Nimrit Goraya 1, Jan Simoni 2, Chan-Hee Jo 3 and Donald E. Wesson 1. clinical investigation

Transcription

1 & 2014 International Society of Nephrology clinical investigation Treatment of metabolic acidosis in patients with stage 3 chronic kidney disease with fruits and vegetables or oral bicarbonate reduces urine angiotensinogen and preserves glomerular filtration rate Nimrit Goraya 1, Jan Simoni 2, Chan-Hee Jo 3 and Donald E. Wesson 1 1 Departments of Internal Medicine, Texas A&M College of Medicine, Scott and White Healthcare, Temple, Texas, USA; 2 Department of Surgery, Texas Tech University Health Sciences Center, Lubbock, Texas, USA and 3 Department of Biostatistics, Scott and White Healthcare, Temple, Texas, USA Alkali therapy of metabolic acidosis in patients with chronic kidney disease (CKD) with plasma total CO 2 (TCO 2 )below 22 mmol/l per KDOQI guidelines appears to preserve estimated glomerular filtration rate (egfr). Since angiotensin II mediates GFR decline in partial nephrectomy models of CKD and even mild metabolic acidosis increases kidney angiotensin II in animals, alkali treatment of CKD-related metabolic acidosis in patients with plasma TCO 2 over 22 mmol/l might preserve GFR through reduced kidney angiotensin II. To test this, we randomized 108 patients with stage 3 CKD and plasma TCO mmol/l to Usual Care or interventions designed to reduce dietary acid by 50% using sodium bicarbonate or base-producing fruits and vegetables. All were treated to achieve a systolic blood pressure below 130 mm Hg with regimens including angiotensin converting enzyme inhibition and followed for 3 years. Plasma TCO 2 decreased in Usual Care but increased with bicarbonate or fruits and vegetables. By contrast, urine excretion of angiotensinogen, an index of kidney angiotensin II, increased in Usual Care but decreased with bicarbonate or fruits and vegetables. Creatinine-calculated and cystatin C calculated egfr decreased in all groups, but loss was less at 3 years with bicarbonate or fruits and vegetables than Usual Care. Thus, dietary alkali treatment of metabolic acidosis in CKD that is less severe than that for which KDOQI recommends therapy reduces kidney angiotensin II activity and preserves egfr. Kidney International advance online publication, 2 April 2014; doi: /ki KEYWORDS: acidosis; albuminuria; angiotensin; chronic kidney disease; glomerular filtration rate; nutrition Correspondence: Donald E. Wesson, Texas A&M College of Medicine, Scott and White Healthcare, 2401 South 31st Street, Temple, Texas 76508, USA. dwesson@sw.org Received 30 July 2013; revised 1 January 2014; accepted 6 February 2014 Oral alkali treatment of the metabolic acidosis of chronic kidney disease (CKD) appears to slow the decline of creatinine clearance 1 and estimated glomerular filteration rate (egfr). 2 Current guidelines recommend alkali treatment of metabolic acidosis in CKD patients with plasma total CO 2 (TCO 2 ) o22 mmol/l. 3 Nevertheless, the risk for and rate of glomerular filteration rate (GFR) decline increases in CKD patients as plasma TCO 2 decreases due to metabolic acidosis within ranges that include values 422 mmol/l. 4 6 Consequently, alkali treatment of metabolic acidosis in CKD patients with plasma TCO mmol/l may also be kidney protective. When alkali treatment of metabolic acidosis in CKD is needed, KDOQI recommends Na þ citrate or Na. 3 Na þ citrate increases plasma TCO 2 and possibly slows GFR decline 2 in CKD but is more costly than Na, and citrate promotes gastric aluminum absorption. 7 On the other hand, Na is cheaper but sometimes causes bloating and flatus. 8 Despite their effectiveness in treating metabolic acidosis, the added Na þ of Na þ citrate and Na may exacerbate volume retention and/or hypertension in CKD. Alternatively, non-na þ -based therapy such as base-inducing fruits and vegetables (F þ V) that reduce urine net acid excretion (NAE) 9 and increase plasma TCO 2 in CKD patients with metabolic acidosis 8 may also effectively treat the metabolic acidosis of CKD. Although the added K þ of F þ V risks hyperkalemia in CKD patients with very low GFR, F þ V did not induce hyperkalemia in carefully selected CKD stage 4 egfr (15 29 ml/min) individuals with serum [K þ ] o4.6 meq/l. 8 Mechanisms by which metabolic acidosis may mediate nephropathy progression in CKD or its purported amelioration with oral alkali include pathophysiologic effects of sustained, high kidney levels of mediators of increased distal nephron acidification in response to GFR reduction, including endothelin and aldosterone. 10 Because AII also mediates Kidney International 1

