Epidemiology of Chronic Kidney Disease Among Normotensives
But What Is Chronic Kidney Disease?
Among adults in the United States, 1 in 4 has hypertension and 1 in 8 has chronic kidney disease (CKD). Although the relationship between hypertension and CKD has been recognized for several hundred years, the prevalence of CKD among patients with normal blood pressure has not been assessed in randomly sampled populations. Crews et al1 in this issue of Hypertension are the first to report such estimates: 13.4% of people who have normal blood pressure have CKD. Among those with prehypertension, the prevalence is 17.3%; among those with undiagnosed hypertension, the prevalence is 22.0%; and among those with diagnosed hypertension, the prevalence is 27.4%. The magnitude of these CKD prevalence estimates is astounding and may even be misleading unless placed in appropriate context. The awareness of CKD diagnosis was dismal: <10% of people were aware of CKD regardless of hypertension category. A thorough analysis of the definition of CKD is necessary to understand how the varying definitions of CKD may have influenced both the prevalence and awareness estimates.
Effect of CKD Definition on Prevalence Estimates
The prevalence estimates may be inflated because of CKD that is so mild that it may not be considered a disease at all. For example, examination of the Figure (derived from Table 3 of the article) shows that if one considers the prevalence of more severe CKD defined as macroalbuminuria or estimated glomerular filtration rate (GFR) <45 mL/min per 1.73 m2, then the prevalence estimates fall dramatically. These stricter definitions of CKD yield the following prevalence estimates of CKD: normal blood pressure 0.5%; prehypertension 1.0%; undiagnosed hypertension 2.0%; and <5.0% prevalence of CKD among those with diagnosed hypertension. These prevalence estimates are much lower than those obtained with the more sensitive definition that is typically used to define CKD. Thus, the prevalence estimates of CKD may be driven up by people in the GFR range of 45 to 60 mL/min per 1.73 m2 and people with microalbuminuria. So what is wrong with these definitions?
These definitions are too sensitive and sometimes not appropriate. For example, a single recording of microalbuminuria is not sufficient to diagnose CKD; microalbuminuria has a high degree of day-to-day variability2 and requires reconfirmation at a later date to be of diagnostic importance. The day-to-day variability is not unique to microalbuminuria. This variability is shared by other analytes, such as glucose. In fact, the prevalence estimates of diabetes mellitus fall by 25% when confirmation of the diagnosis of diabetes mellitus is required by a second measurement.3 GFR in the 45 to 60 mL/min per 1.73 m2 range may be of little clinical consequence, especially when the GFR is stable over years, when it is solely a manifestation of age, sex, and race (as may occur in elderly nonblack women), and when it is unaccompanied by other manifestations of kidney disease, such as albuminuria. The good news is that we do not have so many patients with more severe CKD; the bad news is that there are plenty with mild chronic kidney “disease.”
In addition, there is another caveat in assessing the viability of the definition of hypertension. Blood pressure among patients with CKD is considered hypertensive at ≥130/80 mm Hg. Although the non-CKD definitions of hypertension may have diluted prevalence estimates of CKD in the hypertension groups, this was probably of minor importance. Examination of the Figure will reveal that using more specific definitions (estimated GFR <45 mL/min per 1.73 m2 or macroalbuminuria) may increase the prevalence of CKD by only a slight amount, at most by a few tenths of a percent.
Effect of CKD Definitions on Awareness Estimates
Although patients with CKD are uniformly (≈99%) aware of their diagnosis of hypertension,4 this study suggests that those patients with hypertension, in fact, had poor awareness of CKD. The awareness of CKD among those with normal blood pressure was 2.0%, among those with prehypertension was 2.2%, and among those with undiagnosed hypertension was 3.5%. Even among patients with diagnosed hypertension, the awareness of CKD diagnosis was only 9.1%. Again, the more sensitive definition of CKD may paint a gloomier picture about CKD awareness than a more specific definition. There are several reasons for this. First, physicians may not want to diagnose their patients on the basis of single measurements of estimated GFR or microalbuminuria. Epidemiological studies ignore this fact. As examples, a patient will be classified as having CKD even with a GFR of 59 mL/min on 1 occasion or, even in the absence of diabetes mellitus, a patient will be diagnosed with CKD when the urine albumin/urine creatinine ratio is 40 mg/g on a single test. Second, there may be little reason to screen patients for CKD when risk factors such as hypertension or diabetes mellitus are absent. It is now well established that testing for microalbuminuria among normotensives is not cost-effective and not recommended. Although cost-effective, testing all patients with uncomplicated hypertension for microalbuminuria is not routine, and even if microalbuminuria is detected on a single occasion it may not represent CKD. Third, minimally impaired estimated GFR may not warrant a diagnosis of CKD. Because patients with low estimated GFR in this study, on average, were 72.4 years of age, it is possible that physicians appropriately and wisely avoided labeling their patients with a diagnosis of chronic kidney “disease,” especially when they were elderly. Because the prevalence of kidney disease of greater severity (defined as estimated GFR <45 mL/min per 1.73 m2 or macroalbuminuria) was ≈0.5% among normotensives, 1.0% among prehypertensives, 2.0% among undiagnosed hypertensives, and <5.0% among diagnosed hypertensives, the awareness of CKD diagnosis for more severe impairment in kidney function may have been substantially higher. In fact, analyses from National Health and Nutrition Examination Survey indicate that, even among patients with stage 3 CKD, awareness was increased 3-fold when they had proteinuria and 2-fold when they had diabetes mellitus.5
Tradeoffs With CKD Definitions
The use of sensitive or specific definitions when assessing the awareness and prevalence of CKD has its advantages and its pitfalls. A more specific definition would identify those individuals with CKD who are at increased absolute risk. In such patients, the absolute risk for cardiovascular morbidity and mortality and the risk for ESRD would be increased. Given the heightened absolute risk, the number of people needing treatment to prevent 1 event would be lower. Lower numbers needed to treat may translate to more cost-effective therapies and allow us to target our resources to those individuals who are most in need of care. However, if a more sensitive definition is used, then we may be able to identify CKD at an earlier stage and intervene before it progresses to a later stage. Identification of microalbuminuria or milder degrees of clinical reductions in GFR may be of enormous prognostic importance to patients who have the appropriate clinical context or disease trajectories that suggest progression. Population-based studies suggest that the joint consideration of both parameters of kidney function, albuminuria and estimated GFR, may be most effective in assessing prognosis.6 Thus, sensitive and specific definitions both have a role at the bedside and for clinical epidemiologists. Which definition to use to define CKD depends on the context.
