Defining, Treating, and Understanding Chronic Kidney Disease—A Complex Disorder

Abstract

Chronic kidney disease (CKD) is prevalent in more than 20 million people in the United States. The majority of care provided to patients with this disease comes from primary care physicians, although it is often poorly understood. After an extensive literature review, it is clear that it can be difficult to classify and there are many barriers to care. Risk factors for both incident CKD and disease progression include hypertension, poor glycemic control, sociodemographic factors, acute kidney injury, metabolic acidosis, and possibly hyperuricemia and dietary factors. Treatment of patients with CKD should focus on mitigating risk factors, as well as common comorbidities such as cardiovascular disease, anemia, and bone mineral disease. Novel therapies such as pirfenidone, pentoxifylline, and endothelin‐1 antagonists are being investigated with promising results.

Primary care physicians use terms such as “kidney disease,” “CKD,” and many other phrases to describe when a patient has increased serum creatinine or proteinuria. But what is the significance of the problem? Is it acute or chronic? What are the implications for the patient's long‐term health? What do primary care providers need to know? Chronic kidney disease (CKD) is a pervasive disease, with implications on virtually every organ system in the human body, and it is too much to expect primary care physicians to know every effect, however large or small. Yet, there are key points in the pathophysiological, psychosocial, and treatment aspects of CKD that should be emphasized. This review is an attempt to cover the major aspects of CKD and to focus on major risk factors for progression, including hypertension and diabetes; lesser‐known risk factors such as acidosis, hyperlipidemia, and hyperuricemia; kidney injury events that contribute to the long‐term risk of CKD progression such as acute kidney injury and nephrolithiasis; and treatment strategies such as blood pressure (BP) control, glycemic control, pH neutralizers, and the budding field of antifibrotic and antiviscosity agents (Figure 1).

Figure 1.

Potential chronic kidney disease (CKD) treatment strategies. RAAS indicates renin‐angiotensin‐aldosterone system.

Prevalence

The Centers for Disease Control and Prevention currently estimates that roughly 10% to 16%, or >20 million, Americans 20 years and older have CKD and that these patients are at somewhere between a 16 to 40 times greater risk for dying (mostly from cardiovascular disease) before they progress to end‐stage renal disease (ESRD). Because of these staggering statistics, the overall goal in patients with CKD is to improve primary prevention with early recognition and treatment. Despite this goal, kidney disease is often diagnosed late in its course, because many patients are asymptomatic. The health care costs associated with CKD in a managed care payer format have been estimated at an unadjusted annual per‐patient mean cost difference of $8000 to $11,000 between patients with and without CKD, with the difference mostly driven by hospitalization costs.1 The 2009 United States Renal Data System (USRDS) report showed that costs associated with ESRD included 6% to 7% of the total annual Medicare budget and 1% of the total US budget, with costs of renal replacement therapy (RRT) ranging from roughly $30,000 to $80,000 annually for patients on hemodialysis, peritoneal dialysis, or after receiving a kidney transplant.2

The classification and staging of CKD has recently been revised to include proteinuria as a component of CKD, as this has been shown to be an independent risk for CKD progression and cardiovascular events regardless of baseline glomerular filtration rate (GFR).3 Hemmelgarn and colleagues4 performed a community‐based cohort study among 920,985 adults who had at least 1 outpatient serum creatinine measurement and did not require RRT at baseline and found that the fully adjusted rate of all‐cause mortality was higher in participants with lower estimated GFRs (eGFRs) or heavier proteinuria and was more than two‐fold higher among individuals with heavy proteinuria and eGFR of >60 mL/min/1.73 m² when compared with those with eGFR of 45 mL/min/1.73 m² to 59.9 mL/min/1.73 m² and normal protein excretion. According to the Kidney Disease: Improving Global Outcomes (KDIGO) 2012 clinical guidelines, CKD is defined as abnormalities of kidney structure or function present for >3 months with implications for health. It is classified according to GFR category and albuminuria category. The previously well‐known five‐stage classification has been mostly retained, although the stage 3 category of 30 mL/min/1.73 m² to 59 mL/min/1.73 m² has been divided into 3a (45 to 59 mL/min/1.73 m²) and 3b (30 mL/min/1.73 m² to 44 mL/min/1.73 m²), which was driven by data supporting different outcomes in patients who fell within these two categories (Figure 2). Stages of albuminuria include A1, A2, and A3, based on an albumin excretion ratio (AER; in mg/d) or albumin‐creatinine ratio (ACR; in mg/g) of <30, 30–300, or >300, respectively. The authors also recommended measuring cystatin C to calculate GFR in adults whose eGFR using serum creatinine (eGFRcreat) is 45 mL/min/1.73 m² to 59 mL/min/1.73 m² if they do not have other markers of kidney disease. Current beliefs among many in the CKD community are that the Modification of Diet in Renal Disease (MDRD) equation may overestimate the number of patients with CKD who fall into stage 3a, especially those who are elderly with no other signs of kidney disease. Use of eGFR using cystatin C (eGFRcys) to confirm CKD in populations has found that patients with eGFRcreat <60 mL/min/1.73 m², and confirmed by eGFRcys, have markedly elevated risks for death, cardiovascular disease, and ESRD compared with those with eGFRcys >60 mL/min/1.73 m².5

Figure 2.

Glomerular filtration rate (GFR) and albuminuria categories for staging of chronic kidney disease (CKD).3

As a result of the large number of patients living with CKD, primary care physicians are the backbone of diagnosis and treatment of CKD, and it is not until stage 3 or 4 disease with macroalbuminuria that referral to a nephrologist is warranted. However, CKD patients are often referred to nephrologists too late by primary care physicians, particularly elderly women, minorities, and patients with multiple comorbidities, which is associated with poorer outcomes.6

Underrecognition of CKD has many consequences for patients, including inappropriate medication dosing. In a French study that used the MDRD equation to calculate eGFR, 96% of patients with stage 4 or 5 CKD (n = 25) had either inappropriate drug dosing or were prescribed contraindicated drugs, leading to a 40% higher all‐cause mortality after adjusting for behavioral and sociodemographic factors.7 Suggested interventions to improve this oucome include automated GFR reporting, point‐of‐care physician reminders for measuring ACR values, and prescribing angiotensin‐converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs), although these interventions have not yet shown a significant change in patient outcomes.8 One study that assessed attitudes of nephrologists and primary care physicians collaborating on a theoretical patient who had progressed from stage 3 to stage 4 CKD showed that while both groups desired collaboration, nephrologists were more likely to desire collaborative care for electrolyte management, predialysis discussions, and anemia management, whereas primary care physicians desired more collaboration when patients had both diabetes and hypertension vs hypertension alone. Their belief was that care by nephrologists could slow CKD progression, especially when insurance barriers to nephrology referral were not present.9

It is clear that there are many barriers to caring for patients with CKD, including identification of such patients, defining their disease, and timely referral to nephrologists. However, through modifying common risk factors such as hypertension and diabetes as discussed below, the majority of the CKD population will derive important benefits for health care.