2 clinical investigation N Goraya et al.: Dietary alkali and GFR preservation in CKD increased distal nephron acidification when GFR is reduced 11 and anti-aii drugs slow nephropathy progression in CKD, 12,13 metabolic acidosis may promote and its treatment may reduce nephropathy progression in CKD through kidney AII. We tested the hypothesis that added Na or F þ V in the diet slow egfr decline and reduce kidney AII levels in CKD individuals with stage 3 egfr (30 59 ml/min) and metabolic acidosis characterized by plasma TCO but o24 mmol/l, a range for which current guidelines do not recommend alkali treatment. RESULTS Table 1 shows demographic characteristics and selected parameters at baseline and at 3 years follow-up. There were no differences in the indicated demographics among groups but baseline systolic blood pressure was significantly higher in individuals subsequently given Na ( ) than in the ones in Usual Care. There was no difference in baseline body weight, enalapril daily dose, potential renal acid load, plasma K þ, venous ph, and venous PCO 2 among the three groups. Three-year compared with baseline body weights were lower in Usual Care patients and in individuals given F þ V but not in those given Na ( ). Net body weight loss was greater in the F þ V group than in Usual Care ( 4.0±3.9 vs. 1.9±2.6 kg, Po0.01) and the net loss for both was greater than that for the group ( 0.17±2.7 kg, Po0.048). Body weight did not change significantly for -fed individuals (P ¼ 0.72). The 3-year compared with baseline Enalapril daily dose was lower in the F þ V group but not in the Usual Care and groups and the 3-year dose was lower in F þ V-fed than that in both Usual Care and -fed individuals. Three-year systolic blood pressures were lower than respective baselines for all three groups and the 3-year value was lower in F þ V than in and Usual Care groups. Three-year potential renal acid load was lower than its respective baseline in the F þ V group but was not so for and Usual Care groups. Three-year Plasma [K þ ] was lower than its respective baseline in the group but was not different from baseline in F þ V and Usual Care groups. Three-year venous ph and venous PCO 2 were lower than baseline in Usual Care but each was higher than baseline in the and F þ V groups. Furthermore, 3-year ph and PCO 2 for and F þ V groups were each higher than those in the Usual Care group. Figure 1 shows no difference in baseline plasma TCO 2 (P ¼ 0.62) or 8 h NAE (P ¼ 0.73) among the three groups. Three-year compared with baseline plasma TCO 2 was lower in Usual Care (22.4±0.6 vs. 23.0±0.5, Po0.01) but was higher in (24.0±0.6 vs. 23.1±0.6, Po0.01) and F þ V (23.9±0.6 vs. 23.0±0.6, Po0.01) groups. Three-year compared with baseline 8 h NAE was not different in Usual Care (25.7±2.4 vs. 25.7±2.7, P ¼ 0.86) but was lower in (18.3±2.1 vs. 25.2±2.7, Po0.01) and F þ V (18.2±2.1 vs. 26.0±3.0, Po0.01) groups, consistent with intake of the prescribed Na and F þ V in the two intervention groups. There was no difference in 3-year values for plasma TCO 2 (P ¼ 0.49) or 8 h NAE (P ¼ 0.85) between the and F þ V groups. Figure 2 shows no difference in baseline morning spot urine parameters (per g creatinine) for albumin (Ualb) (P ¼ 0.99), N-acetyl-b-D-glucosaminidase (UNAG) (P ¼ 0.87), or angiotensinogen (UATG) (P ¼ 0.99). Three-year compared with baseline Ualb was lower for Usual Care (300±69 vs. 315±73 mg/g cr, Po0.01), (262±62 vs. 317±72 mg/g cr, Po0.01), and F þ V (242±56 vs. 318±71 mg/g cr, Po0.01) groups. The 3-year values for Ualb were lower than Usual Care for (Po0.01) and F þ V (Po0.01) groups, but 3-year Ualb values were not different between the and F þ V(P ¼ 0.19) groups. Three-year compared with baseline UNAG was higher for Usual Care (9.77±1.14 vs. 8.33±1.02 U/g cr, Po0.01) but was lower for (8.03±0.87 vs. 8.36±0.89 U/g cr, Po0.01) and F