suggesting the importance of early assessment of thyroid function – and, when necessary, initiation of treatment – in hospitalized COVID-19 patients (81).
Pathogenetic mechanisms of SARS-CoV-2 neurologic manifestations include possible spreading of the virus across the blood-brain barrier via leukocyte migration or sluggish movement of blood within the microcirculation, thus binding to endothelial cells. Cells which may present ACE2 receptors, including neurons, astrocytes, and oligodendrocytes, can all be affected directly by viral entry and activate the local immune response. As a consequence of neuronal involvement, several biomarkers of neuroinflammation and damage can be detected (82).
Although COVID-19 rarely affects the brain as a primary manifestation, neurological complications are common in this patient population (82–84). Patients with neurological complications, compared to those without, may experience longer hospital stays, and the duration of mechanical ventilation can be associated with the risk of developing new neurological complications (84, 85). CT and magnetic resonance imaging (MRI) are considered the gold standard for detecting cerebral derangements, although the use of methods which involve exposure to ionizing radiation in non-primarily brain-injured patients can only be justified in case of high suspicion of neurological complications (86). The use of multimodal neuromonitoring has received increasing attention as a means of identifying patients at higher risk of brain derangement because of its low cost, speed, safety, and ready availability. However, the use of neuromonitoring tools is still mainly limited to specific settings (i.e., ICU) and patient populations (i.e., those with primary brain injury) (84).
Other than imaging, blood biomarkers can detect brain damage and predict prognosis efficiently. Blood biomarkers for the study of brain derangements include glial fibrillary acidic protein (GFAP), neurofilament light polypeptide (NfL), tau, S100B calcium binding protein, NSE, and inflammatory markers. Increased GFAP staining has been found in postmortem analysis of brain tissue from patients with COVID-19 (87), and NfL was significantly associated with COVID-19 status (88). Another study reported that GFAP was increased in both moderate and severe COVID-19 cases, whereas serum NfL was increased only in severe cases compared to controls (89). However, another study reported that serum NfL, although elevated across patients hospitalized with COVID-19, was not associated with neurological manifestations. Additionally, the usual close correlation between cerebrospinal fluid and serum NfL was not found, suggesting serum NfL elevation in the non-neurological patients may reflect peripheral nerve damage in response to severe illness (90). In COVID-19 patients with altered NfL and GFAP, values of these markers had normalized in all individuals at 6-month follow-up, suggesting that post-COVID-19 neurological sequelae may be not accompanied by ongoing brain injury (91). Inflammatory and coagulatory markers like D-dimer, LDH, erythrocyte sedimentation rate (ESR), and CRP were independently associated with the occurrence of ischemic stroke in COVID-19 (92, 93), while higher age, diabetes mellitus, and hypertension were found not to be significant predictors of stroke in this population, despite being known predictors of non-COVID-19 stroke (93). Levels of lymphocytes, procalcitonin, and creatinine were higher in COVID-19 stroke patients (94). S100B was higher in patients with mild and severe COVID-19 than in healthy controls, and may be a marker of disease severity (95). Antiphospholipid antibodies (i.e., anti-phosphatidylserine/prothrombin) were higher in COVID-19 patients, particularly those with neurological manifestations, than in controls. In contrast, anticardiolipin antibodies were not associated with neurologic involvement in COVID-19 (96).
Kidney and Liver
COVID-19 may cause kidney and liver injury by either direct infection of cells, via host immune clearance and immune tolerance disorders, endothelium-associated vasculitis, thrombus formation, metabolism and glucose disorder, or tissue hypoxia. As a consequence, biomarkers of endothelial, renal, hepatic, vascular, or hypoxic damage can help in the detection of new organ involvement and assist in determining prognosis (97).
As part of multiorgan involvement in COVID-19, kidney function might be altered directly by viral invasion or may occur secondary to multiple organ failure due to systemic inflammation or aggressive therapies (98). Around 25% of patients hospitalized with COVID-19 were reported to develop acute kidney injury, including low molecular weight proteinuria, Fanconi syndrome, and tubular injury (98). Moreover, regional inflammation, endothelial injury, and microthrombi have been identified as major causative factors of renal pathology in COVID-19. This is also sustained by the fact that anti-inflammatory drugs, such as steroids, play a key role in limiting renal disease progression (98). Classic diagnostic biomarkers of kidney damage include creatinine, neutrophil gelatinase-associated lipocalin (NGAL), cystatin C, kidney injury molecule-1 (KIM-1), blood and urinary urea nitrogen, and urinary proteins (99, 100).
Novel urinary biomarkers have been proposed in COVID-19, including urine 11-dehydro-thromboxane B2, 8-hydroxy-2′-deoxyguanosine, and liver-type fatty acid binding protein (L-FABP) levels, all of which were higher in this patient cohort at the time of hospitalization (101). N-acetyl-β-D-glucosaminidase, β2-microglobulin, α1-microglobulin, and L-FABP, which are all markers of tubular injury, were significantly associated with inflammation, as were IL-6 levels (102). Indeed, another observational study confirmed the association between pro-inflammatory cytokines, urinary cytokines, and urinary kidney injury markers (103). Procalcitonin was associated with acute kidney injury in COVID-19, and a score including simple and easily accessible variables such as procalcitonin, arterial saturation of oxygen, and blood urea nitrogen was shown to be predictive of acute kidney injury (104).
Altered serum creatinine levels with decreased kidney function at admission and up to 24 hours thereafter were significantly associated with acute kidney injury and in-hospital mortality (105). Additionally, urine blood >0.03 mg/dL and urine specific gravity >1.026 were associated with acute kidney injury, ICU admission, and higher mortality (106).