Risk Factors for CKD Progression: Epidemiology

Hypertension

The association between hypertension and CKD appears simple at first glance, but is in fact very complex. The dynamic relationship between the two diseases is not fully known, but there are data to suggest that hypertension may facilitate rather than actually cause the progression of renal disease. Pathologically, most patients with primary hypertension develop benign nephrosclerosis, where the glomeruli are mostly spared and proteinuria has not occurred. Damage progresses fairly slowly with ischemic nephron loss; however, renal function is rarely seriously compromised. Typically, the healthy kidney is able to autoregulate changes in vascular resistance and renal blood flow when there are changes in systemic hypertension. Preglomerular resistance vessels in the afferent glomerular arteriole dilate and constrict to maintain relatively constant glomerular capillary pressures, depending on ambient systemic BP. However, over time, chronic remodeling is seen in these resistance vessels as they are exposed to higher pressures. This renal autoregulation system is believed to be mediated by interaction between a myogenic and tubuloglomerular feedback system. If hypertension becomes severe and exceeds a critical threshold, “malignant” nephrosclerosis may extend to renal arteries, arterioles, and glomeruli.10 This can occur acutely, when BP exceeds the threshold for vascular injury, and the autoregulatory ability of the afferent glomerular arteriole is compromised. In patients with existent diabetes or nondiabetic proteinuric CKD, their susceptibility to renal damage is greatly increased with mild BP elevation. In proteinuric patients, the dominant pathologic lesions associated with their kidney disease are glomerular, suggesting a different pathogenesis than hypertensive injury.11 In African American patients, a proposed risk factor for CKD progression is the higher frequency of expression of both APOL1 G1 and G2 alleles and its association with a higher odds ratio for having hypertensive nephropathy or focal segmental glomerulosclerosis.12

There is evidence suggesting that systolic BP (SBP) is the most damaging component of the BP load, both in vitro and in vivo. A patient‐level meta‐analysis on nondiabetics by Jafar and colleagues13 showed that a SBP of 110 mm Hg to 129 mm Hg and urine protein excretion <2.0 g/d were associated with lowest risk for CKD progression, and that after adjustment for SBP, diastolic pressure was not a risk factor. Elevations in SBP have been found to exhibit the closest correlation with hypertensive kidney injury in a wide variety of studies, which is likely related to the kinetics of pressure‐induced vasoconstriction and vasodilation and the increased delay in onset of relaxation after a pressure change compared with the short delay in onset of relaxation that normally occurs, making it more difficult to resume the steady‐state of the regulatory response.14

The Eighth Joint National Committee Report (JNC 8) from the 2014 Evidence‐Based Guideline for the Management of High Blood Pressure in Adults suggests that patients with diabetic and nondiabetic CKD have a goal BP <140/90 mm Hg.15 In adults older than 60 years, the recommended BP target is higher at <150/90 mm Hg. These are evidence‐based targets as there are limited clinical trial data showing the benefit of lower BP targets on renal disease progression in either diabetic or nondiabetic patients. However, these recommendations have been contested by multiple associations interested in treating hypertension, including the American Heart Association and American College of Cardiology. In particular, the goal BP of <150/90 mm Hg has caused contention, including criticism from five of the 18 original JNC 8 members.16 Secondary analyses of several clinical trials suggest that in patients with proteinuria >1 g/d, lower SBP goals may help slow progression of renal disease.

Diabetes Mellitus

Diabetes mellitus (DM) has long been known to be a cause and risk factor for CKD progression. Diabetic nephropathy, a result of chronic and usually poorly controlled diabetes, is characterized by mesangial expansion, thickening in the glomerulus, and progressive nodular glomerulosclerosis, where tubular epithelial cells are heavily deposited with glycogen and extracellular matrix.17 Poor diabetic glycemic control may accelerate the progression of CKD.

There are much data on preventing CKD progression via glycemic control in patients with type 1 diabetes. The Diabetes Control Complications Trial (DCCT) randomized nonhypertensive type 1 diabetics with normal kidney function to an intensive vs conventional strategy (achieved hemoglobin A_1c 7.2% compared with 9.1%) in a primary prevention cohort (patients with diabetes <5 years, no retinopathy, and albumin excretion ratio <40 mg/d) and secondary prevention cohort (diabetes duration 1–15 years, >1 microaneurysm, and AER >200 mg/d) for a mean duration of 6.5 years (total of 14). Intensive glycemic control resulted in decreased incident microalbuminuria in the primary prevention group and decreased overt nephropathy in the secondary prevention group. These patients were subsequently followed in the Epidemiology of Diabetes Interventions (EDIC) trial,18 where glycemic control eventually became nearly identical between the two study groups to a goal hemoglobin A_1c <8%. However, incident microalbuminuria and overt proteinuria continued to occur at a higher rate in the conventional control arm, with rate of decline of eGFR significantly greater among individuals with past or current macroalbuminuria (5.7% per year, or 44% in 10 years) compared with individuals with normal AER (1.2% per year) or microalbuminuria (1.8% per year).19 This suggested an ongoing benefit or “legacy effect” of strict glycemic control on albuminuria progression and GFR loss after conversion to conventional glycemic control. There was also a decreased risk of hypertension in the intensive arm, and a trend toward reduced rate of doubling of serum creatinine. However, in this cohort, there were some patients with low GFR levels without albuminuria whose renal biopsy results were indistinguishable from patients with similar decreases in GFR and elevated urinary AER. These biopsy findings have been replicated in other studies with similar patients,20 suggesting that there are other factors contributing to diabetic nephropathy besides albuminuria—an idea that warrants future investigative research.