þ V (7.84±0.84 vs. 8.44±0.85 U/g cr, Po0.01) group individuals. The 3-year values for UNAG were lower than Usual Care for (Po0.01) and F þ V(Po0.01) groups, but Table 1 General patient characteristics by group at baseline and at 3 years follow-up Usual Care F þ V 3groups Usual Care F þ V 3 years 3groups 3 yr vs. baseline 3 yr vs. baseline 3yrvs. baseline Males (%) B/W/H (%) 53/14/33 53/22/25 53/17/ Variable Mean±s.d. Mean±s.d. Mean±s.d. Mean±s.d. Mean±s.d. Mean±s.d. Age (years) 53.9± ± ± Body weight 83.1± ± ± ± ± ± o o0.01 Enalapril (mg/dl) 12.8± ± ± ± ± ± o0.01 Sys BP(mm Hg) 158.6± ± ± ± ± ±4.5 o0.01 o0.01 o0.01 o0.01 PRAL (mmol/dl) 60.5± ± ± ± ± ±5.9 o o0.01 PK (meq/l) 4.29± ± ± ± ± ±0.12 o o Venous ph 7.36± ± ± ± ± ±0.01 o0.01 o0.01 o0.01 o0.01 Venous PCO ± ± ± ± ± ±0.6 o0.01 o0.01 o0.01 o0.01 Abbreviations: B/W/H, black, white, Hispanic ethnicity, respectively; F þ V, individuals given fruits þ vegetables in amounts designed to reduce potential renal acid load by 50%;, individuals subsequently given 0.3 meq/kg bw/day oral Na ; PK, plasma [K þ ]; PRAL, potential renal acid load; Sys BP, systolic blood pressure. Presented are P-values between and F þ V at baseline and at 1 year and the 1-year compared with baseline values within and F þ V groups. Po0.05 vs. Usual Care; Po0.05 vs. Usual Care and. 2 Kidney International

3 N Goraya et al.: Dietary alkali and GFR preservation in CKD clinical investigation Plasma TCO 2 (mm) h NAE (meq) Usual care FV Usual care FV Figure 1 Box plots of plasma total CO 2 (TCO 2 ) (left) and 8-hour urine net acid excretion (8 h NAE) (right) in chronic kidney disease (CKD) stage 3 individuals at baseline and 3-year follow-up. Usual Care, individuals without additional dietary intervention;, individuals given oral Na at 0.3 meq/kg lean bw/day; F þ V ¼ subjects given fruits and vegetables in amounts designed to reduce dietary potential renal acid load by half. Po0.05 vs. respective baseline; þ Po0.05 vs. respective Usual Care U alb (mg/g Cr) U NAG (U/g Cr) 9 8 U AGT (ug/g Cr) Usual care FV Usual care FV Usual care FV Figure 2 Boxplots of change of urine albumin (mg)-to-creatinine (g) ratio (Ualb) (left), N-acetyl-b-D-glucosaminidase (U)-to-creatinine (g) ratio (UNAG), and angiotensinogen (lg)-to creatinine (g) ratio (UATG) for the three groups of chronic kidney disease (CKD) individuals with egfr ml/min per 1.73 m 2 at baseline and 3-year follow-up. Po0.05 vs. respective baseline; þ Po0.05 vs. respective Usual Care. 3-year UNAG values were not different between and F þ V (P ¼ 0.69) groups. Similarly, 3-year compared with baseline UATG was higher for Usual Care (38.8±6.3 vs. 35.6±5.9 mg/g cr, Po0.01) but was lower for (32.0±5.3 vs. 35.4±5.7 mg/g cr, Po0.01) and F þ V (32.1± 5.2 vs. 35.5±5.4 mg/g cr, Po0.01) groups. The 3-year values for UATG were lower than Usual Care for (Po0.01) and F þ V (Po0.01) groups but 3-year UATG was not different between and F þ V(P ¼ 0.92) groups. Linear mixed model slopes for and F þ V groups were greater Kidney International 3

4 clinical investigation N Goraya et al.: Dietary alkali and GFR preservation in CKD crgfr (ml/min) 35 cysgfr (ml/min) Usual care FV Year Usual care FV Year Figure 3 Error bars (mean±s.e.) of plasma creatinine-calculated estimated GFR (crgfr) (left) and plasma cystatin C calculated estimated GFR (cysgfr) (right) across four time points including baseline and 3 years follow-up. Po0.05 vs. Usual Care at 2 year; þ Po0.05 vs. Usual Care at 3-year follow-up. (more negative) than those for Usual Care (Po0.01) individuals, indicating a faster decline in each parameter for the two intervention groups. Figure 3 shows no difference in baseline egfr calculated using plasma creatinine (crgfr) or plasma Cystatin C (cysgfr) (P ¼ 0.99) among the three groups. Three-year compared with baseline crgfr was lower in the Usual Care (28.8±7.3 vs. 42.6±7.6, Po0.01), (35.2±6.9 vs. 42.6±7.0, Po0.01), and F þ V (36.9±6.7 vs. 42.3±7.1, Po0.01) groups. Three-year values for crgfr were higher for (Po0.01) and F þ V(Po0.01) groups than those in Usual Care individuals, but 3-year crgfr values were not different between and F þ V(P ¼ 0.57) groups. Threeyear compared with baseline cysgfr was lower in Usual Care (26.6±7.0 vs. 