Abnormal liver and hepatobiliary function have been also identified in COVID-19 (107). A systematic review and meta-analysis showed a cumulative prevalence of liver disease of 24% in COVID-19, with possible alterations in albuminemia, liver enzymes, and total bilirubin (108). Recent findings showed that some liver and renal biomarkers, including albumin, direct bilirubin, neutrophil and lymphocyte counts, and mean corpuscular hemoglobin, are associated with risk of developing severe COVID-19 (107). Moreover, the presence of pre-existing liver fibrosis with silent liver injury significantly influenced mortality in COVID-19 (109)
Future Perspectives: Metabolomic and Proteomic Biomarkers and Machine Learning Models
Given the significant immune dysregulation of COVID-19 patients, the interplay between metabolism and immunity may play a pivotal role in the disease course (110). Additionally, oxygen deprivation may affect homeostasis in tissues and organs such as the lung, brain, kidney, and liver. The modulation of oxygen homeostasis and response to hypoxia is mainly mediated by glycolysis and the lactate cycle. This has increased research interest in proteomic and metabolomic methods to investigate pathways linked to energy production and amino acid metabolism in patients with SARS-CoV-2 infections (110). Metabolomic analyses in COVID-19 patients with and without pulmonary fibrosis revealed that pathways including the peroxisome proliferator-activated receptor (PPAR), D-arginine and D-ornithine metabolism, inflammatory tryptophan metabolic pathway (TRP), and alpha-linolenic acid metabolism were significantly increased in fibrotic lungs, thus suggesting that PPAR signaling is one of the main pathways involved in the formation and development of lung fibrosis in COVID-19 (9). A proteomic and metabolomic analysis identified hypoxanthine and betaine as predictors of ICU stay, and early ICU admission, elevated creatinine, and D-dimer were found to be associated with these pathways (8). Longer duration of invasive mechanical ventilation was associated with the kynurenine and p-cresol sulfate pathways (8). Several markers of metabolic function identified via metabolomic analysis were associated with in-hospital mortality, including cyclic adenosine monophosphate (cAMP), which plays a role in SARS-CoV2 endocytosis in the initial phase of the disease (10). Another major signature of the serum metabolome in COVID-19 was lactic acid, as well as spermidine and spermine. Many other metabolites were commonly increased, including glutamate, aspartate, phenylalanine, β-alanine, ornithine, arachidonic acid, choline, and xanthine (110). Recent machine learning models have been developed to support decision making and risk stratification in COVID-19. Most predictive models rely on demographic and clinical variables. However, biomarkers have recently shown good correlation with severity of disease and mortality in COVID-19 modeling (111). One example was a large study of 2,895 consecutive patients with COVID-19 in whom three biomarkers measured at admission were found to reflect pathobiological axes of myocardial injury, altered coagulation, and inflammation. The machine learning model concluded that patients with low levels of these biomarkers were at lower risk of critical disease and in-hospital mortality (112). In conclusion, the alterations found in the serum metabolome of patients with COVID-19 may reflect a more complex systemic derangement affecting carbon and nitrogen liver metabolism, but further research is needed to completely understand the impact of these alterations on routine clinical practice. Machine learning models can be promising in risk stratification in COVID-19. However, further investigations are needed to develop mathematical models that can help clinicians select the right parameters and interpret results.
Laboratory biomarkers have shown significant diagnostic and prognostic value for risk stratification in COVID-19. Furthermore, novel analytic strategies including metabolomics and proteomics offer interesting insights for early detection of patients at higher risk of severe disease and death. However, their limited availability restricts their widespread clinical use. Further investigations are warranted to identify a core set of laboratory biomarkers which can be used in daily clinical practice to easily predict prognosis and outcome in hospitalized patients with severe COVID-19.
DB and ML-P: review, design, writing, editing. HC-F-N and PP: editing. PR: review, design, editing, senior contribution. All authors contributed to the article and approved the submitted version.
This work was supported by the Brazilian Council for Scientific and Technological Development (COVID-19-CNPq; 401700/2020-8 and 403485/2020-7); Rio de Janeiro State Research Foundation (COVID-19-FAPERJ; E-26/210.181/2020); and Funding Authority for Studies and Projects (01200008.00), Brazil.