There is also a robust number of data from randomized control trials (RCTs) evaluating the effect of glycemic control on the progression of renal disease in patients with type 2 diabetes. A large RCT known as the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE),21 conducted at 215 collaborating centers outside the United States investigated intensive blood glucose control vs standard glucose control in patients with type 2 diabetes mellitus who were matched by demographics and disease burden. Followed for a median of 5 years, the difference in mean hemoglobin A_1c was statistically significant: 6.5% in the intensive arm compared with 7.3% in the control arm. This resulted in a 21% reduction in nephropathy and a 10% relative reduction in the combined outcome of major macrovascular and microvascular events. Hazard ratios comparing the two arms showed the strongest trends toward a benefit from the intensive arm in specific patient populations, including patients younger than 65 years, with BMI <28 and no history of macrovascular disease, suggesting that early initiation of intensive therapy provides the most benefit in preventing or delaying diabetic nephropathy. However, at 5‐year follow‐up, there was no effect on doubling of serum creatinine. The benefits were not without some risk, as the incidence of severe hypoglycemia increased from 1.5% in the control arm to 2.7% in the intensive arm.

A similar trial, the Veterans Affairs Diabetes Trial (VADT),22 conducted at veterans affairs centers throughout the United States, found that intense hemoglobin A_1c control to <6% when compared with <9% was associated with less risk of progression of normal AER at baseline to either microalbuminuria or macroalbuminuria, but did not impact progression to CKD. The VADT trial also showed similar results regarding progression from normoalbuminuria to microalbuminuria or macroalbuminuria in the strict glycemic control group. However, similar to the ADVANCE trial, there was no difference in doubling of serum creatinine or progression to eGFR <50 mL/min/1.73 m². The collective results of these studies, while not showing a significant benefit for strict glycemic control in preventing progression of CKD, may be attributable to an insufficient follow‐up time, as it may take up to 10 years for this legacy effect to occur from more intensive therapy.

The Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, another RCT in patients with type 2 diabetes, enrolled more than 10,000 patients to either an “intensive” hemoglobin A_1c target <6.0% vs a “standard” target of 7.0% to 7.9%. It was stopped early because of increased mortality in the intensive treatment group. Six of the secondary measures at study end favored intensive therapy, including a trend toward decreased risk of incident microalbuminuria and progression to macroalbuminuria, although there was no effect on doubling of serum creatinine, reduction of eGFR <20 mL/min/1.73 m², or new ESRD.23

Overall, these studies show the importance of glycemic control on reducing incident microalbuminuria and macroalbuminuria, which is now seen as an independent predictor of progression of renal disease. The beneficial effects of intensive control seem to be most evident in specific groups of patients, including those with long‐term disease or concurrent high cardiovascular risk (Figure 3).

Figure 3.

Outcomes of major trials assessing strict glycemic control vs traditional glycemic control in patients with chronic kidney disease (CKD). CVD indicates chronic kidney disease. See text for trial expansions.

Demographic Factors

The literature has demonstrated that ESRD disproportionately affects the poor and ethnic minorities, especially African Americans. A large study by Choi and colleagues24 examined a veterans affairs health system database of more than 2 million patients and showed a higher rate of GFR loss among African Americans at any stage of CKD that was not attributable to a survival advantage among patients of that race. A cohort study by Hall and colleagues25 involving 15,000 patients with CKD stage 3 or 4 evaluated time to ESRD or death. After adjusting for sociodemographic factors such as income, housing, type of insurance, disability, comorbid conditions, and laboratory abnormalities, they found that nonwhite race was independently associated with progression to ESRD. Interestingly, among all racial‐ethnic groups in the study, the rates of progression to ESRD peaked in the youngest age group of 20 to 39 years. Hispanics, the fastest‐growing minority in the United States, also have a higher incidence of ESRD than non‐Hispanic whites and are at increased risk for renal failure. It has been suggested that high rates of diabetes and metabolic syndrome are contributing to the burden of disease, as both disproportionately affect the Hispanic population.

Previously published data from a nationally representative sample of Americans from the National Health and Nutrition Examination Survey (NHANES) found that CKD stages 3 to 5 were relatively uncommon in younger persons; however, in the Hall study one half of persons with CKD stage 3 to 5 receiving ambulatory care were younger than 60 years and more than one fourth were younger than 50 years.26 Clearly, it will be important to develop meaningful screening and management programs for these patients living in poor, urban, economically disadvantaged areas, as their disease burden appears to occur at an earlier age. The group of patients in the age group younger than 50 years who appear most likely to be referred for care are male private insurance holders, who collectively appear to have the lowest rate of CKD progression among peers their age. What can explain the increased loss of GFR over time in African Americans? Is it socioeconomic factors or barriers to care? As mentioned earlier, ApoL1 G1 and G2 allele expression has greatly increased prevalence in African Americans and is associated with hypertensive ESRD and focal segmental glomerulosclerosis.

Acute Kidney Injury

Acute kidney injury (AKI) may be another risk factor for CKD progression, especially if there are repeated episodes. Many times, patients have full renal recovery to their baseline creatinine. How are these insults related to risk of developing CKD? Ishari and colleagues27 performed a large study of more than 200,000 patients without a history of AKI or ESRD to investigate this question. The results showed adjusted hazard ratios of 8.4 for developing ESRD in patients with CKD without AKI, 13.0 with AKI but without previous CKD, and 41.2 for patients with both AKI and history of CKD, suggesting that preventing AKI is as important as or more important than preventing early stage CKD for developing ESRD. A study conducted by Wald and colleagues28 in Ontario, Canada, showed that AKI in older adults who required brief dialysis resulted in a hazard ratio of 3.23 for developing a chronic dialysis requirement.

What about repeated kidney injury? One study of 1822 VA patients with diabetes and at least one episode of AKI during hospitalization showed that risk of CKD increases with repeated AKI episodes, and each episode of AKI (up to 3 total) was associated with a doubling in risk of stage 4 CKD after adjustment for baseline creatinine.28

Nephrolithiasis

One study by Alexander and colleagues29 examined the association of new kidney stones with CKD progression in a registry cohort, excluding patients with pyelonephritis. It found that having one or more stones was associated with a two‐fold increased risk of ESRD, new stage 3b to 5 CKD, and doubling of serum creatinine compared with patients who did not form stones during 11 years of follow‐up. Nephrolithiasis appears to put patients at substantially higher risk for developing and progressing CKD. Mechanisms by which nephrolithiasis contributes to CKD are not clear but include medication use, stone type, urinary tract infections, recurrent obstruction, or damage from chronic crystalluria.