39.5±6.8, Po0.01), (32.7±6.7 vs. 39.6±6.6, Po0.01), and F þ V (34.3±6.4 vs. 39.4±6.4, Po0.01) groups. Three-year values for cysgfr were higher for (Po0.01) and F þ V(Po0.01) groups than those in Usual Care group, but 3-year cysgfr was not different between and F þ V(P ¼ 0.34) groups. Linear mixed models for egfr comparisons among Usual Care,, and F þ V groups using creatinine or Cystatin C calculated GFR showed that the rates of egfr decline (slopes) for and F þ V groups were significantly less (less negative) than those for the Usual Care group (Po0.01). DISCUSSION These studies support that 3 years of dietary acid reduction with Na or base-producing F þ V in individuals with metabolic acidosis and stage 3 CKD, due to hypertensive nephropathy, reduced urine excretion of angiotensinogen, an index of kidney AII levels, and preserved egfr. These data add to that from previously published studies 1,2 supporting that oral alkali treatment of metabolic acidosis in CKD is kidney protective. The present studies support at least two additional insights: (1) adding F þ V to the baseline diet is also kidney protective in individuals with CKD and metabolic acidosis; and (2) dietary acid reduction is protective in CKD individuals with metabolic acidosis characterized by plasma TCO mmol/l, a level above which KDOQI does not recommend treatment. 3 Previously described mediators of nephropathy progression in experimental models of CKD with reduced GFR include ammoniagenesis, 14 endothelin, 10,15 and aldosterone. 10 Metabolic acidosis in experimental animals activates the renin angiotensin system including expression of AII type 1 receptors. 16 Because AII stimulates ammoniagenesis 17 and kidney production of endothelin 18 and aldosterone, 19 AII might be the common link for these contributors to nephropathy progression. In addition, kidney AII levels are increased in animals with reduced GFR due to subtotal nephrectomy and AII mediates increased distal nephron acidification associated with reduced GFR in this CKD model. 11 Furthermore, AII receptor antagonism exacerbated metabolic acidosis in human subjects, apparently through a distal nephron mechanism, supporting an important role for AII in mediating acid excretion in human metabolic acidosis. 20 Because AII induces kidney tubuleinterstitial fibrosis, 21 mediates progressive nephropathy in 5/6 Nx, 22 and anti-aii drugs slow nephropathy progression in CKD, 12,13 high kidney AII activity in metabolic acidosis may mediate progressive nephropathy. We hypothesize that high kidney levels of AII that contribute physiologically to increase per nephron kidney acidification in the setting of reduced GFR and metabolic acidosis 11 contribute pathologically to nephropathy progression when metabolic acidosis is sustained. To explore this hypothesis, we measured urine excretion of angiotensinogen (UATG), 4 Kidney International

5 N Goraya et al.: Dietary alkali and GFR preservation in CKD clinical investigation an index of kidney angiotensin levels, 23 in response to chronic dietary acid reduction with Na or F þ V. Figure 2 shows that UATG increased in Usual Care and decreased in the two intervention groups, supporting our hypothesis. Although all study subjects were prescribed angiotensin converting enzyme inhibition, this drug class had no effect on UATG when prescribed to hypertensives. 24 Although all subjects were on angiotensin converting enzyme inhibition, UATG decreased in the two intervention groups but not in the Usual Care group, supporting the more important role for dietary acid reduction in the UATG decrease and not angiotensin converting enzyme inhibition. Furthermore, the dose of angiotensin converting enzyme inhibition was lower at 3 years in F þ V group than in group, yet the UATG level was not different at this time point, again supporting the predominant contribution of dietary acid reduction. These data are consistent with our hypothesis that metabolic acidosis induced AII mediates nephropathy progression in CKD and that dietary acid reduction in CKD patients with metabolic acidosis is an effective strategy to reduce the AII contribution to this progression. Further studies are needed to further test this unifying hypothesis. Previously published studies showed that base-producing F þ V reduced urine NAE and kidney injury in individuals with reduced GFR but no metabolic acidosis 9 and increased plasma TCO 2 in individuals with CKD stage 4 egfr and metabolic acidosis. 