Conflict of Interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
1. Robba C, Battaglini D, Pelosi P, Rocco PRM. Multiple Organ Dysfunction in SARS-CoV-2: MODS-CoV-2. Expert Rev Respir Med (2020) 14:865–8. doi: 10.1080/17476348.2020.1778470 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
2. Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, et al.. Clinical Features of Patients Infected With 2019 Novel Coronavirus in Wuhan, China. Lancet (2020) 395:497–506. doi: 10.1016/S0140-6736(20)30183-5 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
3. Zhang C, Wu Z, Li J-W, Zhao H, Wang G-Q. Cytokine Release Syndrome in Severe COVID-19: Interleukin-6 Receptor Antagonist Tocilizumab May Be the Key to Reduce Mortality. Int J Antimicrob Agents (2020) 55:105954. doi: 10.1016/j.ijantimicag.2020.105954 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
4. Liu J, Li S, Liu J, Liang B, Wang X, Wang H, et al.. Longitudinal Characteristics of Lymphocyte Responses and Cytokine Profiles in the Peripheral Blood of SARS-CoV-2 Infected Patients. EBioMedicine (2020) 55:102763. doi: 10.1016/j.ebiom.2020.102763 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
5. Ackermann M, Verleden SE, Kuehnel M, Haverich A, Welte T, Laenger F, et al.. Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19. N Engl J Med (2020) 383:120–8. doi: 10.1056/NEJMoa2015432 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
6. Leisman DE, Ronner L, Pinotti R, Taylor MD, Sinha P, Calfee CS, et al.. Cytokine Elevation in Severe and Critical COVID-19: A Rapid Systematic Review, Meta-Analysis, and Comparison With Other Inflammatory Syndromes. Lancet Respir Med (2020) 8:1233–44. doi: 10.1016/S2213-2600(20)30404-5 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
7. Andrianto, Al-Farabi MJ, Nugraha RA, Marsudi BA, Azmi Y. Biomarkers of Endothelial Dysfunction and Outcomes in Coronavirus Disease 2019 (COVID-19) Patients: A Systematic Review and Meta-Analysis. Microvasc Res (2021) 138:104224. doi: 10.1016/j.mvr.2021.104224 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
8. Taleb S, Yassine HM, Benslimane FM, Smatti MK, Schuchardt S, Albagha O, et al.. Predictive Biomarkers of Intensive Care Unit and Mechanical Ventilation Duration in Critically-Ill Coronavirus Disease 2019 Patients. Front Med (2021) 8:733657. doi: 10.3389/fmed.2021.733657 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
9. Yang J, Chen C, Chen W, Huang L, Fu Z, Ye K, et al.. Proteomics and Metabonomics Analyses of Covid-19 Complications in Patients With Pulmonary Fibrosis. Sci Rep (2021) 11:14601. doi: 10.1038/s41598-021-94256-8 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
10. Saccon E, Bandera A, Sciumè M, Mikaeloff F, Lashari AA, Aliberti S, et al.. Distinct Metabolic Profile Associated With a Fatal Outcome in COVID-19 Patients During the Early Epidemic in Italy. Microbiol Spectr (2021) 9:e00549–21. doi: 10.1128/Spectrum.00549-21 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
11. Lopes-Pacheco M, Silva PL, Cruz FF, Battaglini D, Robba C, Pelosi P, et al.. Pathogenesis of Multiple Organ Injury in COVID-19 and Potential Therapeutic Strategies. Front Physiol (2021) 12:593223. doi: 10.3389/fphys.2021.593223 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
12. Vianello A, Guarnieri G, Braccioni F, Lococo S, Molena B, Cecchetto A, et al.. The Pathogenesis, Epidemiology and Biomarkers of Susceptibility of Pulmonary Fibrosis in COVID-19 Survivors. Clin Chem Lab Med (2022) 60:307–16. doi: 10.1515/cclm-2021-1021 [PubMed] [CrossRef] [Google Scholar]
13. Chen N, Zhou M, Dong X, Qu J, Gong F, Han Y, et al.. Epidemiological and Clinical Characteristics of 99 Cases of 2019 Novel Coronavirus Pneumonia in Wuhan, China: A Descriptive Study. Lancet (2020) 395:507–13. doi: 10.1016/S0140-6736(20)30211-7 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
17. Gattinoni L, Chiumello D, Caironi P, Busana M, Romitti F, Brazzi L, et al.. COVID-19 Pneumonia: Different Respiratory Treatments for Different Phenotypes? Intensive Care Med (2020) 46:1099–102. doi: 10.1007/s00134-020-06033-2 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
18. Robba C, Battaglini D, Ball L, Patroniti N, Loconte M, Brunetti I, et al.. Distinct Phenotypes Require Distinct Respiratory Management Strategies in Severe COVID-19. Respir Physiol Neurobiol (2020) 279:103455. doi: 10.1016/j.resp.2020.103455 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
19. Tonelli R, Marchioni A, Tabbì L, Fantini R, Busani S, Castaniere I, et al.. Spontaneous Breathing and Evolving Phenotypes of Lung Damage in Patients With COVID-19: Review of Current Evidence and Forecast of a New Scenario. J Clin Med (2021) 10:975. doi: 10.3390/jcm10050975 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
20. George PM, Wells AU, Jenkins RG. Pulmonary Fibrosis and COVID-19: The Potential Role for Antifibrotic Therapy. Lancet Respir Med (2020) 8:807–15. doi: 10.1016/S2213-2600(20)30225-3 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