Metabolic Acidosis

Metabolic acidosis is often seen in patients with CKD. As GFR declines, so does the kidney's ability to regulate acid/base status. There is also a bone‐buffering effect caused by acidosis that results in the release of calcium and phosphate from bone and osteodystrophy.30 Correction of metabolic acidosis appears to improve renal osteodystrophy and delay its progression, as it has been demonstrated that bone parameters such as osteoid volume, osteoid surface, and mineralizing surface are improved with correction of the acidosis.31 Acidosis is particularly deleterious because of its contributions to protein‐energy wasting disorder, increased protein catabolism, excessive oxidation of amino acids, and reduced albumin synthesis. It also induces secretion of endothelin, which, in turn, is thought to mediate kidney injury.32 Metabolic acidosis, when left untreated, likely contributes to progression from early to later stages of CKD, according to multiple small single center trials.33 A study of 134 patients by Brito‐Ashurst and colleagues34 in 2009 found a slower decline in creatinine clearance, decreased progression to ESRD, and improved nutritional parameters with bicarbonate supplementation as compared with placebo, even when stratified according to medication usage, baseline characteristics, and concurrent disease burden. Importantly, these effects occurred without affecting BP or proteinuria levels. Proposed mechanisms for these substantial benefits include preventing maladaptive compensatory changes in tubular function by decreasing ammonia production and subsequent complement cascade activation in remnant tubules. Another single‐center trial by Phisitkul and colleagues35 involving 59 patients with hypertensive nephropathy and metabolic acidosis measured urine endothelin‐1 excretion and N‐acetyl‐B‐D‐glucosaminidase, markers of tubulointerstitial injury, in patients treated with sodium citrate for 24 months vs no alkaline therapy. Levels of the two markers of renal injury were significantly lower in the sodium citrate group, and rate of eGFR decline was significantly slower. While these are both small single‐center trials and did not target the same bicarbonate baseline for their treatment arm, benefits were most pronounced when maintaining patient serum bicarbonate levels at 20 mmol/L and 22 mmol/L.

Dietary Factors

The appropriate diet for patients with CKD remains controversial. Sodium intake, in particular, is a controversial topic of discussion. Traditional thinking has been that sodium restriction has beneficial effects on hypertension, proteinuria, and CKD progression. A study by Slagman and colleagues36 analyzed the effects dietary sodium restriction had on reducing proteinuria when combined with ACE inhibitor therapy compared with combined ARB and ACE inhibitor therapy during four separate 6‐week periods. In patients with a mean protein excretion of 1.68 g/d at baseline, the observed reduction in proteinuria on a low‐sodium diet added to ACE inhibitor was 51% compared with 21% with ARB and ACE inhibitor combined therapy (P < .001). There was also a 7% reduction in SBP when adding a low‐sodium diet to an ACE inhibitor compared with a 2% reduction on ARB/ACE inhibitor therapy. A post hoc analysis of the Ramipril Efficacy in Nephropathy (REIN) trials performed by Vegter and colleagues37 stratified nondiabetic CKD patients according to serial 24‐hour urinary sodium/creatinine excretion as low‐ (<100 mEq/g), medium‐ (100–200 mEq/g), and high‐ (>200 mEq/g) intake groups. They monitored progression to CKD during 4.25 years. The incidence of ESRD was 6.1, 7.9, and 18.2 per 100 patient years in the low‐, medium‐, and high‐ sodium intake groups, respectively. Furthermore, patients in the high‐intake group showed less antiproteinuric effects when treated with ACE inhibitors compared with the other groups.37 The increased progression to ESRD was thought to be the result of the effect of increased sodium intake leading to increased proteinuria. However, not all studies have found that sodium intake correlates with negative outcomes.38 In hypertensive CKD patients, especially those with proteinuria, it is currently recommended to ingest <2300 mg of sodium per day. It is also recommended in CKD patients without oliguria, to maintain a fluid intake of 2 L to 3 L daily. Multiple studies have shown this large volume of fluid intake to increase urine volume and correlate with slower progression of CKD. This, however, is most evident when comparing patients who make more than 1 L/d of urine with those producing <1 L/d. Presently, it is unknown whether dietary protein restriction slows progression of CKD, as the only study to specifically assess this was the MDRD study, which did not show a benefit in CKD progression. However, this study was conducted in patients with advanced, nondiabetic CKD, and it is possible that long‐term protein restriction in earlier stages of CKD may provide some benefit.

Hyperuricemia

The relationship between levels of uric acid and CKD remains complex. What is known is the association between hyperuricemia and cardiovascular disease. In a 2008 article, Feig and colleagues39 noted that hyperuricemia occurs as a result of decreasing renal function, but also postulated that elevated uric acid levels are associated with worsening renal function. The argument was based on lesions seen on kidney biopsies in patients with gout; advanced arteriolosclerosis, glomerulosclerosis, with urate crystals deposited in the outer medulla. There is also an increased risk of developing hypertension within 5 years in patients with hyperuricemia, which can accelerate functional renal decline.39 At the cellular level, uric acid has been shown to induce cellular proliferation, inflammation, oxidative stress, and activation of the local renin‐angiotensin system, which has been theorized to cause vascular oxidative injury and fibrosis in the kidneys. There are also epidemiologic studies suggesting that uric acid has a role in causing renal disease.40 Hyperuricemia is an independent predictor for the development of microalbuminuria and renal dysfunction in persons with normal renal function.41 Hyperuricemia is also associated with impaired GFR in nonproteinuric patients with type 1 diabetes mellitus.42

Does this mean that treating hyperuricemia can slow the progression of CKD? An RCT by Siu and colleagues43 involving 54 patients with stage 3 or 4 CKD and hyperuricemia with baseline creatinine 1.6 to 1.9 found delayed progression of CKD, with a trend toward a lower serum creatinine level in the treatment group receiving allopurinol 100 mg or 300 mg daily added to their medication regimen, compared with controls receiving no allopurinol, after 12 months of therapy (creatinine 1.99 mg/dL vs 2.89 mg/dL at the end of the trial). Although it did not reach statistical significance (P = .08), this difference represented a 20% increased rate of creatinine in the treatment group compared with a > 50% increase of creatinine in the control group. These effects were independent of any statistically significant change in SBP.43 A larger and longer (−2 year) prospective RCT by Goicoechea and colleagues44 randomized 113 patients with eGFR <60 mL/min/1.73 m² and baseline uric acid 7.3 mg/dL to 7.9 mg/dL to receive allopurinol 100 mg/d or usual therapy and observe its effect on eGFR. GFR increased by 1.3 mL/min/1.73 m² in the allopurinol‐treated group, while GFR decreased by 3.3 mL/min/1.73 m² in the control group. While the results of both epidemiologic studies and small prospective trials are promising, there is clearly a need for larger randomized trials assessing the potential renoprotective benefit of allopurinol in: (1) early stages of CKD, (2) across multiple populations, and (3) over a longer duration of therapy. This should occur before allopurinol can be validated as an accepted therapy in hyperuricemic patients with CKD.45

