Whole-body imaging of healthy volunteers produced more biopsies. Most of them found nothing

The promise of whole-body imaging in a checkup is that it looks everywhere at once. The difficulty is the same sentence read differently: it finds everything at once, including a great deal that was never going to matter. What that costs has been measured directly.

A cohort that scanned half its participants#

The Study of Health in Pomerania examined 6,753 participants in northern Germany between 2008 and 2012. All had laboratory testing; 3,371 — just under half — also had whole-body magnetic resonance imaging. Electronic biopsy reports covering 2002 to 2017 were then linked to participants, so biopsy activity could be compared in the two years before their examination and the two years after.

In total 8,208 incidental findings were disclosed to 4,707 participants: 6,839 laboratory values outside reference limits and 1,369 from the MRI. Of the cohort, 1,200 participants — 17.8 percent — had at least one biopsy on record, across 2,271 biopsy reports.

Which kind of biopsy went up#

A rise in biopsies is not by itself a criticism. It would be exactly what a working screening programme produces. The question is what the extra biopsies found, and the study answers it by splitting the increase.

Malignant results also rose, with a rate ratio of 1.74 and a confidence interval from 1.27 to 2.42. Both things are true: the scanning found some cancers, and it produced a larger absolute increase in biopsies that found no cancer. The authors' summary was blunt — most biopsies resulted in no findings and few malignancies were diagnosed, indicating potential overtesting and overdiagnosis — and their recommendation was for a more restrictive policy on disclosing research incidental findings.

Why this matters to a Korean checkup#

This was a research cohort in Germany, not a Korean commercial checkup, and the difference cuts in an uncomfortable direction. Research participants were told about findings by a study team operating under a disclosure policy. A paying customer of a comprehensive package receives a report they have bought, from a hospital, with the expectation that everything found will be acted on.

Korea has already run the largest natural experiment in the world on what happens when an ultrasound is added to a checkup in a healthy population. Between 1999 and 2008 the incidence of thyroid cancer rose 6.4-fold, and 94.4 percent of the increase was in tumours under 20 mm, detected mainly by screening, while mortality did not move. That episode is the reason the incidental-findings literature is read differently in Seoul than elsewhere.

What the study cannot tell a patient#

It is a before-and-after comparison within a cohort, not a randomised trial of scanning, and participation itself — being examined, being given attention — could raise biopsy rates without any imaging at all. The design partly addresses this by comparing participants who received disclosures against those who did not, and it is that comparison, not the overall rise, which carries the finding.

It also does not follow participants long enough to say whether the extra cancers found were ones that would have caused harm. That is the question overdiagnosis turns on, and answering it requires decades rather than the four-year window here.

What it does establish is that disclosing incidental findings from whole-body imaging in people who came in well generates additional biopsies, and that the additional biopsies mostly find nothing. Anybody buying a package because it looks everywhere is buying that outcome as part of it.