A blood test to predict imminent death? Would you want to take it?


Here are some findings that could scare you to death: In a study published this week, Finnish and Estonian researchers report that they have identified specific levels of four chemicals circulating in the blood that offer a reliable signal that death is near. The four harbingers of death can be readily detected in a blood sample, and are even predictive when seen in apparently healthy people, their new study shows.

It’s not just a life insurance saleman’s dream. The study, released this week in the journal PLOS Medicine, suggests that several potentially deadly conditions -- cancer, cardiovascular disease and a welter of non-vascular causes of death -- may share signs, and even origins, that have been hidden in plain sight. If readily detectable physiological clues--called “biomarkers”--could give warning of many dangerous conditions at once, a single blood test might provide a person early warning of a deadly threat, while it could still be averted (or at least delayed).

That said, a blood test to predict death is far from ready for prime time. This is an early effort to glean better ways to screen for and diagnose diseases, not to give notice that the Grim Reaper is stalking you.


For now, however, the “death biomarkers” might make you think twice before giving up a vial of your blood. The four horseman of an individual’s apocalypse were narrowed down from a list of 106 candidate biomarkers--all lipids, proteins and metabolites circulating in the blood. They are: alpha-1-acid glycoprotein, albumin, very low-density lipoprotein (VLDL) particle size, and citrate.

How did the researchers glean the possible significance of these four? They ran the carefully collected blood samples of 9,482 Estonians between the age of 18 and 101 through a scan that used nuclear magnetic resonance spectroscopy, to make measurements of the 106 biomarker candidates in each. Over a median follow-up period of just over five years, 508 of the randomly chosen Estonian subjects died of various causes. The study’s authors compared the biomarker levels of both groups in an effort to identify those that were more common in the dead and less common in the living.

The four biomarkers stood out as unusually common among the dead, and less common in those that survived the five years of follow-up. The researchers created an index of the four measures. They found that compared to a person whose index fell in the bottom 20% of the observed range, the individual whose biomarker index was in the top 20% was 19 times more likely to die in the five years after his or her blood sample was drawn.

Even when the researchers stripped out all participants who had diabetes, cancer or cardiovascular disease, they found that the biomarkers predicted death over the five-year period in apparently healthy people as well. The researchers then repeated their test of biomarkers on a separate population--8,444 Finnish men and women between 24 and 74. The biomarkers were equally predictive of death in this “validation group.”

All of which makes you wonder: Would you want to know?