Scope and Purpose

The objective of this systematic review is to evaluate the evidence for the potential benefits and harms of: (1) screening adults for chronic kidney disease (CKD) stages 1–3, (2) monitoring adults with CKD stages 1–3 for progression of kidney dysfunction and/or damage, and (3) treatment of adults with CKD stages 1–3.

This report's scope is limited to early stage CKD because it is intended to inform patient care decisions of primary care physicians. This report also is intended as background material to assist groups developing clinical practice recommendations.

Definition of CKD

CKD is a condition in which the kidneys are damaged and/or cannot filter blood normally.¹ CKD usually is asymptomatic, except in its most advanced state. Consequently, blood and/or urine tests generally are required to make a diagnosis.

There has been substantial debate regarding how to define early stages of CKD. The definition of CKD developed by Kidney Disease Outcomes Quality Initiative (KDOQI)² was:

1. Kidney damage present at least 3 months, as defined by structural or functional abnormalities (most often based on increased albuminuria, e.g., urinary albumin-creatinine ratio [UACR] ≥30 mg/g); and/or
2. Glomerular filtration rate (GFR) <60 mL/min/1.73 m² present at least 3 months.

Within this framework, KDOQI then classified CKD into five stages, as follows:

• Stage 1: Kidney damage with GFR ≥90 mL/min/1.73 m2.
• Stage 2: Kidney damage with GFR 60-89 mL/min/1.73 m2.
• Stage 3: GFR 30-59 mL/min/1.73 m2.
• Stage 4: GFR 15-29 mL/min/1.73 m2.
• Stage 5: GFR <15 mL/min/1.73 m2 or kidney failure treated by dialysis or transplantation.

A limitation of the KDOQI definition and staging was that they were based on cross sectional data, and that there were limited data associating adverse clinical outcomes with specific levels of GFR, albuminuria, or proteinuria. However, results of a recent series of meta-analyses of multiple large prospective cohort studies clearly demonstrated the independent associations of each level of GFR and albuminuria (or alternatively of dipstick proteinuria), with total and cardiovascular mortality, ESRD and acute kidney injury (AKI).^3-6 These associations were independent of cardiovascular risk factors. Based in part on these data, a consensus conference led by Kidney Disease: Improving Global Outcomes (KDIGO), on Chronic Kidney Disease: Definition, Classification and Prognosis, concluded that the current CKD definition should be preserved. However, the conference recommended that staging be altered to subdivide stage 3 into 3a (GFR 45-59 mL/min/1.73 m²) and 3b (GFR 30-44 mL/min/1.73 m²), to add albuminuria strata within each GFR stage (UACR <30 mg/g, 30-299 mg/g, or ≥300 mg/g), and to assign a cause of CKD when possible.⁷

Prevalence of CKD

In the United States, based on data from the 1999-2006 National Health and Nutrition Examination Survey (NHANES) study, an estimated 11.1 percent (22.4 million) of adults aged 20 or older have CKD stages 1–3.⁸ Because this estimate was based on one-time measurements of urinary albumin-creatinine ratio (UACR) and serum creatinine, and the definition of CKD requires persistent kidney abnormalities, statistical adjustments were made to estimate persistence. An additional 0.8 million U.S. adults aged 20 or older have CKD stage 4, and more than 0.3 million have stage 5 CKD and receive hemodialysis.⁹

Among adults with CKD stages 1–3, approximately half have either stage 1 or 2 CKD (increased albuminuria with normal GFR), and half have stage 3 CKD (low GFR, with approximately one third of these having increased albuminuria and two thirds having normal albuminuria).⁸ Of individuals with albuminuria, nearly 85 percent have microalbuminuria (UACR 30-299 mg/g).

Analyses of NHANES data between 1988-1994 and 1999-2004 suggest that the prevalence of CKD is rising for every CKD stage, but with a particular increase in the prevalence of individuals classified with CKD stage 3.¹⁰ The number of patients with stage 5 CKD requiring dialysis also has increased.⁹ It has been estimated that more than 700,000 individuals will have end-stage renal disease (ESRD) by 2015.¹¹

Factors Associated With CKD

Prevalence of CKD stages 1–3 in U.S. adults rises from 3.1 percent among those aged 20-39 years, to 6.7 percent in those aged 40-59, 17.6 percent in those aged 60-69, and 44.4 percent among adults aged 70 years or older.⁸ CKD prevalence is somewhat higher in women (12.6 percent) than in men (9.7 percent) and is similar in whites (11.6 percent) and blacks (11.2 percent).

