Cardiovascular assessment in living kidney donors: the control group matters

Abstract Background Research on long-term cardiovascular risk among living kidney donors has yielded inconsistent findings, and the selection of control groups might have influenced these results. The study objective was to compare cardiovascular characteristics and cardiac function in living donors with two distinct control groups. This would identify the appropriate control group to be used in long-term studies of cardiovascular disease in donors. Methods This cross-sectional study analysed the predonation characteristics of living kidney donors (“donor candidates”) and compared them to two control groups: healthy individuals who underwent formal donor work-up (“donor-eligible controls”) and an age- and sex-matched control group randomly selected from Norway’s National Population Register (“general-population controls”). Key assessments included comorbidities, blood pressure and echocardiographic parameters. Results Fifty-six donor candidates, 53 donor-eligible controls and 38 general-population controls were included in the study. Donor candidates and donor-eligible controls were similar throughout, differing only in resting heart rate. Both differed from the general-population controls, who had higher systolic blood pressure (mean difference 8.0 mmHg, 95% CI 1.2 to 14.8, and 8.6 mmHg, 95% CI 1.8 to 15.5, respectively). Daily smoking was more common in general-population controls than in donor-eligible controls (22.9% vs. 7.5%). Ejection fraction was higher in donor candidates than in the general-population controls (61.5 ± 4.3% vs. 58.2 ± 4.8%, mean difference 3.3% points, 95% CI 1.1 to 5.5), with all group means within the normal range. Of secondary outcomes, left ventricular stroke volume was higher in donor-eligible controls than in general-population controls (mean difference 8.8 ml, 95% CI 1.4 to 16.2), although this was attenuated after adjusting for heart rate. Conclusion An unscreened general-population control group underestimates the baseline health advantage conferred by donor screening. The differences we observed were small and of uncertain clinical significance for an individual, but they were systematic and consistent in direction. The choice of control group matters, but should follow explicitly from the causal question. While screened controls isolate the effect of donation itself, a population comparator captures the effect of donor selection or population-level risk.