21. Cione E, Siniscalchi A, Gangemi P, Cosco L, Colosimo M, Longhini F, et al.. Neuron-Specific Enolase Serum Levels in COVID-19 Are Related to the Severity of Lung Injury. PLoS One (2021) 16:e0251819. doi: 10.1371/journal.pone.0251819 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
22. Topp G, Bouyea M, Cochran-Caggiano N, Ata A, Torres P, Jacob J, et al.. Biomarkers Predictive of Extubation and Survival of COVID-19 Patients. Cureus (2021) 13:e15462. doi: 10.7759/cureus.15462 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
23. Alay H, Laloglu E. The Role of Angiopoietin-2 and Surfactant Protein-D Levels in SARS-CoV-2-Related Lung Injury: A Prospective, Observational, Cohort Study. J Med Virol (2021) 93:6008–15. doi: 10.1002/jmv.27184 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
24. Kerget F, Kerget B, İba Yılmaz S, Kızıltunç A. Evaluation of the Relationship Between TREM-1/TREM-2 Ratio and Clinical Course in COVID-19 Pneumonia. Int J Clin Pract (2021) 75(10):e14697. doi: 10.1111/ijcp.14697 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
25. Martinez Mesa A, Cabrera César E, Martín-Montañez E, Sanchez Alvarez E, Lopez PM, Romero-Zerbo Y, et al.. Acute Lung Injury Biomarkers in the Prediction of COVID-19 Severity: Total Thiol, Ferritin and Lactate Dehydrogenase. Antioxidants (2021) 10:1221. doi: 10.3390/antiox10081221 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
26. V’kovski P, Kratzel A, Steiner S, Stalder H, Thiel V. Coronavirus Biology and Replication: Implications for SARS-CoV-2. Nat Rev Microbiol (2021) 19:155–70. doi: 10.1038/s41579-020-00468-6 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
28. Hamming I, Timens W, Bulthuis M, Lely A, Navis G, van Goor H. Tissue Distribution of ACE2 Protein, the Functional Receptor for SARS Coronavirus. A First Step Understanding SARS Pathogen J Pathol (2004) 203:631–7. doi: 10.1002/path.1570 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
29. Ippolito M, Misseri G, Catalisano G, Marino C, Ingoglia G, Alessi M, et al.. Ventilator-Associated Pneumonia in Patients With COVID-19: A Systematic Review and Meta-Analysis. Antibiotics (2021) 10:545. doi: 10.3390/antibiotics10050545 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
30. Ippolito M, Simone B, Filisina C, Catalanotto FR, Catalisano G, Marino C, et al.. Bloodstream Infections in Hospitalized Patients With COVID-19: A Systematic Review and Meta-Analysis. Microorganisms (2021) 9:2016. doi: 10.3390/microorganisms9102016 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
31. Shen Y, Cheng C, Zheng X, Jin Y, Duan G, Chen M, et al.. Elevated Procalcitonin Is Positively Associated With the Severity of COVID-19: A Meta-Analysis Based on 10 Cohort Studies. Med (B Aires) (2021) 57:594. doi: 10.3390/medicina57060594 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
32. Waris A, Din M, Iqbal N, Yar L, Khalid A, Nawaz M, et al.. Evaluation of Serum Procalcitonin Level as a Biomarker for Disease Severity in COVID-19 Patients. New Microbes New Infect (2021) 43:100922. doi: 10.1016/j.nmni.2021.100922 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
33. Calderon M, Li A, Bazo-Alvarez JC, Dennis J, Baker KF, Schim van der Loeff I, et al.. Evaluation of Procalcitonin-Guided Antimicrobial Stewardship in Patients Admitted to Hospital With COVID-19 Pneumonia. JAC-Antimicrobial Resist (2021) 3(3):dlab133. doi: 10.1093/jacamr/dlab133 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
34. Ticinesi A, Nouvenne A, Prati B, Guida L, Parise A, Cerundolo N, et al.. The Clinical Significance of Procalcitonin Elevation in Patients Over 75 Years Old Admitted for COVID-19 Pneumonia. Mediators Inflamm (2021) 2021:1–10. doi: 10.1155/2021/5593806 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
35. Kåsine T, Dyrhol-Riise AM, Barratt-Due A, Kildal AB, Olsen IC, Nezvalova-Henriksen K, et al.. Neutrophil Count Predicts Clinical Outcome in Hospitalized COVID-19 Patients: Results From the NOR-Solidarity Trial. J Intern Med (2021) 291(12):241–3. doi: 10.1111/joim.13377 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
36. Ulloque-Badaracco JR, Ivan Salas-Tello W, Al-kassab-Córdova A, Alarcón-Braga EA, Benites-Zapata VA, Maguiña JL, et al.. Prognostic Value of Neutrophil-to-Lymphocyte Ratio in COVID-19 Patients: A Systematic Review and Meta-Analysis. Int J Clin Pract (2021) 75:e14596. doi: 10.1111/ijcp.14596 [PubMed] [CrossRef] [Google Scholar]
37. Liu K, Yang T, Peng X, Lv S, Ye X, Zhao T, et al.. A Systematic Meta-Analysis of Immune Signatures in Patients With COVID-19. Rev Med Virol (2021) 31:e2195. doi: 10.1002/rmv.2195 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
38. Iwamura APD, Tavares da Silva MR, Hümmelgen AL, Soeiro Pereira PV, Falcai A, Grumach AS, et al.. Immunity and Inflammatory Biomarkers in COVID-19: A Systematic Review. Rev Med Virol (2021) 31:e2199. doi: 10.1002/rmv.2199 [PubMed] [CrossRef] [Google Scholar]
39. Gorog DA, Storey RF, Gurbel PA, Tantry US, Berger JS, Chan MY, et al.. Current and Novel Biomarkers of Thrombotic Risk in COVID-19: A Consensus Statement From the International COVID-19 Thrombosis Biomarkers Colloquium. Nat Rev Cardiol (2022), 1–21. doi: 10.1038/s41569-021-00665-7 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
40. Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, et al.. Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus–Infected Pneumonia in Wuhan, China. JAMA (2020) 323:1061. doi: 10.1001/jama.2020.1585 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
41. Zhou F, Yu T, Du R, Fan G, Liu Y, Liu Z, et al.. Clinical Course and Risk Factors for Mortality of Adult Inpatients With COVID-19 in Wuhan, China: A Retrospective Cohort Study. Lancet (2020) 395:1054–62. doi: 10.1016/S0140-6736(20)30566-3 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
42. Yang J, Zheng Y, Gou X, Pu K, Chen Z, Guo Q, et al.. Prevalence of Comorbidities and its Effects in Patients Infected With SARS-CoV-2: A Systematic Review and Meta-Analysis. Int J Infect Dis (2020) 94:91–5. doi: 10.1016/j.ijid.2020.03.017 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
43. Zhu Z, Wang M, Lin W, Cai Q, Zhang L, Chen D, et al.. Cardiac Biomarkers, Cardiac Injury, and Comorbidities Associated With Severe Illness and Mortality in Coronavirus Disease 2019 (COVID-19): A Systematic Review and Meta-Analysis. Immun Inflammation Dis (2021) 9(4):1071–100. doi: 10.1002/iid3.471 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
44. Schiavone M, Gobbi C, Biondi-Zoccai G, D’Ascenzo F, Palazzuoli A, Gasperetti A, et al.. Acute Coronary Syndromes and Covid-19: Exploring the Uncertainties. J Clin Med (2020) 9:1683. doi: 10.3390/jcm9061683 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
45. Nishiga M, Wang DW, Han Y, Lewis DB, Wu JC. COVID-19 and Cardiovascular Disease: From Basic Mechanisms to Clinical Perspectives. Nat Rev Cardiol (2020) 17:543–58. doi: 10.1038/s41569-020-0413-9 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
46. Jabri A, Kalra A, Kumar A, Alameh A, Adroja S, Bashir H, et al.. Incidence of Stress Cardiomyopathy During the Coronavirus Disease 2019 Pandemic. JAMA Netw Open (2020) 3:e2014780. doi: 10.1001/jamanetworkopen.2020.14780 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
47. De Marzo V, Di Biagio A, Della Bona R, Vena A, Arboscello E, Emirjona H, et al.. Prevalence and Prognostic Value of Cardiac Troponin in Elderly Patients Hospitalized for COVID-19. J Geriatr Cardiol (2021) 18:338–45. doi: 10.11909/j.issn.1671-5411.2021.05.004 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
49. Gilad V, De Marzo V, Guglielmi G, Della BR, Giovinazzo S, Pescetelli F, et al.. Cardiac Point-of-Care Ultrasound in Hospitalized Coronavirus Disease-2019 Patients: Findings and Association With Outcome. J Cardiovasc Med (2022) 23:e3–7. doi: 10.2459/JCM.0000000000001177 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
50. Metkus TS, Sokoll LJ, Barth AS, Czarny MJ, Hays AG, Lowenstein CJ, et al.. Myocardial Injury in Severe COVID-19 Compared With Non–COVID-19 Acute Respiratory Distress Syndrome. Circulation (2021) 143:553–65. doi: 10.1161/CIRCULATIONAHA.120.050543 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
51. Shi S, Qin M, Shen B, Cai Y, Liu T, Yang F, et al.. Association of Cardiac Injury With Mortality in Hospitalized Patients With COVID-19 in Wuhan, China. JAMA Cardiol (2020) 5:802. doi: 10.1001/jamacardio.2020.0950 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
52. Guan W, Ni Z, Hu Y, Liang W, Ou C, He J, et al.. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med (2020) 382:1708–20. doi: 10.1056/NEJMoa2002032 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
53. Ni W, Yang X, Liu J, Bao J, Li R, Xu Y, et al.. Acute Myocardial Injury at Hospital Admission Is Associated With All-Cause Mortality in COVID-19. J Am Coll Cardiol (2020) 76:124–5. doi: 10.1016/j.jacc.2020.05.007 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
54. Doyen D, Moceri P, Ducreux D, Dellamonica J. Myocarditis in a Patient With COVID-19: A Cause of Raised Troponin and ECG Changes. Lancet (2020) 395:1516. doi: 10.1016/S0140-6736(20)30912-0 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
55. Cipriani A, Capone F, Donato F, Molinari L, Ceccato D, Saller A, et al.. Cardiac Injury and Mortality in Patients With Coronavirus Disease 2019 (COVID-19): Insights From a Mediation Analysis. Intern Emerg Med (2021) 16:419–27. doi: 10.1007/s11739-020-02495-w [PMC free article] [PubMed] [CrossRef] [Google Scholar]
56. Lombardi CM, Carubelli V, Iorio A, Inciardi RM, Bellasi A, Canale C, et al.. Association of Troponin Levels With Mortality in Italian Patients Hospitalized With Coronavirus Disease 2019. JAMA Cardiol (2020) 5:1274. doi: 10.1001/jamacardio.2020.3538 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
57. Cameli M, Pastore MC, Mandoli GE, D’Ascenzi F, Focardi M, Biagioni G, et al.. COVID-19 and Acute Coronary Syndromes: Current Data and Future Implications. Front Cardiovasc Med (2021) 7:593496. doi: 10.3389/fcvm.2020.593496 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
58. Chieffo A, Stefanini GG, Price S, Barbato E, Tarantini G, Karam N, et al.. EAPCI Position Statement on Invasive Management of Acute Coronary Syndromes During the COVID-19 Pandemic. Eur Heart J (2020) 41:1839–51. doi: 10.1093/eurheartj/ehaa381 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
