Two fitness numbers predict surgical risk best

Cardiopulmonary exercise testing is routinely used to decide who is fit enough for major lung surgery, but the thresholds clinicians rely on have largely come from expert consensus and scientific statements rather than graded evidence. That leaves an awkward gap: a test that can influence whether a patient is offered an operation has not had its individual variables ranked by how well they actually predict harm. Exercise testing produces many numbers at once — peak oxygen uptake, ventilatory efficiency, anaerobic threshold, peak ventilation — and it has not been clear which of them carry the signal. This review set out to grade that evidence formally.

The authors searched electronic databases through 20 December 2025 for cohort studies linking cardiopulmonary exercise testing variables to outcomes after lung cancer surgery. Of 3080 records screened, 47 studies were included, covering 11,356 surgeries and 13 distinct exercise-test variables. Study quality was rated with the Newcastle-Ottawa Scale, heterogeneity assessed with the Q-test and I² statistic, and effects pooled using fixed or random effects models as appropriate. Publication bias was examined with funnel plots and Egger's test, with trim-and-fill adjustment where needed, and the certainty of evidence for each variable was graded using the GRADE approach.

Lower peak oxygen uptake was associated with a 2.04 to 3.02 times higher risk of perioperative complications, whether expressed against a cutoff of 15 mL/kg/min or as a percentage of predicted. A higher ratio of minute ventilation to carbon dioxide production — the ventilatory efficiency slope — was associated with a 2.44-fold increased risk, with a 95% confidence interval of 1.82 to 3.27 and a cutoff of 35. Both findings were rated high-certainty. Evidence linking peak oxygen uptake in litres per minute and the ventilatory slope to cancer-specific and all-cause mortality came from only limited studies. Variables including oxygen uptake at the anaerobic threshold and peak ventilation remained low-certainty.

This puts graded evidence underneath thresholds that were previously supported mainly by consensus, and it narrows the field: of 13 candidate variables, only two survived high-certainty grading for perioperative risk. It broadly endorses the widely used 15 mL/kg/min peak oxygen uptake cutoff while adding the ventilatory efficiency slope of 35 as an equally well-supported marker. It also complicates the picture for the many other variables that appear on a standard test report, which the data do not yet justify weighting heavily. The role of exercise testing in predicting cancer-specific mortality remains an open question the authors explicitly flag for further study.

Measured cardiorespiratory fitness is not just a performance statistic; here it tracks the odds of complications after a major operation, and it does so more reliably than most of the other numbers a lab test produces. For a general reader, this is a reason to think of peak oxygen uptake as a clinically meaningful index of physiological reserve rather than an athletic vanity metric. Note the direction of the evidence, though: these are observational cohorts showing that lower fitness marks higher risk, not trials showing that improving fitness before surgery changes what happens. How much of that risk is modifiable is a separate question this review did not address.