Thread by @ZdenekVrozina

A new narrative review in Communications Medicine sums up where the field stands on long COVID.
Not as one single, uniform diagnosis, but as a complex, multisystem condition after SARS2 infection🧵

Its value is in the synthesis. It brings together immunology, neurology, vascular biology, metabolism, and clinical medicine into one framework.

The review covers prevalence, pathophysiology, biomarkers, treatment strategies, and future research directions. It is a broad interpretation of the current literature.

The first important shift.
A strong emphasis on viral persistence. The paper takes seriously the possibility that SARS-CoV-2 remnants or viral antigens may persist in tissues.

If viral components remain present, they could keep stimulating the immune system. That moves long COVID away from a post infectious echo to ongoing biological activity.

The authors also discuss reactivation of latent viruses, such as EBV, HHV-6, or CMV. The issue may be a longer-lasting virological and immune dysregulation, not just damage left behind by the acute infection.

The second major theme.
The link between microclots, the endothelium, exercise intolerance. One striking feature of the review is how central endothelial injury, platelet activation, and coagulation abnormalities are.

The authors discuss abnormal fibrin or microclot deposits, glycocalyx damage, increased platelet activation, and impaired microcirculation as possible contributors to tissue hypoxia.

They connect these vascular and coagulation mechanisms with fatigue, PEM, reduced exercise tolerance, multisystem symptoms. In simple words, these are not symptoms without findings, but findings with concrete biology behind them.

The third important point.
PEM is not deconditioning. It is described as a worsening of symptoms after physical, mental, cognitive, or emotional exertion. It can appear with a delay and last for days, weeks, or sometimes longer.

The paper discusses muscle, mitochondrial, microvascular, and inflammatory abnormalities. It also warns that graded exertion may harm some patients if it pushes them beyond their PEM threshold.

The fourth shift.
Long COVID is framed as a set of subtypes, not one disease. The authors do not expect one universal biomarker that cleanly separates all patients from controls.

Instead, they point toward subtyping based on combinations of symptoms, biomarkers, and mainly mechanisms - immunological, vascular, neurological, metabolic, viral-persistence-related, or overlapping forms.

That matters for clinical trials. If biologically different patients are placed into one broad study group, a treatment may look ineffective overall even if it helps one specific subgroup.

The fifth notable point.
The neurological mechanisms are becoming more biologically specific. The review summarizes evidence for neuroinflammation, blood-brain barrier disruption, and possible viral components in nervous tissue.

It also discusses the skull–meninges–brain axis and vascular immune mechanisms in the CNS. That gives brain fog a concrete biological frame.

The sixth point. The review is cautious about treatments. It mentions IVIg, low-dose naltrexone, BC007, apheresis, antivirals, and other approaches, but treats most of them as preliminary.

The review says there are several biologically plausible therapeutic directions that need proper testing.

Without better patient stratification, clinical trials will remain hard to interpret. Long COVID is probably not driven by one mechanism, so one intervention is unlikely to work across the entire patient population.

So long COVID is presented here as a biologically measurable condition with multiple mechanistic subtypes.
We still lack validated diagnostics, robust biomarkers, and treatments proven for specific subtypes.

Faghy at al., Current status and future perspectives on the mechanistic and pathophysiological understanding of long COVID. nature.com/articles/s4385…