Age-related changes in plasma biomarkers and their association with mortality in COVID-19
Graphical Abstract Sample Work
COVID-19 mortality increases with age, with ≥70-year-olds accounting for ∼89% of deaths in the Netherlands. While ageing-related comorbidities explain part of the risk, COVID-19 triggers inflammation, cytokine release, coagulation disturbances, and endotheliopathy. Dexamethasone and tocilizumab show limited benefits in older patients. This study explores age-related changes in key pathophysiological pathways and their association with 30-day mortality in noncritically ill hospitalised patients.
What this abstract has to show in one frame
This graphical abstract condenses an observational study of how plasma biomarkers change with age in patients hospitalised with COVID-19, and whether those changes help explain why older patients are more likely to die. The study combines three ideas that are hard to hold together in prose: age stratification, a large panel of 43 biomarkers grouped into biological domains, and a mediation analysis linking age, biomarkers and 30-day mortality. The figure has to present all three without asking the reader to parse a results table.
It does this with a simple two-column frame headed "Methods" and "Results", a structure familiar to clinicians and to editors of infectious disease and critical care journals.
The clinical puzzle: why age drives COVID-19 deaths
COVID-19 mortality rises steeply with age. The study's background notes that people aged 70 or over accounted for roughly 89% of COVID-19 deaths in the Netherlands. Age-related comorbidities explain part of that risk, but SARS-CoV-2 infection also provokes systemic inflammation, cytokine release, coagulation disturbances and damage to the vascular endothelium (endotheliopathy). Treatments aimed at the host response, such as dexamethasone and tocilizumab, appear to offer limited benefit in older patients.
If ageing alters these host-response pathways, measuring them could show which mechanisms connect age to death and point towards targets for older patients. The study focuses on non-critically ill patients admitted to general wards, a group in whom early differences may be more informative than in those already in intensive care.
Left column: cohort, validation and biomarker domains
The Methods column is divided into three stacked boxes, read from top to bottom:
- Derivation cohort. 464 COVID-19 patients admitted to the general ward, stratified into four age bands: under 50, 50 to under 60, 60 to under 70, and 70 years or older. Each band is shown with paired male and female figures surrounded by virus particles.
- Validation. Findings were checked in an intensive care unit cohort (n=157) and an external cohort (n=196), which signals to readers that the associations were not confined to one dataset.
- Biomarker panel. 43 plasma biomarkers reflecting key pathophysiological domains, drawn as four labelled groups around icons: coagulation and endothelium (a blood vessel), systemic inflammation and organ damage (lungs), cytokines and chemokines (immune cells).
Right column: an alveolus that carries the results
The Results column uses a single anatomical illustration rather than charts. A coughing patient with inflamed lungs is linked by dashed lines to an enlarged view of an alveolus and adjacent capillary, and an icon of an older couple anchors the comparison to age. Arrows placed around this scene summarise the direction of change in older patients:
| Domain | Change shown with older age |
|---|---|
| Endothelial cell and coagulation activation markers | Increased |
| Systemic inflammation and organ damage markers | Increased |
| Cytokines | Increased or unchanged |
| Chemokines | Increased, unchanged or decreased |
The mixed symbols for cytokines and chemokines are an honest touch: rather than implying that every inflammatory marker rises with age, the figure shows that responses differed within those groups.
Showing a mediation analysis without a path diagram
The bottom strip of the Results column handles the most technical element. An older couple on the left is connected by three arrows to three membrane-bound molecules, each marked with an upward arrow: soluble TNF receptor 1 (sTNF-R1), soluble TREM-1 and soluble thrombomodulin. These then converge on an upward arrow beside a cross, the conventional symbol for death. Read left to right, the strip states that these three markers lay on the pathway between older age and higher mortality.
Mediation analysis estimates how much of an exposure's association with an outcome runs through an intermediate variable. In an observational cohort, such results describe statistical pathways and do not by themselves prove causation, which is why the figure presents them as arrows rather than as claims of mechanism. Studies of this design are normally reported against the STROBE checklist, and a graphical abstract should not state more than the paper does.
Points to borrow for a biomarker cohort figure
For authors with a similar dataset, the sample suggests three transferable choices: group a large biomarker panel into a handful of named domains before drawing; put sample sizes for derivation and validation cohorts in the figure itself; and reserve one compact strip for the main inferential result. It also avoids clutter by omitting p-values and confidence intervals, which belong in the paper.
This page presents a writing and design sample and is not clinical guidance. Pubrica prepares figures like this through its graphical abstract service, and mediation or stratified survival analyses of the kind summarised here fall under biostatistics and statistical programming.
Questions about this sample
Why are patients split into four age bands rather than treated as a continuous age variable?
The figure shows four bands (under 50, 50–59, 60–69 and 70 or older) because bands make age-related trends easy to see visually. They let readers compare biomarker patterns across clearly defined groups. The underlying paper may also analyse age continuously, but for a graphical abstract the stratified presentation communicates the gradient more directly.
Which biomarkers does the abstract identify as mediators of mortality?
The mediation strip names three: soluble TNF receptor 1, soluble TREM-1 and soluble thrombomodulin. Each is shown rising with age and linked to higher 30-day mortality. Together they represent inflammatory signalling, innate immune activation and endothelial injury, which matches the domains the study set out to examine.
What do the validation cohorts add to the figure?
Showing an intensive care cohort of 157 patients and an external cohort of 196 tells readers that the age-related patterns found in the 464 general ward patients were tested elsewhere. Including these numbers in the Methods column gives an immediate sense of the study's robustness without requiring any additional charts.
Give yourself the academic edge today
Each order includes
- On-time delivery or your money back
- A fully qualified writer in your subject
- In-depth proofreading by our Quality Control Team
- 100% confidentiality, the work is never re-sold or published
- Standard 7-day amendment period
- A paper written to the standard ordered
- A detailed plagiarism report
- A comprehensive quality report
Enquire Now