Interventions to Slow Progression of CKD

Blood Pressure

The evidence on hypertension and its relationship with loss of renal function is compelling. Likewise, there is abundant evidence of the use of ACE inhibitors and ARBs in patients with diabetic nephropathy. In conjunction with lowering SBP to 140 mm Hg, a 20% relative risk reduction in the doubling of serum creatinine or progression to ESRD was well demonstrated in the Irbesartan Diabetic Nephropathy Trial (IDNT)46 and the Reduction in Endpoints in Patients With Non‐Insulin Dependent Diabetes Mellitus With the Angiotensin Antagonist Losartan (RENAAL) trial.47 These two studies collectively found that in patients with type 2 diabetes and an eGFR 50 mL/min/1.73 m² to 60 mL/min/1.73 m² with >1 g/24 h proteinuria, achieving an SBP of 140 mm Hg with ARB therapy slowed progression of CKD and incident ESRD (Figure 4).

Figure 4.

Incidence of diabetic nephropathy in the Irbesartan Diabetic Nephropathy trial.

However, there are no studies in diabetic patients with CKD that examine the benefit of lower BP goals on renal disease progression. In patients with nondiabetic CKD, there are only three RCTs that have evaluated patients with nondiabetic nephropathy to two BP goals. These were the MDRD trial in 1993, Ramipril Efficacy and Nephropathy (REIN‐2) trial in 2005, and the African‐American Study of Kidney Disease and Hypertension (AASK) trial in 2002. The MDRD trial randomized 840 patients with a measured GFR 25 mL/min/1.73 m² to 55 mL/min/1.73 m² to a BP goal of either <140/90 mm Hg or <125/75 mm Hg. After a mean of 2.2 years of follow‐up, the group with the lower BP goal had a nonstatistically significant reduction in the rate of GFR loss compared with patients with the traditional BP goal. While at 10‐year follow‐up, there was a 32% reduction in the incidence of ESRD in the lower pressure group compared with the traditional group, there were limited BP readings obtained in the later years of the study.48 Other endpoints such as doubling of serum creatinine, ESRD, and death were not examined, and confounding variables such as ACE inhibitor use (48% in the treatment group and 28% in the control group) were not controlled for. Despite these shortcomings, it was interesting that the benefit was most pronounced in proteinuric patients with >1 g of protein excreted per day.

The REIN trial demonstrated that use of ramipril in patients with chronic, nondiabetic, proteinuric CKD slowed the rate of CKD progression independent of the resultant BP.49 The REIN‐2 trial randomized 338 patients with nondiabetic kidney disease with a mean GFR of 34.1 mL/min.1.73 m² and proteinuria of ≥3 g/d to intensive BP control <130/80 mm Hg vs conventional control, with the intensified BP control achieved with the addition of calcium channel blockers. Again, there was no significant difference in the endpoint of ESRD in either study group.50

Finally, the AASK trial randomized 1094 African Americans with hypertensive nephrosclerosis to either a conventional (140/98 mm Hg) vs an intensive (125/75 mm Hg) goal using β‐blockade, calcium channel blockade, or ACE inhibitor therapy as initial treatment. The intensive group did not show a significant reduction in endpoints (50% decrease in GFR, ESRD, or death) compared with the conventional group, independent of the initial treatment option.51 More recently, a systematic review that included the MDRD, REIN‐2, and AASK trials in more than 2300 patients with CKD found that a BP target of <125–130/75–80 mm Hg did not prevent CKD progression when compared with a target of <140/90 mm Hg. The ongoing Systolic Blood Pressure Intervention Trial (SPRINT), a two‐arm, National Heart, Lung, and Blood Institute multicenter RCT, has enrolled 10,000 nondiabetic hypertensive patients older than 50 years, one third of whom have underlying CKD, to determine whether an SBP goal of <120 mm Hg improves cardiovascular and renal outcomes compared with an SBP goal of <140 mm Hg.52

Thus, it seems clear that in patients with CKD and hypertension, renin‐angiotensin system blockade should occur, especially in patients who have not progressed to ESRD. What other medication strategies are effective at reducing BPs and slowing progression of kidney disease? The pathophysiology of hypertension in CKD patients is primarily caused by positive sodium balance and expanded plasma volume, and use of loop diuretics is often necessary to control BP. A 2003 study demonstrated that single daily‐dose torsemide or furosemide for 3 weeks reduced extracellular water by 1.7 L, and systolic BP was reduced from an average of 147 mm Hg to 138 mm Hg in patients taking furosemide (P = .021) and from 143 mm Hg to 133 mm Hg in patients taking torsemide (P = .007) compared with patients not taking loop diuretics.53 But what about patients with stage 3 CKD who require only one additional agent to an ACE inhibitor or ARB? Weir and colleagues54 reported on renal outcomes when combining benazepril plus amlodipine vs benazepril plus HCTZ in hypertensive black (1414) and nonblack (6711) patients in the Avoiding Cardiovascular Events Through Combination Therapy in Patients Living With Systolic Hypertension (ACCOMPLISH) trial. Terminated at 2.9 years because of significantly better cardiovascular outcomes in the ACE inhibitor/calcium channel blocker group, decreased progression of CKD was noted in the benazepril plus amlodipine group in nonblack patients. Of note, the composite endpoint of doubling of serum creatinine, ESRD, or death was not different between blacks and nonblacks; however, black patients were more likely to develop a > 50% increase in serum creatinine to >2.6 mg/dL.54 Whether amlodipine provides a greater advantage for renal disease progression over HCTZ remains unclear, especially in black patients. Lastly, a systematic review of 85 RCTs including nearly 22,000 patients showed that the combination of ACE inhibitors and ARBs demonstrated no benefit in preventing progression of albuminuria or development of ESRD compared with either therapy alone and had a higher risk of adverse events including cough and hypotension.55