8 As previous studies support that the treatment of metabolic acidosis in CKD with Na þ -based alkali preserved GFR, 1,2 the present study supports that F þ V treatment of metabolic acidosis in CKD also preserved GFR. Individuals given F þ V also had lower systolic blood pressure after 3 years, likely contributing to the kidney-protective effect of this intervention. 25 The F þ V given to these CKD stage 3 individuals did not induce hyperkalemia despite the increased dietary K þ associated with this intervention, as was observed in CKD stage 4 individuals given F þ V. 8 Nevertheless, individuals in the present and previous studies were selected to be at low risk for hyperkalemia as described in Materials and Methods. Consequently, clinicians must use caution in recommending F þ V for individuals with low egfr. Nevertheless, the data support that F þ V are an effective alternative to a Na þ -based alkali to improve metabolic acidosis and preserve GFR in stage 4 CKD due to hypertensive nephropathy. The KDOQI guidelines that recommend alkali treatment of metabolic acidosis in CKD patients with plasma TCO 2 o22 mmol/l were based on evidence and opinion but cautioned that more research is needed on the long-term effects of correcting acidemia on clinical outcomes. 3 Studies in animal models of CKD since issuance of these recommendations suggest that alkali treatment earlier in the course of CKD may be kidney protective. Animals with 2/ 3 Nx have reduced GFR without metabolic acidosis yet have progressive GFR decline that is ameliorated with dietary alkali. 10,26 Epidemiologic studies show that CKD patients with ranges of plasma TCO 2 that include those 422 mmol/l are associated with greater risk for and faster rate of GFR decline. 4 6 In addition, Na slowed the rate of egfr decline in individuals with reduced egfr but without metabolic acidosis. 27 Individuals for the present studies were selected with CKD stage 3 (30 59 ml/min per 1.73 m 2 ) and metabolic acidosis characterized by plasma TCO but o24 mmol/l, a range for which current KDOQI guidelines do not recommend treatment. Figure 3 shows that egfr declined in all three groups but was higher at 3 years for both intervention groups compared with that in Usual Care group, supporting that dietary alkali was kidney protective for these patients for whom current guidelines do not recommend treatment. Further and larger scale studies will be needed to better determine whether the treatment of CKD patients with plasma TCO mmol/l should become standard care. Three-year Ualb values were lower than baseline in all groups, consistent with angiotensin converting enzyme inhibition 13 and systolic blood pressure reduction 28 achieved in each group. On the other hand, 3-year values for UNAG, an index of kidney tubule-interstitial injury, 29 compared to baseline were higher in Usual Care group, consistent with worsening tubule-interstitial injury. By contrast, UNAG was lower than baseline at 3 years in both intervention groups, consistent with ameliorated tubule-interstitial injury. Oral alkali reduced tubule-interstitial injury in the five-sixths nephrectomy (5/6 Nx) model of CKD 14,15 and reduced UNAG in CKD individuals with metabolic acidosis. 2 These data support that the course of tubule-interstitial injury helps predict nephropathy progression as previously suggested. 30 Furthermore, these data are consistent with studies in the 5/6 Nx CKD model, which showed that it was the tubuleinterstitial component of progressive nephropathy that was most closely associated with metabolic acidosis and its correction. 