59. Chung MK, Zidar DA, Bristow MR, Cameron SJ, Chan T, Harding CV, et al.. COVID-19 and Cardiovascular Disease. Circ Res (2021) 128:1214–36. doi: 10.1161/CIRCRESAHA.121.317997 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
60. Zinellu A, Sotgia S, Fois AG, Mangoni AA, Serum CK-MB. COVID-19 Severity and Mortality: An Updated Systematic Review and Meta-Analysis With Meta-Regression. Adv Med Sci (2021) 66:304–14. doi: 10.1016/j.advms.2021.07.001 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
61. Zhan L, Liu Y, Cheng Y, Guo W, Yang J. Predictive Value of Neutrophil/Lymphocyte Ratio (NLR) on Cardiovascular Events in Patients With COVID-19. Int J Gen Med (2021) 14:3899–907. doi: 10.2147/IJGM.S317380 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
62. Yu J-S, Chen R-D, Zeng L-C, Yang H-K, Li H. Myoglobin Offers Higher Accuracy Than Other Cardiac-Specific Biomarkers for the Prognosis of COVID-19. Front Cardiovasc Med (2021) 8:686328. doi: 10.3389/fcvm.2021.686328 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
63. García de Guadiana-Romualdo L, Martínez Martínez M, Rodríguez Mulero MD, Esteban-Torrella P, Hernández Olivo M, Alcaraz García MJ, et al.. Circulating MR-proADM Levels, as an Indicator of Endothelial Dysfunction, for Early Risk Stratification of Mid-Term Mortality in COVID-19 Patients. Int J Infect Dis (2021) 111:211–8. doi: 10.1016/j.ijid.2021.08.058 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
64. Sorrentino S, Cacia M, Leo I, Polimeni A, Sabatino J, Spaccarotella CAM, et al.. B-Type Natriuretic Peptide as Biomarker of COVID-19 Disease Severity—A Meta-Analysis. J Clin Med (2020) 9:2957. doi: 10.3390/jcm9092957 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
65. Robba C, Battaglini D, Ball L, Valbusa A, Porto I, Della Bona R, et al.. Coagulative Disorders in Critically Ill COVID-19 Patients With Acute Distress Respiratory Syndrome: A Critical Review. J Clin Med (2021) 10:140. doi: 10.3390/jcm10010140 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
66. Poudel A, Poudel Y, Adhikari A, Aryal BB, Dangol D, Bajracharya T, et al.. D-Dimer as a Biomarker for Assessment of COVID-19 Prognosis: D-Dimer Levels on Admission and Its Role in Predicting Disease Outcome in Hospitalized Patients With COVID-19. PLoS One (2021) 16:e0256744. doi: 10.1371/journal.pone.0256744 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
67. Sui J, Noubouossie DF, Gandotra S, Cao L. Elevated Plasma Fibrinogen Is Associated With Excessive Inflammation and Disease Severity in COVID-19 Patients. Front Cell Infect Microbiol (2021) 11:734005. doi: 10.3389/fcimb.2021.734005 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
68. Blot M, de Maistre E, Bourredjem A, Quenot J-P, Nguyen M, Bouhemad B, et al.. Specific Features of the Coagulopathy Signature in Severe COVID-19 Pneumonia. Front Med (2021) 8:675191. doi: 10.3389/fmed.2021.675191 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
69. Fernández S, Moreno-Castaño AB, Palomo M, Martinez-Sanchez J, Torramadé-Moix S, Téllez A, et al.. Distinctive Biomarker Features in The Endotheliopathy of COVID-19 and Septic Syndromes. Shock (2021) 57(1):95–105. doi: 10.1097/SHK.0000000000001823. Online ahead of print. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
70. Billoir P, Alexandre K, Duflot T, Roger M, Miranda S, Goria O, et al.. Investigation of Coagulation Biomarkers to Assess Clinical Deterioration in SARS-CoV-2 Infection. Front Med (2021) 8:670694. doi: 10.3389/fmed.2021.670694 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
71. Joly BS, Darmon M, Dekimpe C, Dupont T, Dumas G, Yvin E, et al.. Imbalance of Von Willebrand Factor and ADAMTS13 Axis Is Rather a Biomarker of Strong Inflammation and Endothelial Damage Than a Cause of Thrombotic Process in Critically Ill COVID-19 Patients. J Thromb Haemost (2021) 19:2193–8. doi: 10.1111/jth.15445 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
72. Ambrosino P, Calcaterra I, Molino A, Moretta P, Lupoli R, Spedicato GA, et al.. Persistent Endothelial Dysfunction in Post-Acute COVID-19 Syndrome: A Case-Control Study. Biomedicines (2021) 9:957. doi: 10.3390/biomedicines9080957 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
73. Janneh AH, Kassir MF, Dwyer CJ, Chakraborty P, Pierce JS, Flume PA, et al.. Alterations of Lipid Metabolism Provide Serologic Biomarkers for the Detection of Asymptomatic Versus Symptomatic COVID-19 Patients. Sci Rep (2021) 11:14232. doi: 10.1038/s41598-021-93857-7 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
74. Razdan K, Singh K, Singh D. Vitamin D Levels and COVID-19 Susceptibility: Is There Any Correlation? Med Drug Discov (2020) 7:100051. doi: 10.1016/j.medidd.2020.100051 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
76. Battaglini D, Robba C, Fedele A, Trancǎ S, Sukkar SG, Di Pilato V, et al.. The Role of Dysbiosis in Critically Ill Patients With COVID-19 and Acute Respiratory Distress Syndrome. Front Med (2021) 8:671714. doi: 10.3389/fmed.2021.671714 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