Although CKD can be caused by primary kidney disease (predominantly glomerular diseases, tubulointerstitial diseases, obstruction, and polycystic kidney disease), in the vast majority of patients with CKD, the kidney damage is associated with other medical conditions such as diabetes and hypertension. Other risk factors for CKD include older age, cardiovascular disease, obesity, family history, and African American, Native American, or Hispanic ethnicity. With respect to diabetes as a CKD risk factor, based on NHANES 1999-2006 data, prevalence of diabetes was approximately 5 percent in individuals without CKD and 20 percent in individuals with CKD stages 1–3.¹² Prevalence of hypertension was 24 percent among individuals without CKD, but rose from 36 percent in those with CKD stage 1 to 64 percent in those with CKD stage 3. Similarly, prevalence of cardiovascular disease was 6 percent among individuals without CKD, and rose from 7 percent in those with CKD stage 1 to 36 percent in those with CKD stage 3. Compared with the NHANES population, the prevalence of comorbidities was higher in the older Medicare population. Excluding those with ESRD, in 2008, 48 percent of Medicare patients with CKD had diabetes, 91 percent had hypertension, and 46 percent had atherosclerotic heart disease.¹²

Association of CKD With Adverse Outcomes

CKD has been associated with numerous adverse health outcomes. Many studies have reported that a GFR of 30-59 mL/min/1.73 m² is associated with an increased risk of mortality,^3,13 cardiovascular disease,¹⁴ fractures,¹⁵ bone loss,¹⁶ infections,¹⁷ cognitive impairment,¹⁸ and frailty.¹⁹ Similarly, there appears to be a graded relationship between the severity of proteinuria or albuminuria and adverse health outcomes, including mortality,^3,20 ESRD,²¹ and cardiovascular disease.²² Further, the risk for adverse outcomes conferred by reduced GFR and increased albuminuria (or proteinuria) appears to be independent and multiplicative.^3,21

A number of possible explanations exist for the observed association of CKD with adverse health outcomes. First, CKD shares many of the same risk factors as other vascular diseases, such as older age, hypertension, and diabetes, so CKD may be a marker for undiagnosed vascular disease or for a worsened prognosis among individuals with known vascular disease. Second, CKD may be associated with a number of nontraditional risk factors for vascular disease and mortality, such as increased inflammation or bone mineral disorders. Third, CKD may be a marker for individuals less likely to receive proven medical therapies. For example, among individuals with myocardial infarction, those with CKD are less likely to receive proven effective therapies such as coronary artery bypass grafting, angiotensin converting enzyme inhibitors (ACEI), beta-blockers, or HMG CoA-reductase inhibitors (i.e., statins).²³ Therefore, systematic undertreatment may in part underlie the association between CKD and adverse health outcomes. Finally, the associations of CKD with adverse health outcomes and increased healthcare costs may be related to a combination of the above mechanisms.

Rationale for CKD Screening

Factors that impact the potential benefit of screening adults for CKD stages 1–3 include: (1) whether undiagnosed CKD is sufficiently prevalent in the population, overall or in certain high risk groups; (2) whether CKD is associated with significant adverse health consequences and/or healthcare costs; (3) whether CKD is accurately diagnosable while asymptomatic; (4) whether there are valid and reliable screening tests for CKD that are acceptable to patients and available in primary care settings; and (5) whether there are treatments for patients with CKD that improve clinically important health outcomes.

Going further, determination that CKD screening is beneficial would require evidence that treatment of screen-detected CKD is associated with an improvement in health outcomes compared with treatment initiated once an individual is symptomatic or has CKD detected through usual care, while limiting harms. In addition, since potential CKD treatments often are indicated for conditions associated with CKD, such as diabetes, hypertension, or cardiovascular disease, demonstration that CKD screening is beneficial may require evidence that treatment benefits CKD populations who don't have another indication for treatment or, that among patients with another indication for treatment, those with CKD experience a greater relative treatment benefit than those without CKD. Alternatively, because patients with diabetes, hypertension, and/or cardiovascular disease who also have CKD are at significantly higher risk for adverse health outcomes than patients with these comorbid conditions who don't have CKD, diagnosis of CKD resulting from screening patients with these conditions would identify a group, if currently untreated, who could derive a greater absolute benefit in health outcomes even if the relative benefit of treatment versus no treatment was similar in CKD and non-CKD patients.

Several organizations have made recommendations regarding screening for CKD. KDIGO recommends screening for CKD in patients with hypertension, diabetes, or cardiovascular disease using both a urine test for proteinuria and a blood test for creatinine to estimate GFR.²⁴ KDIGO further recommends that CKD screening be considered in patients who are older, have a family history of kidney disease, have other cardiovascular disease risk factors, have certain chronic infections or cancers, or are treated with potentially nephrotoxic drugs, and that screening need not be performed more often than annually. The American Diabetes Association (ADA) recommends that all adults with diabetes undergo annual measurement of serum creatinine to estimate GFR, and that all type 2 diabetics and all type 1 diabetics with a diabetes duration of at least 5 years undergo annual measurement of urinary albumin excretion.²⁵ Ongoing CKD screening programs include the National Kidney Foundation's Kidney Early Evaluation Program (KEEP^®), which offers free screening for all adults with hypertension, diabetes, or a first degree relative with a history of kidney disease, hypertension, or diabetes.²⁶

Rationale for Monitoring for Progression of CKD

Because CKD in stages 1–3 is usually asymptomatic, monitoring these patients for worsening kidney function or damage requires laboratory testing (i.e., measures to estimate GFR, albuminuria).