15 Consequently, the lower Ualb values compared with the Usual Care group at 3 years in the and F þ V groups in Figure 2 may be due to improved metabolic acidosis. Considering other factors that might have contributed to nephropathy progression in the studied subjects, all were nonsmokers and none of the few patients hospitalized during follow-up had acute kidney injury as an admitting or discharge diagnosis. Nevertheless, we cannot comment as to whether undetected acute kidney injury occurred but was not captured in the patient record. We suspect, however, that acute kidney injury had little impact on the reported data. In summary, these studies show that dietary acid reduction with Na or base-inducing F þ V reduces urine excretion of angiotensinogen, a marker of kidney AII levels, and preserved egfr in individuals with CKD stage 3 egfr due to hypertensive nephropathy. Because all study subjects underwent blood pressure reduction with regimens that included angiotensin converting enzyme inhibition, both interventions appear to be kidney-protective adjuncts to currently recommended kidney-protective strategies. The F þ V did not induce hyperkalemia in these individuals, who were carefully selected to be at low risk for hyperkalemia. Consequently, clinicians must use caution when considering Kidney International 5

6 clinical investigation N Goraya et al.: Dietary alkali and GFR preservation in CKD F þ V as a treatment alternative to Na þ -based alkali therapy for metabolic acidosis in CKD patients with low GFR. These studies also support that dietary acid reduction is kidney protective in CKD patients with plasma TCO 2 higher than that for which current guidelines recommend alkali treatment. MATERIALS AND METHODS This interventional study compared the effect of 36 months of dietary acid reduction with added daily oral Na ( ) or added F þ V on GFR estimated by using plasma cysgfr, urine parameters of kidney injury, and urine excretion of angiotensinogen, an index of kidney angiotensin II levels. 23 Study subjects were individuals with CKD stage 3 (estimated GFR ml/min per 1.73 m 2 ), macroalbuminuric (urine albumin-to-creatinine ratio 4200 mg/g creatinine), hypertensive nephropathy and metabolic acidosis characterized by plasma TCO mmol/l but o24 mmol/l. We chose individuals fitting these plasma TCO 2 criteria to examine subjects who had metabolic acidosis, although not of a severity warranting alkali treatment (plasma TCO 2 o22 mmol/l) according to current guidelines, 3 thereby allowing us to ethically randomize some to receive no alkali treatment. Plasma Cystatin C calculated egfr was the primary end point because it is superior to creatinine as a marker of kidney function 31 and because among adults diagnosed with CKD using creatinine-based equations, those with CKD according to the Cystatin C based equation have the highest risk for CKD-related complications, including kidney failure. 32 Primary outcome was the effect of 3 years of these interventions on cysgfr. Secondary outcomes were plasma creatinine-calculated egfr (crgfr), 33 Ualb, and UNAG, two urine indices whose increasing levels are associated with worsening kidney injury, 30,34 and urine angiotensinogen. Earlier studies 2 suggested the need for 33 patients in each of the three groups to achieve an 80% probability of seeing a Po0.05 difference of the and F þ V groups in the primary outcome (assuming that Na and F þ V were equally effective in egfr preservation) compared with Usual Care group. Individuals were identified from our clinic system, as described, 27 who met these inclusion criteria: (1) nonmalignant hypertension; (2) egfr by MDRD formula ml/min; (3) plasma TCO 2 o25 mmol/l (the lower limit of normal in our clinical laboratory) and 422 mmol/l but o24 mmol/l upon follow-up measurement using ultrafluoremetry 35 ; (4) macroalbuminuria because of the associated higher risk for subsequent GFR decline than normoalbuminuria or microalbuminuria in individuals with hypertension-associated nephropathy 36 ; (5) able to tolerate angiotensin converting enzyme inhibition; (6) nonsmoking defined as not having smoked tobacco for X1 year prior to study entry because smoking is associated with exacerbation of hypertension-associated nephropathy 36 ; (7) no diabetes or cardiovascular disease on their problem lists; (8) X2 primary care physician visits in the preceding year, showing compliance with clinic visits; (9) age X18 years and able to give consent. Exclusion criteria were: (1) primary kidney disease or findings consistent thereof such as Xthree red blood cells per highpowered field of urine or urine cellular casts; (2) history of diabetes or fasting