77. Alcántara-Alonso E, Molinar-Ramos F, González-López JA, Alcántara-Alonso V, Muñoz-Pérez MA, Lozano-Nuevo JJ, et al.. High Triglyceride to HDL-Cholesterol Ratio as a Biochemical Marker of Severe Outcomes in COVID-19 Patients. Clin Nutr ESPEN (2021) 44:437–44. doi: 10.1016/j.clnesp.2021.04.020 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
78. Tepasse P-R, Vollenberg R, Fobker M, Kabar I, Schmidt H, Meier JA, et al.. Vitamin A Plasma Levels in COVID-19 Patients: A Prospective Multicenter Study and Hypothesis. Nutrients (2021) 13:2173. doi: 10.3390/nu13072173 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
79. Murai IH, Fernandes AL, Sales LP, Pinto AJ, Goessler KF, Duran CSC, et al.. Effect of a Single High Dose of Vitamin D 3 on Hospital Length of Stay in Patients With Moderate to Severe COVID-19. JAMA (2021) 325:1053. doi: 10.1001/jama.2020.26848 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
80. Zelzer S, Prüller F, Curcic P, Sloup Z, Holter M, Herrmann M, et al.. Vitamin D Metabolites and Clinical Outcome in Hospitalized COVID-19 Patients. Nutrients (2021) 13:2129. doi: 10.3390/nu13072129 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
81. Beltrão FE de L, Beltrão DC de A, Carvalhal G, Beltrão FE de L, Brito A da S, da Capistrano KHR, et al.. Thyroid Hormone Levels During Hospital Admission Inform Disease Severity and Mortality in COVID-19 Patients. Thyroid (2021) 31:1639–49. doi: 10.1089/thy.2021.0225 [PubMed] [CrossRef] [Google Scholar]
82. Battaglini D, Brunetti I, Anania P, Fiaschi P, Zona G, Ball L, et al.. Neurological Manifestations of Severe SARS-CoV-2 Infection: Potential Mechanisms and Implications of Individualized Mechanical Ventilation Settings. Front Neurol (2020) 11:845. doi: 10.3389/fneur.2020.00845 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
83. Huth SF, Cho S-M, Robba C, Highton D, Battaglini D, Bellapart J, et al.. Neurological Manifestations of Coronavirus Disease 2019: A Comprehensive Review and Meta-Analysis of the First 6 Months of Pandemic Reporting. Front Neurol (2021) 12:664599. doi: 10.3389/fneur.2021.664599 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
84. Battaglini D, Santori G, Chandraptham K, Iannuzzi F, Bastianello M, Tarantino F, et al.. Neurological Complications and Noninvasive Multimodal Neuromonitoring in Critically Ill Mechanically Ventilated COVID-19 Patients. Front Neurol (2020) 11:602114. doi: 10.3389/fneur.2020.602114 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
87. Reichard RR, Kashani KB, Boire NA, Constantopoulos E, Guo Y, Lucchinetti CF. Neuropathology of COVID-19: A Spectrum of Vascular and Acute Disseminated Encephalomyelitis (ADEM)-Like Pathology. Acta Neuropathol (2020) 140:1–6. doi: 10.1007/s00401-020-02166-2 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
88. Ameres M, Brandstetter S, Toncheva AA, Kabesch M, Leppert D, Kuhle J, et al.. Association of Neuronal Injury Blood Marker Neurofilament Light Chain With Mild-to-Moderate COVID-19. J Neurol (2020) 267:3476–8. doi: 10.1007/s00415-020-10050-y [PMC free article] [PubMed] [CrossRef] [Google Scholar]
89. Kanberg N, Ashton NJ, Andersson L-M, Yilmaz A, Lindh M, Nilsson S, et al.. Neurochemical Evidence of Astrocytic and Neuronal Injury Commonly Found in COVID-19. Neurology (2020) 95:e1754–9. doi: 10.1212/WNL.0000000000010111 [PubMed] [CrossRef] [Google Scholar]
90. Paterson RW, Benjamin LA, Mehta PR, Brown RL, Athauda D, Ashton NJ, et al.. Serum and Cerebrospinal Fluid Biomarker Profiles in Acute SARS-CoV-2-Associated Neurological Syndromes. Brain Commun (2021) 3:fcab099. doi: 10.1093/braincomms/fcab099 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
91. Kanberg N, Simrén J, Edén A, Andersson L-M, Nilsson S, Ashton NJ, et al.. Neurochemical Signs of Astrocytic and Neuronal Injury in Acute COVID-19 Normalizes During Long-Term Follow-Up. EBioMedicine (2021) 70:103512. doi: 10.1016/j.ebiom.2021.103512 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
92. Esenwa C, Cheng NT, Luna J, Willey J, Boehme AK, Kirchoff-Torres K, et al.. Biomarkers of Coagulation and Inflammation in COVID-19–Associated Ischemic Stroke. Stroke (2021) 52:e706–709. doi: 10.1161/STROKEAHA.121.035045 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
93. Goyal N, Sodani AK, Jain R, Ram H. Do Elevated Levels of Inflammatory Biomarkers Predict the Risk of Occurrence of Ischemic Stroke in SARS-CoV2 ?: An Observational Study. J Stroke Cerebrovasc Dis (2021) 30:106063. doi: 10.1016/j.jstrokecerebrovasdis.2021.106063 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
94. Yassin A, Ghzawi A, Al-Mistarehi A-H, El-Salem K, Y Benmelouka A, M Sherif A, et al.. Mortality Rate and Biomarker Expression Within COVID-19 Patients Who Develop Acute Ischemic Stroke: A Systematic Review and Meta-Analysis. Futur Sci OA (2021) 7:FSO713. doi: 10.2144/fsoa-2021-0036 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
95. Mete E, Sabirli R, Goren T, Turkcuer I, Kurt O, Koseler A. Association Between S100b Levels and COVID-19 Pneumonia: A Case Control Study. Vivo (Brooklyn) (2021) 35:2923–8. doi: 10.21873/invivo.12583 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
