Low Sodium Diet, Vitamin D, or Both for RAASi-Resistant, Residual, Proteinuria in CKD? The ViRTUE Trial Points the Way Forward but Is Not the Last Word

• Vitamin D
• proteinuria
• Trial

Persistent albuminuria-proteinuria (not explained by intercurrent changes in clinical status, such as fever or urinary sepsis) is regarded as a cardinal sign of real or potential kidney damage. Persistent albuminuria-proteinuria is also considered a major prognostic marker for inferior clinical outcomes, including premature cardiovascular and renal disease.¹

The association between albuminuria and increased risk of progressive loss of renal function over time is not only found in various pathophysiologic conditions, including diabetic nephropathy, hypertensive nephropathy, and various primary renal diseases such as IgA nephropathy, but also may be present in the general, otherwise healthy populations. Emerging data suggest that elevated albuminuria causes tubulointerstitial damage through activation of proinflammatory mediators, which may ultimately lead to a progressive decline in renal function. Thus, early detection of albuminuria-proteinuria should be considered a cornerstone tool for prevention and management of chronic cardiorenal conditions, including hypertension and diabetes mellitus, as well as a number of chronic GNs.²

Few pharmaceutical interventions are available to slow the rate of renal functional decline (as estimated by a decrease in GFR) in patients with kidney disease. The renin-angiotensin-aldosterone system inhibitors (RAASis) are, of course, well known in this context and applicable across the spectrum of primary and secondary renal pathologies. A hallmark feature of these types of antihypertensive drugs is their ability to reduce albuminuria-proteinuria,³ although nondihydropyridine drugs (diltiazem and verapamil) also have useful antihypertensive and antiproteinuric actions.⁴

Other potential antiproteinuric interventions have been applied (e.g., such as those tried in diabetic nephropathy, including intensive glucose control and anti-inflammatory agents [pentoxifylline]). In a wide spectrum of conditions leading to CKD, other interventions, including a low-protein diet, corticosteroids (for IgA nephropathy), and endothelin antagonists (e.g., FSGS), have been tested.⁵ No other antiproteinuric renoprotective strategy is as well known or well established as renin-angiotensin-aldosterone system (RAAS) blockade, which in many situations is simplistically equated with nephroprotection.

Experimental evidence strongly supports the biologic plausibility of proteinuria as a surrogate for ESRD, showing the relationship between accumulated protein during heavy proteinuria and progressive nephropathy. Clinical evidence also supports a dose-dependent association between the degree of proteinuria at baseline and the risk of clinically important outcomes, including cardiovascular disease and ESRD.⁶ The prognostic importance of proteinuria has been supported by evidence showing the effectiveness of BP-lowering therapies, particularly on slowing the progression of CKD in people with proteinuria at baseline.^7,8

Analyses of data from clinical trials show that the reduction in albuminuria observed during the first months of treatment with RAASis correlates with the degree of long-term renal protection: the larger the initial reduction in albuminuria, the lower the risk of ESRD during treatment. In addition, in treated patients, residual albuminuria is the strongest predictor of risk for renal disease progression. Taken together, these observations provide a strong argument that albuminuria is an appropriate therapeutic target in patients with CKD.⁹

The importance of proteinuria in CKD and the role that it plays in the subsequent development of ESRD have also identified proteinuria as a potential surrogate outcome in clinical trials; indeed, proteinuria change has been used in many small-scale, earlier-phase trials as a predictive marker for ESRD.^10,11 However, proteinuria is currently not accepted as a surrogate outcome in broader registration, large-scale, multicenter trials of diabetic nephropathy.^12,13

Whereas it is clear that RAAS blockade can often achieve useful results in slowing down the otherwise relentless progressive loss of kidney function that is a hallmark of many (although by no means all) renal conditions, it also has strict limitations and some attendant dangers if inappropriately applied or used without clinical oversight. These latter arguments have recently been extensively articulated in the particular context of dual- or triple-RAASi therapy.¹⁴ Both AKI and hyperkalemia can result from an intensified dual- or triple-RAASi blockade approach (as well as with single RAASi therapy).¹⁵