blood glucose X110 mg/dl; (3) current pregnancy, history of malignances, chronic infections, or clinical evidence of cardiovascular disease; (4) peripheral edema or diagnoses associated with edema such as heart/liver failure or nephrotic syndrome; (5) baseline plasma [K þ ] 44.6 meq/l because observations of our cadre of such individuals, collected over many years, and published studies 8 show that those with CKD stage 4 and with [K þ ] o4.6 meq/l were at very low risk of subsequently developing [K þ ] 45.0 meq/l; and (6) individuals taking or those who could not stop taking drugs other than angiotensin converting enzyme inhibitors that limit urine K þ excretion. Preliminary studies showed that no subject fitting the egfr criteria with plasma [K þ ] p4.6 meq/l and given F þ V developed plasma [K þ ] 45.0 meq/l after four weekly measurements of plasma [K þ ]. Nine hundred and sixty-one subjects meeting study criteria were identified over 8 years, of which 108 were matched for age, sex, ethnicity (black vs. white or Hispanic), egfr, and Ualb and agreed to randomization. When three so-matched individuals were identified, they were prospectively randomized to Usual Care,,orFþVgroups and began follow-up. The 36 F þ V-fed individuals received an amount of fresh F þ V free of charge to reduce their dietary acid by 50% as done previously, 8,9 focusing primarily on F þ V that are particularly base-inducing. 37 Fruits that were provided predominantly were apples, apricots, oranges, peaches, pears, raisins, and strawberries. Vegetables that were provided predominantly were carrots, cauliflower, eggplant, lettuce, potatoes, spinach, tomatoes, and zucchini. Subjects did not receive specific dietary instructions and were allowed to integrate the provided F þ V into their ad lib diets as they wished. The necessary amount of F þ V was prescribed by a dietician and distributed from a community center food bank. To assure that each F þ V-fed subject received sufficient F þ V and did not spread their prescribed amount throughout their families, we gave the prescribed amount for each household member. The 36 -group individuals received oral Na 0.3 meq/kg/d (average dose per patient was 25.2 meq/d), an amount estimated to reduce their dietary acid about 50%, as scored tablets containing 10 meq Na with sucrose in three divided daily doses. Each subject was dosed to the nearest one-half tablet by lean body weight in Kg (e.g., a 70 kg subject received two tablets daily). Because receipt of F þ V could not be blinded, study subjects were randomized and the overall study was controlled but not blinded. Dietary acid was measured as potential renal acid load, pre- and post-interventions, using 3-day diaries of food intake and a published formula. 37 We assessed compliance to the prescribed Na and F þ V diet through reductions in urine NAE and did not monitor subject intake of Na or F þ V. All subjects received furosemide for blood pressure. The diagnosis of hypertension-associated nephropathy as the exclusive nephropathy cause was made clinically by excluding individuals with systemic diseases associated with nephropathy, nephrotic-range proteinuria, and urine abnormalities except albuminuria. None had a kidney biopsy. Secondary causes of hypertension such as renal artery stenosis and hyperaldosteronism were excluded clinically. Kidney Doppler studies and plasma aldosterone-to-renin ratios were not performed. All patients had systolic blood pressure reduced toward o130 mm Hg using a described protocol 38 with regimens including angiotensin converting enzyme inhibitors as recommended for hypertensives with albuminuria. 39 All subjects were prescribed enalapril for angiotensin converting enzyme inhibition because it was the only angiotensin converting enzyme inhibitor on our indigent formulary when these studies began. No subjects were prescribed angiotensin II receptor antagonists. Eight-hour ( hours) urine NAE) was measured and calculated from urine titratable acidity, ammonium (NH þ 4 ), and ([NH þ 4 ] þ [TA] [ ]) as described 2 at baseline and at the end of the 3-year follow-up. Venous plasma acid base 6 Kidney International