96. Benjamin LA, Paterson RW, Moll R, Pericleous C, Brown R, Mehta PR, et al.. Antiphospholipid Antibodies and Neurological Manifestations in Acute COVID-19: A Single-Centre Cross-Sectional Study. EClinicalMedicine (2021) 39:101070. doi: 10.1016/j.eclinm.2021.101070 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
97. Faour WH, Choaib A, Issa E, El Choueiry F, Shbaklo K, Alhajj M, et al.. Mechanisms of COVID-19-Induced Kidney Injury and Current Pharmacotherapies. Inflammation Res (2022) 71:39–56. doi: 10.1007/s00011-021-01520-8 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
98. Legrand M, Bell S, Forni L, Joannidis M, Koyner JL, Liu K, et al.. Pathophysiology of COVID-19-Associated Acute Kidney Injury. Nat Rev Nephrol (2021) 17:751–64. doi: 10.1038/s41581-021-00452-0 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
99. Liu Y, Xia P, Cao W, Liu Z, Ma J, Zheng K, et al.. Divergence Between Serum Creatine and Cystatin C in Estimating Glomerular Filtration Rate of Critically Ill COVID-19 Patients. Ren Fail (2021) 43:1104–14. doi: 10.1080/0886022X.2021.1948428 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
100. Yasar E, Ozger HS, Yeter HH, Yildirim C, Osmanov Z, Cetin TE, et al.. Could Urinary Kidney Injury Molecule-1 be a Good Marker in Subclinical Acute Kidney Injury in Mild to Moderate COVID-19 Infection? Int Urol Nephrol (2021) 54(3):627–36. doi: 10.1007/s11255-021-02937-0 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
101. Tantry US, Bliden KP, Cho A, Walia N, Dahlen JR, Ens G, et al.. First Experience Addressing the Prognostic Utility of Novel Urinary Biomarkers in Patients With COVID-19. Open Forum Infect Dis (2021) 8:ofab274. doi: 10.1093/ofid/ofab274 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
102. Fukao Y, Nagasawa H, Nihei Y, Hiki M, Naito T, Kihara M, et al.. COVID-19-Induced Acute Renal Tubular Injury Associated With Elevation of Serum Inflammatory Cytokine. Clin Exp Nephrol (2021) 25:1240–6. doi: 10.1007/s10157-021-02101-z [PMC free article] [PubMed] [CrossRef] [Google Scholar]
103. Gradin A, Andersson H, Luther T, Anderberg SB, Rubertsson S, Lipcsey M, et al.. Urinary Cytokines Correlate With Acute Kidney Injury in Critically Ill COVID-19 Patients. Cytokine (2021) 146:155589. doi: 10.1016/j.cyto.2021.155589 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
104. Wang RR, He M, Kang Y. A Risk Score Based on Procalcitonin for Predicting Acute Kidney Injury in COVID-19 Patients. J Clin Lab Anal (2021) 35:e23805. doi: 10.1002/jcla.23805 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
105. Alfano G, Ferrari A, Fontana F, Mori G, Ligabue G, Giovanella S, et al.. Twenty-Four-Hour Serum Creatinine Variation Is Associated With Poor Outcome in the Novel Coronavirus Disease 2019 (COVID-19) Patients. Kidney Res Clin Pract (2021) 40:231–40. doi: 10.23876/j.krcp.20.177 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
106. Morell-Garcia D, Ramos-Chavarino D, Bauça JM, Argente del Castillo P, Ballesteros-Vizoso MA, García de Guadiana-Romualdo L, et al.. Urine Biomarkers for the Prediction of Mortality in COVID-19 Hospitalized Patients. Sci Rep (2021) 11:11134. doi: 10.1038/s41598-021-90610-y [PMC free article] [PubMed] [CrossRef] [Google Scholar]
107. Wang K, Qu M, Ding L, Shi X, Wang C, Cheng S, et al.. Liver and Kidney Function Biomarkers, Blood Cell Traits and Risk of Severe COVID-19: A Mendelian Randomization Study. Front Genet (2021) 12:647303. doi: 10.3389/fgene.2021.647303 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
108. Kumar- MP, Mishra S, Jha DK, Shukla J, Choudhury A, Mohindra R, et al.. Coronavirus Disease (COVID-19) and the Liver: A Comprehensive Systematic Review and Meta-Analysis. Hepatol Int (2020) 14:711–22. doi: 10.1007/s12072-020-10071-9 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
109. Romero-Cristóbal M, Clemente-Sánchez A, Piñeiro P, Cedeño J, Rayón L, del Río J, et al.. Possible Unrecognised Liver Injury Is Associated With Mortality in Critically Ill COVID-19 Patients. Ther Adv Gastroenterol (2021) 14:175628482110234. doi: 10.1177/17562848211023410 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
110. Caterino M, Costanzo M, Fedele R, Cevenini A, Gelzo M, Di Minno A, et al.. The Serum Metabolome of Moderate and Severe COVID-19 Patients Reflects Possible Liver Alterations Involving Carbon and Nitrogen Metabolism. Int J Mol Sci (2021) 22:9548. doi: 10.3390/ijms22179548 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
111. Lasso G, Khan S, Allen SA, Mariano M, Florez C, Orner EP, et al.. Longitudinally Monitored Immune Biomarkers Predict the Timing of COVID-19 Outcomes. PLoS Comput Biol (2022) 18:e1009778. doi: 10.1371/journal.pcbi.1009778 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
112. Smilowitz NR, Nguy V, Aphinyanaphongs Y, Newman JD, Xia Y, Reynolds HR, et al.. Multiple Biomarker Approach to Risk Stratification in COVID-19. Circulation (2021) 143:1338–40. doi: 10.1161/CIRCULATIONAHA.120.053311 [PMC free article] [PubMed] [CrossRef] [Google Scholar]