The fundamental reasons for attempting to intensify RAAS blockade seem to be twofold: the first is to ensure mitigation of putative endocrinologic escape from the action of a single drug (usually angiotensin-converting enzyme inhibitors [ACEis]) and the second is when either BP or albuminuria-proteinuria levels remain elevated (i.e., when monotherapy fails to meet optimal treatment goals). In 2017, the practice of prescribing dual- or triple-RAASi therapy has fallen into some opprobrium—but in terms of ACEi- or angiotensin receptor blocker (ARB)–resistant albuminuria-proteinuria, the need to interrupt this pathologic process is still seen as the best option for arresting renal disease progression.¹⁶

Two potential clinical interventions have recently excited considerable interest among clinical and research nephrologists in terms of augmenting the antiproteinuric potency of RAAS blockade: namely dietary salt restriction and administration of vitamin D.

A systematic review of 16 studies addressing salt intake and kidney disease set out to establish whether variations in dietary sodium consumption influence renal outcomes in people with CKD.¹⁷

Despite marked heterogeneity of the data across study populations, the review suggested that higher salt intake was associated with worsening albuminuria and an increased likelihood of GFR reduction. Although the quality of the studies included was insufficient to support the authors’ hypothesis that increased salt intake is nephrotoxic, the results were strong enough to suggest that modest dietary avoidance of salt should be encouraged in people with CKD, especially those with hypertension and/or proteinuria.

In a randomized, double-blind, placebo-controlled trial of a rather meager degree of salt reduction in 40 Afro-Caribbean subjects who were hypertensive, a salt-restricted diet (approximately 5 g daily) significantly reduced 24-hour urinary protein excretion by 19% and led to falls in systolic and diastolic BP of 8 and 3 mmHg, respectively.¹⁸

The fall in urinary protein excretion correlated with a reduction in urinary sodium excretion but did not correlate with BP reduction. Individuals with metabolic syndrome may be especially sensitive to the effects of sodium intake. Hoffman and Cubeddu¹⁹ examined the role of salt intake with increased BP in 109 subjects with metabolic syndrome. Restriction of usual salt intake from an average of 8.2 to nearly 2.3 g/d reduced the percentage of subjects who were hypertensive from 23.8% to 8.2%. In a 6-month prospective, controlled trial, 110 patients with GFR categories G4 or G5 (GFR<30 ml/min per 1.73 m²) followed a low-sodium (LS; approximately 1 g/d) diet as part of either a low-protein diet (0.6 g/kg per day) or very low–protein diet supplemented with essential amino acids (0.35 g/kg per day) or followed a protein-free diet.²⁰

BP fell significantly in the very low–protein diet group from 143±19/84±10 to 128±16/78±7 mmHg (P<0.001), despite reduction of antihypertensive drugs. The improved BP correlated with decreased urinary sodium, leading the authors to conclude that the antihypertensive effect was due to reduced salt intake independent of actual protein intake. Finally, a randomized, controlled, crossover trial in 52 nondiabetic subjects with CKD compared the effects of an LS diet (target 50 mmol [1.15 g] sodium/d) with those of a regular sodium (RS) diet (target 200 mmol [4.60 g] sodium/d) on lowering of proteinuria through RAAS blockade.²¹

The reduction of proteinuria by the addition of an LS diet to an ACEi was significantly larger (P<0.001) than that by the addition of an ARB to ACEi. Similarly, the reduction of systolic BP by the addition of the LS diet was significantly larger (P=0.003) than that by the addition of an ARB. The authors concluded that sodium restriction to a level recommended in guidelines was more effective than dual-RAAS blockade for reduction of proteinuria and BP in nondiabetic nephropathy. Overall, a low-salt diet should surely by now be accepted as having a place in the clinical management of proteinuria.

