OVER THE PAST DECADE, CHRONIC KIDNEY DISEASE (CKD), defined as impaired glomerular filtration rate (GFR) or elevated urine albumin excretion, has been increasingly recognized as an important risk factor contributing to cardiovascular disease and death. Moreover, the prevalence of CKD has increased, with CKD now affecting approximately 13.1% of Americans. This increase has been driven by the increasing prevalence of the most common risk factors for CKD, including diabetes, hypertension, and—perhaps most important—older age. Among US adults aged 70 years or older, the prevalence of CKD has been estimated to be more than 45%. However, the clinical implications of CKD in elderly persons remain uncertain. In particular, substantial controversy exists regarding whether CKD represents a true disease or whether loss of GFR is simply part of the natural aging process. In this issue of JAMA, the CKD Prognosis Consortium provides valuable new data regarding the importance of CKD in older adults. Based on findings from more than 2 million participants from 46 cohorts, the authors report that among older adults, CKD is associated with excess mortality risks that are as high as or higher than the excess risks observed among middle-aged adults. For example, after accounting for a number of potential confounding variables, for participants aged 18 to 54, 55 to 64, 65 to 74, or 75 or more years at baseline, an estimated GFR of 45 vs 80 mL/min/1.73 m was associated with 9.0, 12.2, 13.3, or 27.2 additional estimated deaths per 1000 person-years of follow-up, respectively. These data, combined with knowledge that CKD is more prevalent with advancing age, identify older adults as a population for which CKD has exceptional individual and public health importance. Risks of end-stage renal disease (ESRD) were also examined by age. For older adults, absolute incidence rates of ESRD were substantially lower than absolute mortality rates. For example, for participants aged 75 years or older with an estimated GFR of 45 mL/min/1.73 m, the estimated average incidence rate of ESRD was 9.8 per 1000 personyears of follow-up, compared with an estimated average mortality rate of 85.0 per 1000 person-years. In addition, associations of CKD with risk of ESRD were attenuated comparing older with younger participants. These data are consistent with prior reports demonstrating that older adults are less likely to receive kidney replacement treatment for ESRD than younger individuals and that the poor outcomes of older adults with CKD occur predominantly without progression to treated ESRD. The CKD Prognosis Consortium study has 2 main salient features. First, it analyzes data across 46 cohort studies that include more than 2 million participants. This yields results that are broadly applicable and provides adequate statistical power to evaluate whether the association of CKD with health outcomes varies by age. Second, excess risk is evaluated on the clinically relevant additive scale, in addition to the more commonly applied relative scale. Differences in absolute risk calculated on the additive scale reflect the marginal risk of a risk factor, such as CKD, accounting for the “baseline” risk among unexposed members of a population. These absolute risk differences can be easily compared with those of other risk factors or, in the case of treatment studies, can be used to directly compare benefits with harms and calculate the number needed to treat. Moreover, if an exposure is assumed to be causally related to the outcome examined, the absolute risk difference most meaningfully captures the public health benefit that could be achieved by preventing or effectively treating the risk factor. What should the medical community conclude from the new data? Older adults with CKD should be recognized as patients at high risk of death. In CKD, excess mortality risk is largely attributable to cardiovascular disease, although other disease pathways are also important. Therefore, all efforts should be made to apply treatment strategies that are proven effective for the primary or secondary prevention of cardiovascular disease in this population and are appropriate to the individual. These may
Read full abstract