Glycemic Control

While the bulk of evidence suggests that strict glycemic control may prevent progression of renal disease, there is a lack of long‐term evidence in patients with type 2 diabetes, as larger RCTs with longer enrollment periods are necessary to fully understand the effect of tight glucose control on CKD. Because there is always a higher risk of hypoglycemic events in patients with stricter glucose control, this risk must be balanced with the potential benefits a strict glucose control strategy might yield, not just for renal outcomes, but cardiovascular events, cognitive decline, and overall quality of life. The most important interventions in patients with type 2 diabetes will always be lifestyle modifications such as diet, exercise, and weight control. Primary care clinics have a multitude of resources to support patients in these interventions, and they should be aggressively pursued in all diabetic patients, especially those who have not yet been affected by the diseases’ more serious long‐term effects. Oral agents such as sulfonylureas (glipizide, repaglinide) appear to be safe in patients with CKD because of their short half‐life. Metformin is a valuable oral agent early in the disease course, but it should be avoided in CKD patients with eGFR <30 mL/min/1.73 m² as its use can result in lactic acidosis. Dipeptidyl peptidase‐IV inhibitors (gliptins) do not have a sufficient amount of evidence to recommend their use in CKD, although some data suggest that they are safe with dosage reduction in patients with CKD and ESRD.56 Sodium glucose cotransporter 2 (SGLT‐2) inhibitors are a novel therapy that may be beneficial and safe in patients with CKD, with a GFR >45 mL/min/1.73 m². Canagliflozin is an SGLT‐2 inhibitor undergoing phase 3 trials. A recent double‐blinded placebo‐controlled trial measured the change from baseline hemoglobin A_1c in diabetic patients randomized to canagliflozin or placebo. A higher proportion of the treatment arm reached a hemoglobin A_1c target of <7% compared with placebo, with similar rates of adverse events in each group.57 SGLT‐2 inhibitors may also have a modest antihypertensive effect, likely caused by a variety of mechanisms including natriuresis, osmotic dieresis, and protection against renal remodeling. Interestingly, this effect has been demonstrated in patients with either normal or reduced GFR and high or low levels of glycosuria.58 Exogenous insulin is often used when oral therapy has failed or earlier in treatment, depending on the level of the initial A_1c level. There are a multitude of insulin treatment regimens to choose from that may require direction and close follow‐up from an endocrinologist. Almost all diabetic patients with ESRD on hemodialysis require insulin therapy, as the majority of the oral agents are renally cleared.59

Coronary Artery Disease Treatment

Various studies have shown that patients with nondialysis‐dependent CKD have a cardiovascular risk profile similar to those with known coronary artery disease, and patients receiving hemodialysis have a 40 to 50 times greater cardiovascular event risk than persons in the non‐CKD population of the same age and sex. The Study of Heart and Renal Protection (SHARP) trial randomized 9270 CKD patients (3023 dialysis, 6247 nondialysis) with no known history of myocardial infarction or coronary revascularization to receive either simvastatin 20 mg plus ezetimibe 10 mg vs placebo for a median follow‐up of 4.9 years. The lipid‐lowering group had a statistically significant reduction in major atherosclerotic events, including nonfatal myocardial infarction or coronary death, nonhemorrhagic stroke, or any arterial revascularization procedure, compared with the placebo group (11.3% vs 13.4%, or 17% proportional reduction; P = .0021), with similar results in dialysis and nondialysis patients.60 The trial showed no change in progression of CKD in the treatment or control groups; however, it was not powered to examine this question. A subsequent meta‐analysis from 2012 by Palmer and colleagues61 evaluated 80 trials with 51,099 participants who received either statin or placebo for a median of 5 years of treatment and showed with moderate‐ to high‐quality evidence that statin treatment reduced all‐cause mortality, cardiovascular mortality, and cardiovascular events by about 20% to 25% in patients who were not dialysis‐dependent, and had little to no effect on persons receiving dialysis. This suggests that the benefits of statin therapy for mortality and cardiovascular outcomes may differ depending on the stage of CKD. The reasoning behind why such a difference exists among patients with different stages of CKD is currently unknown but may be the result of a different pathobiology in those with advanced CKD and preexisting cardiovascular disease, dominated by vascular calcification, cardiac hypertrophy, and arterial stiffening. Once these lesions occur, lipid‐lowering therapy may have less effect on patient outcomes. These trials do, however, show a substantial cardiovascular benefit of using statins in patients with early‐stage CKD. Multiple trials have also been conducted to detect a possible antihypertensive effect of statin therapy. A recent meta‐analysis of prospective controlled studies showed a small, statistically significant reduction of SBP in patients taking statins,62 suggesting a pleiotropic effect of the drug class.

Metabolic Bone Disease Treatment

Fibroblast growth factor 23 (FGF‐23) is a protein responsible for regulating phosphorus concentrations in plasma that is secreted by osteocytes in response to a decreased GFR. Increased concentrations of FGF‐23 cause decreased expression of NPT2, a sodium‐phosphate cotransporter in the renal proximal tubule, resulting in decreased reabsorption and increased excretion of phosphorus. From an epidemiological standpoint, patients with chronically high phosphorus levels are at higher risk for cardiovascular disease and vascular calcinosis.63 In CKD patients, FGF‐23 levels increase early in the disease to stimulate phosphaturia and maintain serum phosphorus levels (Figure 5). While there is evidence that higher serum levels of FGF‐23 correlate with increased CKD progression,64 FGF‐23 may also cause direct kidney injury. Two mechanisms have been proposed: the peptide may reduce 1,25(OH)₂ vitamin D levels. It may also cause left ventricular hypertrophy through growth‐related mechanisms.65 A large cohort study comprising 3800 patients with stage 2 to 4 CKD examined the association between FGF‐23 levels and ESRD progression or mortality. After adjusting for eGFR, demographic characteristics, use of phosphate binders, vitamin D supplementation, and CKD‐specific risk factors, there was a relationship between FGF‐23 levels and risk of death across all stages of CKD. However, it was associated with increased risk of ESRD progression only in patients with eGFR >30 mL/min/1.73 m². At more advanced stages of CKD, the eGFR itself was the primary determinant of ESRD progression, independent of FGF‐23 level.66 It is unknown whether reducing FGF‐23 would be beneficial on kidney disease progression since it would cause serum phosphate and calcium levels to further increase. A recent meta‐analysis comparing elevated serum phosphate levels in patients with CKD found an independent association between hypophosphatemia and mortality.67

Figure 5.

