Abstract
Diagnostic criteria for chronic kidney disease (CKD) continue to evolve, particularly for women. In 2002 the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (KDOQI) published guidelines that defined CKD as either kidney damage (proteinuria, hematuria, or structural abnormalities of the kidney) with or without impaired glomerular filtration rate (GFR) or impaired GFR<60 mL/min/1.73 m2 with or without kidney damage that persists over 3 months. The four-variable modification of diet in renal disease (MDRD) equation was used to estimate GFR, moving away from the use of serum creatinine concentration as a marker for CKD. It must be pointed out that the MDRD equation was never validated fully for the elderly, diabetics, children, pregnant women, or Native Americans or Hispanics. The use of these guidelines has caused overdiagnosis of CKD in women and in the elderly. Despite these concerns, it is clearly evident that the prevalence of CKD is high, estimated at 10%–13% worldwide, and that CKD is an important independent risk factor for cardiovascular disease (CVD).
In this issue, Gulati et al. 1 present an important study involving a cohort of 5716 predominantly white volunteer women who at study initiation had no known CKD or CVD. 2 The mean follow-up period for this group of women was 15.9 years, and all-cause mortality was the end point for the study. They found that the estimated glomerular filtration rate (eGFR) was highly predictive of all-cause mortality and that fitness, as measured by symptom-limited treadmill testing, significantly modified the association between eGFR and mortality. Although there were limitations to the study, the findings provide important implications to women's health.
The study examined the relationship between eGFR and all cause mortality both as a continuous variable and as a categorical variable, depending on the CKD stage. The majority of the subjects were classified as having stage 3a CKD (eGFR 45–59 mL/min/1.73 m2). As of 2011, the vast majority of laboratories used the MDRD formula to estimate GFR. A new equation, the Chronic Kidney Disease Epidemiology Collaboratiion (CKD-EPI) equation has been shown to more accurately estimate GFR than the MDRD equation. A recent meta-analysis confirmed this in a large cohort of 1.1 million adults. Compared to the MDRD classification, 24.4% of individuals were reclassified into higher eGFR categories, and the prevalence of CKD stages 3–5 was reduced from 8.7% to 6.3%. 2 This effect was particularly evident in women, the elderly, and subjects with reduced muscle mass. Applied to the study population examined by Gulati et al., 1 this analysis would predict that a third of the subjects in their CKD stage 3a would move to a eGFR of>60 mL/min/1.73 m2. Although this reclassification might affect their categorical analysis, the relationships between eGFR analyzed as a continuous variable would not be affected, and their conclusions remain highly significant.
Several points should be stressed. First, the use of serum creatinine to define CKD should be abandoned in clinical practice. This was clearly shown in this study as well as in many other studies. Second, the equations used to estimate GFR continue to be refined. Although the CKD-EPI formula appears to be the best formula currently in use, it still relies on a single determination of serum creatinine. It is expected that continued improvement in our ability to more accurately define CKD is forthcoming. For now, the CKD-EPI formula should be used to categorize subjects based on their eGFR. Current proposals to subdivide stage 3 CKD into 3a (45–59 mL/min/1.73 m2) and stage 3b (30–44 mL/min/1.73 m2) and use the degree of microalbuminuria/proteinuria as a way to better define patients that are more likely to progress are being evaluated.
Gulati et al. 1 also examined the relationship between proteinuria as measured by a single dipstick determination and all-cause mortality. Proteinuria occurred in 2% of the population and was not an independent risk factor for mortality. It has been shown in a variety of other populations that the presence and degree of proteinuria, particularly albuminuria, are predictive of cardiovascular and cerebrovascular morbidity and mortality. Use of the urinary albumin/creatinine ratio to define high-risk populations has come into vogue since the study was initiated, and it would have been an interesting variable in this study. That fact that this study found no independent association should not detract from the validity of using a measure of proteinuria as a way to define risk for our patients.
Although all-cause mortality was the end point for Gulati et al.'s study 1 based solely on social security records, CVD continues to be the leading cause of death in women. Kurth et al. 3 explored the relationship between eGFR and CVD and mortality in 27,939 women participating in the Women's Health Study. Although the risk of myocardial infarction, stroke, coronary revascularlization, or noncardiovascular mortality was not associated with an eGFR<60 mL/min/1.73 m2, cardiovascular mortality was highly correlated with eGFR. Perhaps the most important aspect of the study by Gulati et al. 1 is their finding that baseline cardiovascular fitness modified patients' risk at all levels of eGFR. The subjects in the best cardiovascular health had lesser risk for all-cause mortality despite having CKD. This has very important public health implications and indicates that lifestyle modification, particularly fitness, might modify risk in our high-risk CKD patients. A prospective study would be necessary to better explore this issue.
In summary, the study by Gulati et al. 1 adds to our understanding of the interplay between risk factors, such as CKD, and fitness level. Future studies are necessary to explore the therapeutic impact of modifying fitness level in our high-risk populations. For now, though, women should be encouraged that fitness reduces risk and that lifestyle modification for risk reduction is appropriate.