|Home | About | Journals | Submit | Contact Us | Français|
In the general population, approximately 38% of adults aged 70 or older have an estimated glomerular filtration rate (eGFR) < 60 ml/min/1.73 m2, with most having moderate reductions in eGFR in the 30-59 ml/min/1.73 m2 range (1). Some have argued that such moderate reductions in eGFR can occur as the result of normal aging and should not be equated with chronic kidney disease (CKD) in the absence of other abnormalities (2-4). Others point out that reductions in eGFR in the elderly may reflect the high prevalence of kidney disease risk factors at older ages and should not necessarily be considered normal simply because they are common (5).
In this issue of the American Journal of Kidney Diseases, Roderick and colleagues take a different approach to this issue (6). Instead of asking whether older patients with moderate reductions in eGFR do or do not have CKD, they ask whether reductions in eGFR in the elderly are associated with a clinically significant outcome--mortality. These authors address this question using data collected as part of a randomized trial that compared two methods of multi-dimensional assessment in a representative sample of adults aged 75 and older drawn from primary care practices across Britain. More than half of the members of this cohort had an eGFR < 60 ml/min/1.73 m2. Among these, most had mild to moderate decrements in eGFR (45-59 ml/min/1.73 m2) and most were women (Figure 1). Baseline characteristics did not differ greatly between those with an eGFR 45-59 ml/min/1.73 m2 and the referent group with an eGFR ≥ 60 ml/min/1.73 m2, particularly among women. The prevalence of traditional complications of CKD was very low. For example, only 1% of women with an eGFR 45-59 ml/min/1.73 m2 had hyperphosphatemia and less than 6% had a hemoglobin <11 g/dL. The prevalence of cognitive and functional impairment was by comparison much higher. For example, almost 20% of women with an eGFR 45-59 ml/min/1.73 m2 had cognitive impairment, 33% were partially or fully dependent in two or more activities of daily living and 8% had fallen at least twice during the previous six months. However, as for many baseline characteristics, these conditions were no more common in this group than they were in women with an eGFR ≥ 60 ml/min/1.73 m2.
Crude death rates were high for all groups but increased with falling eGFR from 8% among those with an eGFR ≥ 60 ml/min/1.73 m2, to 21% among those with an eGFR < 30 ml/min/1.73 m2. Mortality rates were consistently higher and increased more steeply across eGFR categories among men compared with women. Compared with the referent group, absolute mortality rates for those with an eGFR 45-59 ml/min/1.73 m2 were approximately 1% per year higher for women and 3% per year higher for men. After adjustment for baseline characteristics, the risk for all-cause mortality relative to the referent category with an eGFR ≥ 60 ml/min/1.73 m2 was only minimally elevated for those with an eGFR 45-59 ml/min/1.73 m2, and this did not reach statistical significance for women. Nevertheless, for both men and women in this group, death was more likely to be due to cardiovascular causes than for patients with higher levels of eGFR, particularly if this occurred within the first two years of follow-up. The presence of dipstick proteinuria did not contribute substantial additional prognostic information beyond that provided by eGFR. Roderick and colleagues conclude that in the elderly, identification and management of CKD should focus on the select group with more severe reductions in eGFR.
Unlike many studies examining the relationship between eGFR and mortality in the elderly, members of the cohort described by Roderick and colleagues underwent detailed baseline assessment for both clinical and functional status measures and are thus extraordinarily well characterized. Furthermore, while clinical trial populations are often poorly representative of the general population, this study population was deliberately selected so as to be representative of older adults in general practice. Nevertheless, some caution is warranted in interpreting the results of this study. While mortality is clearly an important and reliably ascertained outcome, other clinical outcomes may be more important in the elderly, including hospitalization, non-fatal cardiovascular events, disability, reduced quality of life, cognitive decline and loss of independence. In addition, because ascertainment of proteinuria was based only on the results of dipstick, this study may underestimate the prognostic value of urine protein measurements in this cohort. Finally, as the authors point out, heterogeneity in mortality rates among those with an eGFR ≥ 60 ml/min/1.73 m2 render estimates of relative risk sensitive to the referent category selected.
Notwithstanding these limitations, the findings reported here are very consistent with the results of earlier studies conducted in more broadly defined cohorts demonstrating that the relative risk of death associated with a given level of eGFR is attenuated in the elderly (7-10). The significance of this finding is that in the elderly, the threshold level of eGFR below which mortality risk exceeds that in the referent category is lower than in younger patients ((9, 10). Collectively, these studies raise the question of whether a disease-oriented approach is appropriate for many elderly individuals who meet criteria for CKD. For patients with moderate reductions in eGFR, the National Kidney Foundation's Kidney Disease Outcomes Quality Initiative (KDOQI) clinical action plan recommends the following strategies: diagnosis and treatment of kidney disease, treatment of comorbid conditions, slowing progression, cardiovascular risk reduction, estimating progression and evaluating and treating complications (11). However, the value of this approach in older adults with mild to moderate reductions in eGFR is uncertain if most are at no higher risk of death than their age peers with higher levels of eGFR, do not have a treatable complication of CKD and have a less than one percent per year chance of progressing to end-stage renal disease (as demonstrated in prior studies) (12, 13).
Tinetti and Fried have pointed out that disease-oriented models of care evolved in an era when life was short and life expectancy was impacted by dominant disease processes (14). They argue that the management of complex comorbidity in the elderly calls for a more individualized approach to care focused on attainment of patient goals and on impacting modifiable processes—both biologic and non-biologic (14). Table 1 highlights the differences between these two models of care. It is important to note that an individualized approach does not preclude the application of disease-specific treatment strategies when these may impact outcomes of importance to the individual patient.
Under the individualized care model, knowing whether or not an elderly person with moderate reductions in eGFR has CKD assumes less importance than it might under a disease-oriented model. Instead, greater emphasis is placed on assessing each patient's risk for particular outcomes, eliciting which outcomes are most important to the patient and identifying processes that can be modified. Thus, for example, kidney disease-specific strategies may be very appropriate for an elderly person with an eGFR of 55 ml/min/1.73 m2 who has evidence for progressive loss of kidney function and/or metabolic complications of CKD, particularly if there are no competing health concerns. Conversely, in an elderly person with a stable eGFR of 55 ml/min/1.73 m2 and no complications of CKD, other health concerns may reasonably take precedence.
In conclusion, an individualized model may be more appropriate than a disease-oriented model of care for many elderly people who meet criteria for CKD. The substantial prognostic significance of eGFR in the elderly suggests that it may be very helpful in crafting individualized treatment plans. The clinical utility of this measure could be further enhanced by more studies like this one. Studies that embrace the complexity of how eGFR relates to significant outcomes in older adults in relation to potential modifying factors. The better we understand what eGFR and urinary protein can tell us about what lies in store for our elderly patients, the more helpful these measures will be in the clinical setting.
Ann M. O'Hare, MA, MD
University of Washington, VA/Puget Sound Medical Center
Financial[ND1] Disclosure: None.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.