Fibroblast growth factor 23 (FGF‐23), parathyroid hormone (PTH) secretion, and serum calcium in chronic kidney disease (CKD). GFR indicates glomerular filtration rate; LVH, left ventricular hypertrophy.

Deficiency of vitamin D has previously been postulated as a risk factor for CKD progression, but despite theorized benefits including blocking the renin‐angiotensin system, anti‐inflammatory effects, and reducing podocyte injury, no large trial has yet shown a decreased rate of progression to ESRD in patients using vitamin D analogues. However, observational studies demonstrate decreased mortality in CKD patients receiving concurrent phosphate binder and vitamin D therapy, primarily through fewer cardiovascular events.68 These results are independent of phosphorus levels and parathyroid hormone levels. Some experts recommend initiating these therapies before phosphorus and parathyroid hormone levels are pathologically altered, for a potentially greater benefit. However, the data used to recommend these interventions so far is mostly from uncontrolled cohort studies vulnerable to confounding. A large placebo‐controlled randomized trial of vitamin D, calcimimetic agents, and phosphate binders is needed to provide guidelines for treatment strategies in CKD‐mineral bone disease.69

Anemia Management

Treatment with erythropoiesis‐stimulating agents (ESAs) to treat anemia of CKD was previously thought to slow progression of CKD before clinical trial data were available to assess these outcomes. The mechanisms of anemia in CKD are multifactorial and include decreased renal synthesis of erythropoietin, chronic inflammation, and hyperuricemia causing bone marrow suppression.70 It is now known that hemoglobin target levels >12.0 are generally thought to be too high to warrant treatment with ESAs in CKD patients. The Cardiovascular Risk Reduction by Early Anemia Treatment With Epoetin Beta (CREATE) study randomized 603 CKD patients with anemia to a higher (13.0–15.0) vs lower hemoglobin target group (10.5–11.5). They observed that the higher target group progressed to dialysis and ESRD more rapidly.71 Another large trial, the Trial to Reduce Cardiovascular Events With Aranesp Therapy (TREAT), showed no difference in preventing ESRD progression in patients randomized to hemoglobin target levels of 13 g/dL with darbepoeitin or placebo and no difference in mortality.72 The Correction of Hemoglobin and Outcomes in Renal Insufficiency (CHOIR) trial was performed at 130 US medical centers and randomized 1432 patients with CKD and anemia to a target hemoglobin of 13.5 g/dL vs 11.3 g/dL. It found a nonsignificant trend toward a higher risk of requiring dialysis for ESRD in the higher target arm. In addition, the trial was terminated early after only 16 months because of a 48% higher risk of death in the high‐target hemoglobin arm. Importantly, post hoc analysis of the CHOIR trial by Inrig and colleagues73 suggested that a higher hemoglobin level may be an independent predictor for progression of kidney disease in patients with CKD treated for anemia (Table). The CHOIR study also linked more adverse outcomes with current smoking. Although this association between smokers and poor outcomes with ESA administration for anemia of CKD has not yet been explained, a molecular mechanism has been hypothesized: smoking has been shown to decrease endothelial nitric oxide (NO) synthase (eNOS) activity in mice, an enzyme that increases levels of NO. During renal vascular injury, eNOS levels are low as a result of endothelial damage to the vessels. Endothelial behavior is known to be highly dependent on NO activity. Since it has been shown that erythropoietin administration enhances medial thickening of injured carotid arteries in eNOS‐deficient mice, it is possible that the decrease in eNOS activity caused by smoking would result in erythropoietin producing detrimental effects directly on “injured” blood vessels in the kidney via ESA‐induced vasoconstriction.73

Table 1.

Comparison of Three Major Trials of Anemia in Chronic Kidney Disease

Trial Name	ESA	Hemoglobin Targets	Study Population	Primary Outcomes	Primary Outcome Results	Progression to ESRD
CREATE	Erythropoietin beta	Higher target 13.0–15.0 g/dL vs lower target 10.5–11.5 g/dL	603 patients with hemoglobin 11–12.5 g/dL, eGFR 15–35 mL/min/1.73 m2	Sudden death, MI, acute heart failure, CVA, TIA, arrhythmia	No change in likelihood of cardiac event among groups (HR, 0.78; P=.20)	No change in eGFR between groups, slightly more patients in high hemoglobin target group required HD initiation (127 vs 111; P=03)
CHOIR	Erythropoietin alfa	Higher target 13.0–13.5 g/dL, lower target 10.5–11.0 g/dL	1432 patients with hemoglobin <11.0, eGFR 15–50 mL/min/1.73 m2	Time to death or MI, hospitalization for CHF, stroke	Higher risk of death or MI in higher target g roup compared with lower target group (17.5% vs 13.5%, HR, 1.34; CI, 1.03–1.74)	Higher incidence of renal replacement therapy in higher hemoglobin target arm (21.7% vs 18.7%, nonsignificant)
TREAT	Darbopoetin alfa	Higher target 13.0 g/dL, lower target 9 g/dL	4038 patients with type 2 diabetes mellitus, eGFR 20–60 mL/min/1.73 m2, hemoglobin <11.0 g/dL, transferrin sat >15%	Composite outcomes of death or CV event, and death or ESRD progression	Death or nonfatal CV event nonsignificantly elevated in higher target group, death or ESRD nonsignificantly elevated in higher target group (P=.29)	ESRD initiation unchanged between groups

Abbreviations: CHF, congestive heart failure; CHOIR, Correction of Hemoglobin and Outcomes in Renal Insufficiency trial; CI, confidence interval; CREATE, Cardiovascular Risk Reduction by Early Anemia Treatment With Epoetin Beta trial; CV, cardiovascular; CVA, cardiovascular accident; eGFR, estimated glomerular filtration rate; ESA, erythropoiesis‐stimulating agent; ESRD, end‐stage renal disease; HD, hemodialysis; HR, hazard ratio; MI, myocardial infarction; TIA, transient ischemic attack; TREAT, Trial to Reduce Cardiovascular Events With Aranesp Therapy.