Several recent studies have highlighted the importance of vitamin D therapy in areas beyond traditional bone and mineral metabolism in humans. These endeavors have certainly looked at albuminuria-proteinuria. Several small randomized clinical trials have evaluated the effect of active vitamin D therapy on albuminuria as a marker of kidney damage. A single-center study of 61 patients showed lower urine protein-to-creatinine ratios and lower serum parathyroid hormone (PTH) concentrations in patients randomized to paricalcitol, a synthetic vitamin D receptor activator, compared with placebo.²²

Another small single-center study of 24 patients randomized to two different doses of paricalcitol or placebo showed lower high-sensitivity C-reactive protein levels and lower rates of 24-hour albumin excretion in the paricalcitol group.²³

A large, placebo-controlled, double-blinded, randomized clinical trial of two different doses of paricalcitol in 281 participants with type 2 diabetes mellitus showed similar results.²⁴ In this multicenter, multinational study, all patients had to have albuminuria and be taking an ACEi or ARB at baseline. The study population in the trial had a mean age of 64 years old, 69% of participants were men, 72% were white, 14% were black, and the median urinary albumin excretion was approximately 700 mg/24 h. The results showed that there was a significant reduction in the urinary albumin-to-creatinine ratio in participants taking the 2-mg/d dose of paricalcitol compared with placebo. This was associated with a significant and dose-dependent lowering of eGFR as estimated from serum creatinine. Twelve weeks after randomization, the eGFR was 2 ml/min per 1.73 m² lower in the participants receiving the 1-mg dose and 4 ml/min per 1.73 m² lower in participants receiving the 2-mg dose. Both resolved after the drug was stopped, suggesting a hemodynamically related fall in GFR that is not unlike changes after RAAS blockade. BP was also significantly lower in the participants randomized to the 2-mg dose by a mean of approximately 8 mmHg. This multicenter paricalcitol study is important, in that it showed a decrease in albuminuria and eGFR in patients with type 2 diabetes and nephropathy in a nicely designed and executed randomized trial. It does not, however, answer the more important question of whether this decrease in albuminuria will translate to better clinical outcomes, such as less rapid progression to dialysis. Interestingly, a recent study showed that, in a small group of patients with kidney disease, paricalcitol increased serum creatinine levels without affecting iothalamate GFR measurements, suggesting that changes in eGFR were not truly functional.²⁵ This area has been expertly reviewed by de Borst et al.²⁶

In this issue of the Journal of the American Society of Nephrology, Keyzer et al.,²⁷ forming a consortium of clinical researchers led from Groningen, The Netherlands—an academic center with a long tradition of nephrology innovation starting with the clinical observations on artificial dialysis treatment by Jan Kolff in the 1940s and a strong tradition of clinical research related to BP, albuminuria, and CKD reported on the individual and combined effects of paricalcitol and dietary sodium restriction on residual albuminuria in CKD. Their Vitamin D Receptor Activator and Sodium Restriction for Treatment of Urinary Albumin Excretion in CKD (ViRTUE-CKD) Trial was a multicenter, randomized, placebo-controlled, crossover trial in which 45 patients with nondiabetic CKD stages 1–3 and albuminuria <300 mg/24 h, despite ramipril at 10 mg/d and BP<140/90 mmHg, were treated for four 8-week periods with paricalcitol (2 mg/d) or placebo, each combined with an LS or RS diet.

In their intention to treat analysis, albuminuria (geometric mean) was 1060 mg/24 h (95% confidence interval [95% CI], 778 to 1443) during RS intake and when taking placebo versus 990 mg/24 h (95% CI, 755 to 1299) during RS intake but taking paricalcitol (P=0.20 versus RS and placebo).The combination of LS and placebo reduced albuminuria to 717 mg/24 h (95% CI, 512 to 1005; P<0.001 versus RS and placebo) compared with 683 mg/24 h (95% CI, 502 to 929) with LS and paricalcitol (P<0.001). The reduction by paricalcitol beyond the effect of LS intake was nonsignificant (P=0.60). In the per-protocol analysis restricted to participants with >95% compliance with study medication, paricalcitol was observed to provide further albuminuria reduction (marginally positive P=0.04, LS and paricalcitol versus LS and placebo). Dietary adherence in the study was good as reflected by urinary excretion of 174±64 mmol Na⁺/d in the combined RS intake groups and 108±61 mmol Na⁺/d in the LS groups (P<0.001). The investigator group concluded in their report that moderate dietary sodium restriction substantially reduced residual albuminuria during therapy with fixed-dose ACEis. The additional effect of paricalcitol on albuminuria was small and nonsignificant.