Link Between Hypertension and CKD: Estimated GFR Versus Albuminuria
An important finding of this report was that hypertension was more strongly related to albuminuria rather than the estimated GFR. These results confirm earlier observations made through ambulatory blood pressure monitoring, which disclose that proteinuria was the most important correlate of systolic hypertension among patients with CKD; proteinuria trumps estimated GFR.7 Analysis of National Health and Nutrition Examination Survey data also suggests that albuminuria is an important correlate of hypertension control.8 In the more recently reported Chronic Renal Insufficiency Cohort (CRIC) study, hypertension prevalence and control assessed by clinic blood pressure measurement confirms that albuminuria is an important correlate of poor hypertension control.4 Although only increasing severity of albuminuria is related to the mean level of ambulatory blood pressure, any CKD, whether detected by the presence of albuminuria or an impairment in estimated GFR, is associated with marked impairment in the circadian rhythm of blood pressure, including dipping.9 These data point out the important relationships between albuminuria and hypertension above and beyond estimated GFR.
Taken together, the results of the study indicate that the prevalence estimates of CKD are strongly dependent on the definitions that we use to define this disorder. More specific definitions yield low estimates, more sensitive definitions yield ≤26-fold higher estimates especially among normotensive and prehypertensive individuals. The study provides strong evidence that hypertension is not sine qua non of CKD. Because the prehypertensive state can identify patients who have mild impairments of kidney function or microalbuminuria, physicians must not rely solely on the presence of hypertension to screen for CKD among patients who are at high risk, such as those with diabetes mellitus. Finally, given the higher prevalence of white coat and masked hypertension among patient with CKD,10 the assessment of cardiovascular and renal risks by measurement of blood pressure in the clinic can be enhanced by home blood pressure monitoring.11
Sources of Funding
The author was supported by National Institutes of Health grant 2RO1-062030-06 and a grant from US Veterans’ Affairs Merit Review.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
Crews DC, Plantinga LC, Miller ER III, Saran R, Hedgeman E, Saydah SH, Williams DE, Powe NR, for the Centers for Disease Control and Prevention Chronic Kidney Disease Surveillance Team. Prevalence of chronic kidney disease in persons with undiagnosed or prehypertension in the United States. Hypertension. 2010; 55: 1102–1109.
Muntner P, Anderson A, Charleston J, Chen Z, Ford V, Makos G, O'Connor A, Perumal K, Rahman M, Steigerwalt S, Teal V, Townsend R, Weir M, Wright JT Jr. Hypertension awareness, treatment, and control in adults with CKD: results from the Chronic Renal Insufficiency Cohort (CRIC) Study. Am J Kidney Dis. 2010; 55: 441–451.
Agarwal R, Andersen MJ. Correlates of systolic hypertension in patients with chronic kidney disease. Hypertension. 2005; 46: 514–520.
Peralta CA, Hicks LS, Chertow GM, Ayanian JZ, Vittinghoff E, Lin F, Shlipak MG. Control of hypertension in adults with chronic kidney disease in the United States. Hypertension. 2005; 45: 1119–1124.
Agarwal R, Light RP. GFR, proteinuria and circadian blood pressure. Nephrol Dial Transplant. 2009; 24: 2400–2406.
Bangash F, Agarwal R. Masked hypertension and white-coat hypertension in chronic kidney disease: a meta-analysis. Clin J Am Soc Nephrol. 2009; 4: 656–664.
Pickering TG, Miller NH, Ogedegbe G, Krakoff LR, Artinian NT, Goff D. Call to action on use and reimbursement for home blood pressure monitoring: a joint scientific statement from the American Heart Association, American Society of Hypertension, and Preventive Cardiovascular Nurses Association. Hypertension. 2008; 52: 10–29.