Factors that impact the potential benefit of monitoring adults with CKD stages 1–3 for worsening kidney function or damage include: (1) whether undiagnosed progression of patients with CKD stages 1–3 to worse kidney function or damage is sufficiently frequent in the population, overall or in certain high risk groups; (2) whether CKD that has progressed from stages 1–3 is associated with significant adverse health consequences and/or healthcare costs; (3) whether CKD that has progressed from stages 1–3 is diagnosable while asymptomatic; (4) whether there are valid and reliable monitoring tests for CKD stages 1–3 that are acceptable to patients and available in primary care settings; and (5) whether there are treatments for patients whose CKD has progressed from stages 1–3 that improve clinically important health outcomes.

Strictly considered, determination that monitoring patients with CKD stages 1–3 for worsened kidney function or damage is beneficial would require evidence that modified treatment of worsened CKD detected by monitoring is associated with an improvement in health outcomes compared with treatment modified once an individual becomes symptomatic or has CKD worsening detected through usual care, while limiting harms.

Several organizations have made recommendations regarding monitoring kidney function and/or damage in patients with CKD. KDOQI recommends that adults with CKD receive monitoring of urinary albumin or protein to creatinine ratio, though no frequency of monitoring was recommended.²⁷ The U.K. National Health Service (NHS) National Institute for Health and Clinical Excellence (NICE) guidelines suggest “more frequent monitoring” in CKD patients with worsening kidney function and a “relaxed frequency” of estimated GFR measurements in patients with stable kidney function.²⁸

Rationale for Treatment of CKD

In patients treated for nonprimary CKD, treatment most often is not directed specifically at the CKD but rather at the associated underlying conditions or cardiovascular risk factors, such as hypertension or diabetes,²⁹ with therapeutic goals for these conditions sometimes set more strictly for CKD patients than for non-CKD patients.³⁰ An aim of this systematic review is to evaluate the evidence regarding whether the benefits and harms of treatment differ between patients with and without CKD, both in patients with and without other indications for treatments. Medications such as ACEI and angiotensin receptor blockers (ARB) potentially could be directed specifically towards treatment of CKD. However, whether their impact on CKD outcomes or markers (e.g., incident ESRD, albuminuria severity³¹) is independent of their blood pressure lowering effect is not clear.³² Additional nonspecific therapies may include other medications and nonpharmacological interventions targeted, for example, at blood pressure control, glycemic control, cholesterol control, and obesity treatment.

Analytic Framework and Key Questions

During this project's topic development, the topic nominators and other interested parties agreed that an independent, comprehensive review of the issues introduced above would provide helpful guidance to clinicians and policymakers regarding diagnosis and management of early stage CKD. There was consensus that the following analytic framework (Figure 1) and Key Questions addressed the most important issues regarding CKD stages 1–3:

Figure 1

Analytic framework for screening, monitoring, and treatment of chronic kidney disease stages 1–3.

Key Question 1.

In asymptomatic adults with or without recognized risk factors for chronic kidney disease (CKD) incidence, progression or complications, what direct evidence is there that systematic CKD screening improves clinical outcomes?

• In asymptomatic adults with or without risk factors for CKD incidence, progression, or complications, what is the accuracy and reliability of CKD screening and the prevalence of CKD identifiable by screening?
• Does initiating treatment for CKD as a result of systematic screening improve clinical outcomes compared with treatment initiated after incidental CKD diagnosis during routine clinical practice?
• How do patient factors and CKD screening thresholds modify the yield of CKD screening and its association with clinical benefits?

Key Question 2.

What harms result from systematic CKD screening in asymptomatic adults with or without recognized risk factors for CKD incidence, progression or complications?

• How do patient factors and CKD screening thresholds modify the association of CKD screening with harms?

Key Question 3.

Among adults with CKD stages 1–3, whether detected by systematic screening or as part of routine care, what direct evidence is there that monitoring for worsening kidney function and/or kidney damage improves clinical outcomes?

• How do patient factors, CKD severity/stage, and CKD monitoring intervals modify the association of CKD monitoring with clinical benefits?

Key Question 4.

Among adults with CKD stages 1–3, whether detected by systematic screening or as part of routine care, what harms result from monitoring for worsening kidney function/kidney damage?

• How do patient factors, CKD severity/stage, and CKD monitoring intervals modify the association of CKD monitoring with harms?

Key Question 5.

Among adults with CKD stages 1–3, whether detected by systematic screening or as part of routine care, what direct evidence is there that treatment improves clinical outcomes?

• Does the presence of CKD modify the likelihood of improvement in clinical outcomes associated with treatment of vascular disease or vascular risk factors?
• Among adults with CKD, what patient factors modify the association of specific treatments with improved clinical outcomes?

Key Question 6.

Among adults with CKD stages 1–3, whether detected by systematic screening or as part of routine care, what harms result from treatment?

• Does the presence of CKD modify the likelihood of harms associated with treatment of vascular disease or of vascular risk factors?
• How do patient factors and CKD severity/stage modify the association of CKD treatment with harms?