As atmospheric carbon dioxide has risen over two decades, a marker in human blood appears to have shifted alongside it — a subtle change researchers say is worth watching but not a cause for alarm.

Scientists at the Kids Research Institute Australia, with Curtin University and the Australian National University, analysed US NHANES health-survey data from about 7,000 people collected every two years between 1999 and 2020. Blood bicarbonate rose roughly 7% over that period while calcium and phosphorus edged down. The work appeared in Air Quality, Atmosphere and Health.

What bicarbonate does

Blood pH is held within an extremely narrow range — deviation of a few tenths in either direction is life-threatening — and the bicarbonate buffer system is the main mechanism maintaining it.

The chemistry is a balance: carbon dioxide dissolved in blood forms carbonic acid, which dissociates into bicarbonate and hydrogen ions. Raising CO2 pushes the reaction toward acid, and the body compensates by retaining bicarbonate through the kidneys.

That relationship is textbook physiology, and it is the basis of the proposed link: as environmental CO2 rises, the body retains more bicarbonate to keep pH stable.

The correlation

Atmospheric CO2 rose from about 369 parts per million in 2000 to over 420 today, and the blood changes tracked that rise.

“What we’re seeing is a gradual shift in blood chemistry that mirrors the rise in atmospheric carbon dioxide,” the researchers said. Their modelling suggests bicarbonate could approach the upper end of the healthy range within about 50 years, a concern they flag especially for children, who face a lifetime of exposure.

Why the magnitude invites scepticism

The physiological argument deserves scrutiny, because the numbers involved are small relative to normal variation.

Atmospheric CO2 at 420 ppm is 0.042% of air. Alveolar CO2 in the lungs is around 5% — more than a hundred times higher — because the body continuously produces CO2 through metabolism at rates that dwarf ambient concentrations.

Against that, a change of 50 ppm in inspired air is a very small perturbation, and the respiratory system adjusts CO2 elimination continuously by varying breathing. Whether a shift of that size could produce a measurable systemic change is the question the study raises rather than answers.

The alternative explanations

Twenty years is long enough for a great deal besides atmospheric composition to change, and any of it could produce a trend in a population marker.

Laboratory methods change over decades. NHANES has run for many years, and analytical platforms, reagents and calibration standards are periodically updated — systematic drift in measurement is a well-recognised hazard in long-running surveys.

The population changed too. Rates of obesity, chronic kidney disease and diabetes rose substantially over the same period, and all affect acid-base balance. Medication use also shifted, and several widely prescribed drug classes influence bicarbonate.

Diet changed, and dietary acid load is a recognised determinant of bicarbonate levels.

Any of these would produce a trend correlating with atmospheric CO2 for the simple reason that both are trending over the same period. Two variables rising together across twenty years is a weak basis for inferring one causes the other.

The researchers’ own framing

To their credit, they are explicit: this is a correlation, not proof of cause. The study shows two trends moving together across a large population, not that rising CO2 directly changes blood chemistry or harms health.

They describe it as preliminary epidemiological evidence warranting continued monitoring — an intriguing signal rather than a settled finding, which is a fair characterisation.

What would strengthen or refute it

Several tests are available. A dose-response gradient would help — people in higher-CO2 environments such as poorly ventilated buildings or dense urban areas should show larger shifts if the mechanism is real.

Controlled exposure studies could measure bicarbonate in volunteers breathing air at different CO2 concentrations, directly testing whether the physiological effect exists at these levels.

And replication in other national datasets with independent laboratory methods would address the measurement-drift concern, which is currently the most plausible competing explanation.

Indoor CO2 is the more plausible exposure

If ambient carbon dioxide affects human physiology at all, the relevant exposure is probably not outdoor air — and that is where this line of research has firmer ground.

Concentrations indoors routinely exceed outdoor levels by a large margin, because people exhale CO2 and buildings are ventilated imperfectly. Poorly ventilated meeting rooms, classrooms and bedrooms regularly reach 1,000 to 2,000 parts per million, several times the outdoor concentration, and levels above 3,000 have been measured in crowded spaces.

There is a separate literature on cognitive effects at those indoor concentrations, with some studies reporting impaired decision-making and others failing to replicate it — the evidence is genuinely mixed.

The relevance here is that people in developed countries spend the overwhelming majority of their time indoors, so their actual CO2 exposure is dominated by building ventilation rather than atmospheric concentration. Any real physiological effect would be expected to show up first as a difference between well- and poorly-ventilated environments — a comparison that is straightforward to make, and one that would test the hypothesis far more directly than a population trend.

Until then it is a hypothesis generated from an observation, presented as such. Research news, not medical advice.