Review of 223 studies weighs blood vessel aging signs against death risk

A team of researchers has pooled the results of 223 earlier studies to ask a simple question: of the many ways scientists try to measure how “old” a person’s blood vessels are, which ones actually track with dying sooner? Their answer, posted on November 12, 2026, as a preprint (a paper not yet checked by outside experts), points to three measures, with links of very different strength.¹

The three were a coronary artery calcium score above zero, a low ankle-brachial index and shorter leucocyte telomere length.¹ In plain words: calcium deposits showing up in the arteries that feed the heart, blood pressure at the ankle that is low compared with blood pressure at the arm, and shorter protective caps on the ends of chromosomes in white blood cells. Each was tied to a higher risk of dying from any cause and of dying from heart and blood vessel disease.¹

What the researchers did

The team searched three large medical databases, CENTRAL, MEDLINE and EMBASE, plus other sources, for prospective studies, meaning studies that measured people first and then followed them forward in time. Each study had to follow people for at least six months.¹ They registered their plan in advance in PROSPERO, a public registry for reviews, under the number CRD42022377292.¹

In total they included 223 studies covering 1,325,186 participants.¹ They grouped the measures into 16 categories of blood vessel aging markers and looked at five outcomes: death from any cause, death from cardiovascular disease, stroke, heart attack (myocardial infarction), and a combined group of major heart and blood vessel events.¹

To combine the studies they used a random-effects meta-analysis, a statistical method that merges results while allowing for the fact that the original studies differ. They also ran trial sequential analysis, which checks whether enough data have piled up to trust a pooled result, along with subgroup and sensitivity analyses that test whether the findings hold when the data are sliced or handled differently.¹

What the numbers show

The results are reported as hazard ratios. A hazard ratio compares the rate of an event, such as death, between two groups; a value of 1.0 means no difference, and 1.5 means roughly a 50% higher rate.² Here is what the review reported for the three main markers:¹

The ranges are 95% confidence intervals, the span in which the true value most likely sits. None of them crosses 1.0, which is why the authors describe all three as associated with both kinds of death.¹

The sizes differ a lot. A low ankle-to-arm ratio went with about 2.3 times the rate of cardiovascular death. Shorter telomeres, a measure often discussed in the longevity world, went with a rate only about 11% to 13% higher.¹ Artery calcium showed a stronger link with death from any cause than with cardiovascular death specifically.¹

The authors reported that trial sequential analyses supported these associations and that sensitivity analyses did not materially change the main findings.¹ For stroke, heart attack and the combined event group, they said the picture varied by marker.¹ The abstract does not give those individual results.

What it does not show yet

All the pooled studies observed people rather than testing a treatment. That means the results show which measures travel with higher risk, not that changing a measure would change anyone’s risk. The authors are direct about the limits:

Heterogeneity and residual confounding limit clinical interpretation. Their combined predictive value and usefulness for guiding interventions require prospective validation.¹

“Heterogeneity” means the original studies differed from each other in who they studied and how. “Residual confounding” means other factors, such as existing illness or lifestyle, may explain part of the link even after adjustment. The authors call their findings “candidate biomarkers for further vascular aging research,” not ready-made clinical tools.¹

Several things also remain unclear from the abstract. It does not say how the 16 marker categories compared beyond the three highlighted, whether using the three markers together predicts risk better than using any one, or how well they perform against standard heart risk checks. The authors say that combined predictive value still needs testing.¹ And because this is a preprint, its methods and numbers have not been checked by peer reviewers.¹

Who is behind it

The preprint lists 17 authors, beginning with T. Wang and ending with L. Jiao.¹ The text available to Biohack does not state the authors’ institutions, who funded the work, or any conflicts of interest. Readers should check the full paper, and any peer-reviewed version, for that information.

How it fits the field

Heart and blood vessel diseases are the leading cause of death worldwide, according to the World Health Organization figures cited by the diagnostics company Biomedica.³ Atherosclerosis, a buildup of fat, fibrous tissue and calcium in large arteries, narrows vessels and drives many of those deaths.³ That is part of why measures like artery calcium attract interest as signs of how blood vessels are aging.

The search for a simple measure that predicts death has a long history in aging research. In a 2019 overview, writer and researcher Sarah Constantin ranked dozens of markers by their links to death from any cause in general-population studies.² In her list, markers of heart strain such as troponin T (3.5) and NT-proBNP (2.3) and a simple frailty score (2.9) showed larger hazard ratios, while grip strength came in at 1.7 and the Horvath “epigenetic clock” at 1.4.²

Those numbers cannot be lined up directly against the new review. Constantin compared the top and bottom groups of each marker and averaged studies that were not strictly comparable, and she said her overview was not a true meta-analysis.² The new review used formal pooling with registered methods, but compared different groupings, such as any artery calcium versus none.¹ Constantin also raised a caution that applies here: when a marker is used as a stand-in without knowing the mechanism, it is hard to be sure which way cause and effect run.²

Mixed evidence is common for markers meant to predict death. A separate systematic review of blood markers in people on heart-lung support machines for severe heart failure, for example, described the evidence for their predictive value as inconsistent.⁴ The new review’s authors likewise opened by noting that the evidence behind candidate vascular aging markers varies.¹

What to watch next

• Peer review. Whether the paper is accepted by a journal, and whether reviewers ask for changes to the numbers or methods.¹
• The full results. How the other marker categories performed, and the marker-by-marker results for stroke, heart attack and combined events, which the abstract only summarizes as varying.¹
• Testing in new groups of people. The authors say whether these markers add predictive value together, and whether they can help guide treatment, needs to be tested in future forward-looking studies.¹

Biohack will update this thread if the review is published or if follow-up studies test these markers directly.