Newer Therapeutic Opportunities

Despite the vast array of modifiable risk factors, preventing CKD progression remains difficult. Barriers include patient compliance, racial disparities in treatment, and lack of effective interventions. Presently, increasing attention is being paid to newer compounds that may provide incremental benefits to patients with CKD when combined with established therapies. While the compounds do not all have the same molecular mechanism, in general their function is based on anti‐inflammatory and antioxidative effects.

Pentoxifylline (PTF) is a potential therapeutic agent in CKD because of its anti‐inflammatory and antiproteinuric effects. A study by Perkins and colleagues74 in 2009 documented slower progression of CKD when taking PTF vs placebo for 1 year, with the PTF treatment group showing an eGFR decline of −1.2 mL/min/1.73 m² compared with −7.2 mL/min/1.73 m² in the control group.74 In 2012, Goicoechea and colleagues75 conducted a prospective RCT in 91 patients with eGFR <60 mL/min/1.73 m² to receive PTF or placebo and found that high‐sensitivity C‐reactive protein, serum fibrinogen, and tumor necrosis factor α (TNF‐α) decreased significantly in patients taking PTF. Renal function did not change significantly in the PTF group after 12 months and eGFR decreased by an average of 6 mL/min/1.73 m² to 7 mL/min/1.73 m² in the control group.75 While these studies are promising, they were small cohorts with a short duration of follow‐up. A newer study by Chen and colleagues,76 a single‐center retrospective analysis examining the combination of PTF with ACE inhibitors or ARBs in patients with advanced CKD, found no change in overall mortality or risk of incident cardiovascular events. The addition of PTF, however, showed better renal outcomes, demonstrating a 40% lower risk of developing ESRD than treatment with an ACE inhibitor or ARB alone, especially in patients with large amounts of proteinuria.76

Pirfenidone is an antifibrotic and anti‐inflammatory compound that was initially developed and recently approved by the US Food and Drug Administration for patients with interstitial pulmonary fibrosis. It inhibits transforming growth factor α and free radical oxygen species, as well as reduces interleukin 1 and TNF‐α levels. Pirfenidone has been studied in clinical trials and has shown some promise in treating cirrhosis and pulmonary fibrosis.77 Experimental data from RamachandraRao and colleagues78 in 2009 showed that pirfenidone inhibited mesangial matrix expansion and reduced levels of type I collagen, type IV collagen, and fibronectin gene expansion in kidneys of mice with diabetic nephropathy. This likely occurs via inhibiting phosphorylation of eIF4E, which prevents mRNA processing and thus expression of fibrotic proteins. A study on nephrectomized rats by Chen and colleagues79 in 2012 found that pirfenidone inhibits M1 and M2 macrophage infiltration in 5/6 of the rats. The only human trial to date was an RCT in 77 patients with diabetic nephropathy showing that after 1 year of therapy with pirfenidone in patients with eGFR ranging from 20 mL/min/1.73 m² to 75 mL/min/1.73 m² resulted in an average increase in GFR of +8.5 mL/min/1.73 m² compared with a mean eGFR decrease of −2.2 mL/min/1.73 m² in the placebo group.79 More experimental and human trials with longer study durations in wider populations need to be performed in order to eventually recommend pirfenidone for open‐label use.

Bardoxolone methyl was, until recently, a promising antioxidant and anti‐inflammatory transcription factor. A phase 2, double‐blinded RCT published in 2011 by Pergola and colleagues80 found that bardoxolone methyl was associated with improvement in eGFR in patients with advanced CKD and type 2 diabetes at 24 weeks of follow‐up. However, in 2013, an RCT by de Zeeuw and colleagues81 randomized more than 2000 patients with stage 4 CKD and diabetes to either bardoxolone methyl or placebo. It found that bardoxolone failed to reduce progression to ESRD or death, and was also associated with a higher rate of cardiovascular events (mostly incident heart failure). While the mechanism linking the drug to heart failure is unknown, the authors postulated that an increase in preload caused by volume expansion and an increase in afterload related to increased mean BP precipitated heart failure in the at‐risk population. While these results are disappointing, this study did show a trend toward preserving eGFR and slowing the rate of progression to ESRD in patients treated with bardoxolone, questioning whether there may be a place for this therapy in lower‐risk patients.

Endothelin‐1 is a vasoconstrictor that acts on receptors in the kidney vasculature and collecting system. Numerous studies have suggested that endothelin antagonists are effective in reducing albuminuria in diabetic nephropathy although fluid retention has limited their use.82 A double‐blinded RCT by Kohan and colleagues83 examined the effect of atrasentan, a selective endothelin A receptor antagonist on albuminuria in 89 patients with diabetic nephropathy. At higher doses, after 8 weeks of therapy, treatment with atrasentan resulted in a significant decrease in urinary ACR, and there were fewer cases of peripheral edema seen than in earlier trials with nonselective endothelin antagonists. A longer study that tests for endpoints involving renal function and progression to ESRD is needed to evaluate whether endothelin receptor antagonists may have a role in slowing CKD progression. Of note, a recent 2014 study by Baltatu and colleagues84 suggested that low‐dose avosentan is renal protective in transgenic rats with hypertensive nephropathy. They noted a reduction in renal damage and mortality, while not causing evidence of volume overload at the doses studied.

Conclusions

CKD affects a large population that continues to grow. Hypertension and diabetes mellitus are two risk factors with abundant data documenting their relationship with CKD progression. Treating them more intensively in an individualized fashion with close follow‐up is paramount to reduce risk of CKD progression. Statin therapy has been shown to reduce both all‐cause and cardiovascular mortality in nondialysis‐dependent CKD patients. Bone and mineral disorders, such as hypophosphatemia and vitamin D deficiency, occur as a consequence of CKD and should be treated. However, there are fewer data about how early, how intensively, and which are the preferred therapies. Anemic patients with CKD often require exogenous ESA to prevent severe anemia and transfusions, although the optimal hemoglobin level remains unclear, and these patients should always be evaluated for other possible causes of anemia. Metabolic acidosis should be recognized and treated as an independent risk factor for CKD progression. Lastly, novel agents such as pirfenidone, pentoxifylline, and others may offer a new approach to reducing progression of CKD. By identifying barriers to coordinated care, treatable risk factors, and novel molecular therapies, primary care physicians who treat patients with CKD will be able to improve outcomes in this complex disease.

J Clin Hypertens (Greenwich) 2015;17:514–527. DOI: 10.1111/jch.12560. © 2015 Wiley Periodicals, Inc.