This was an interesting well conducted but challenging study. Although 212 eligible patients were invited to participate, only 45 were randomized. It is important to note that, although these patients might be thought of as renal in phenotype, their mean creatinine clearance was a robustly normal 101±41 ml/min (perhaps higher than desirable, because administration of synthetic vitamin D receptor agonists to subjects with largely normal kidney function would be unusual in the real world). Therefore, although patients with a creatinine clearance down to 30 ml/min could theoretically have been recruited, the main CKD element chosen in this study for investigation was residual albuminuria (defined as being refractory to the use of ramipril at 10 mg/d) without the loss of excretory renal function. Thus, in terms of comparing data from this study with those of the Vitamin D and Omega-3 Trial (VITAL),²⁴ which focused on the effect of paricalcitol on proteinuria, two major differences are worth noting. First, in the VITAL, the mean eGFR of subjects was 40 ml/min and the mean PTH levels were approximately 100 pg/ml (where administration of a Vitamin D analog is justified) compared with the ViRTUE-CKD Trial, in which subjects had a mean eGFR of 100 ml/min and a normal PTH. Second, all subjects in the VITAL had type 2 diabetes compared with the ViRTUE-CKD Trial, where none of the patients had diabetes.

The study design of the ViRTUE-CKD Trial was good, albeit complex; moreover, the lack of any washout period between the four different phases/interventions was potentially confounding; the lack of any washout period would only be justified if it is very clearly known that 8 weeks of treatment per drug/diet cycle is considered sufficient to eliminate the order effects of treatment. Nevertheless, good separation between the LS and normal sodium intake cohorts was achieved (as judged by urinary sodium measurements), and the normal sodium and LS diets were both representative and potentially achievable in clinical practice.

The most important findings from the study are that the deployment of clinically relevant dietary sodium restriction led to a reduction in BP, eGFR, and albuminuria—acting exactly as expected for the RAASis—and in this context, no further important effect on albuminuria was achieved by adding paricalcitol at a dose strength proven to be biologically active in the VITAL. This is an important observation, because the 2010 VITAL was only 6 months in duration.²⁴

Although the ViRTUE-CKD Trial is certainly of considerable interest and real value, it is not necessarily the last word on this topic; those who believe that vitamin D might well have clinical utility in this context (targeting residual proteinuria) should not yet lose hope. There are some potential issues in a broader interpretation and implementation of the ViRTUE-CKD Trial findings. The ViRTUE-CKD Trial was very short term indeed, and it would be necessary to have a much longer period of vitamin D use to help determine the overall effects (both good and bad) of restricted dietary salt and vitamin D administration. This type of interventional study should be done separately for patients who are microalbuminuric and those with overt proteinuria. It should also be done on cohorts with some significant impairment of excretory kidney function (e.g., when eGFR is in the range of CKD stages 3a and 3b) with elevation of PTH, for whom a vitamin D–based approach makes clinical sense. Finally, although paricalcitol is typically chosen for such clinical trials, there is certainly a place to consider high-dose supplementation of cholecalciferol as an alternative form of vitamin D (because it is typically associated with fewer signs of toxicity).

The search for effective, safe, and nephroprotective interventions to deploy in CKD remains active. Indeed, as a public health intervention, it is vital. We should make a virtue of the fact that hope remains for the discovery of an effective, applicable, and affordable clinical suite of interventions to slow down the otherwise remorseless progression seen in most patients with CKD and by so doing, reduce the burden of disease in terms of dialysis and the need for organ transplantation.

• Copyright © 2017 by the American Society of Nephrology