7 N Goraya et al.: Dietary alkali and GFR preservation in CKD clinical investigation parameters and Ualb, UNAG, and UATG were each measured before and after intervention, as were urine albumin (Ualb), UNAG, and UATG measured in a morning urine specimen and factored per gram of creatinine. Our Institutional Review Board approved the protocol. Analytical methods Plasma and urine creatinine and urine albumin were measured using the Sigma Diagnostics Creatinine Kit (Procedure No. 555, Sigma Diagnostics, St Louis, MO). 38 The IRMA SL Series 2000 blood analysis system (Edison, NJ) measured venous plasma/blood ph and PCO 2. Urine and plasma TCO 2 were measured using ultrafluoremetry. 35 Urine titratable acidity was measured by þ correction to the ambient plasma ph by NaOH addition, NH 4 by the Formalin titrametric (to ambient plasma ph) method. 40 Urine NAG was measured using a colorimetric assay (Boehringer Mannheim, Mannheim, Germany). 2 Urine ATG was measured using RIA quantitation of Ang I generation 23 after the addition of excess exogenous porcine renin (R 2761; Sigma, St Louis, MO) using a commercially available kit (Incstar, Stillwater, MI). Urine samples were incubated with renin at 37 1C, removed at 0, 10, 30, 60, and 120 min, and diluted with reagent blank in such a way that the RIA results were on the linear portion of a previously determined standard curve. The amount of Ang I generated at each of the above time points was determined by comparison with the standard curve. The amount of Ang I produced was plotted vs. time. Saturation kinetics due to conversion of all angiotensinogen to Ang I were obtained by the 60-min point. Statistical methods Patient characteristics at the time of study enrollment were tabulated by percentages or described by mean and s.d.. We first compared baseline characteristics among the three subject groups with the w 2 test or one-way analysis of variance, as appropriate. Transformations of variables were examined to ensure that normality assumptions were satisfied. If necessary, nonparametric tests were also considered. The primary outcome was cysgfr in response to 3 years of an oral F þ V or Na diet. Secondary outcomes included Ualb, UNAG, and UATG. The changes from pre to post for each group were described by mean and s.d. and they were considered with a paired t-test. The differences among the three groups, F þ V,, and Usual Care, were considered with one-way analysis of variance followed by post hoc Tukey s test. We fit linear mixed models with terms for group and time (0, 1, 2, and 3 years), and the interaction between the two. A P-value of less than 0.05 indicates a statistical significance. SAS version 9.3 (SAS Institute, Cary, NC) was used for the statistical analysis and graphs were created using R (R Core Development Team, 2013). DISCLOSURE All the authors declared no competing interests. ACKNOWLEDGMENTS We thank the nursing, dietary, and clerical staff of the Department of Internal Medicine at Texas Tech University Health Sciences Center for their assistance and to the Inside Out Community Outreach Program and Food Bank of Lubbock, Texas for making these studies possible.this work was supported by funds from the Larry and Jane Woirhaye Memorial Endowment in Renal Research the Texas Tech University Health Sciences Center, by the Statistics Department of Scott and White Healthcare, and by the Academic Operations Division at Scott and White Healthcare. REFERENCES 1. de Brito-Ashurst I, Varagunam M, Raferty MJ et al. Bicarbonate supplementation slows progression of CKD and improves nutritional status. J Am Soc Nephrol 2009; 20: Phisitkul S, Khanna A, Simoni J et al. Amelioration of metabolic acidosis in subjects with low GFR reduces kidney endothelin production, reduces kidney injury, and better preserves GFR. Kid Int ; Kopple JD, Wolfson M, Chertow EC et al. National Kidney Foundation: K/DOQI Clinical Practice Guidelines for Nutrition in Chronic Renal Failure. Am J Kid Dis 2000; 35: S1 S Raphael K, Wei G, Baird B et al. Higher plasma bicarbonate levels within the normal range are associated with better survival and renal outcomes in African Americans. Kid Int 2011; 79: Shah SN, Abramowitz M, Hostetter TH et al. Plasma bicarbonate levels and the progression of kidney disease: a cohort study. Am J Kid Dis 2009; 54: Dobre MD, Yang W, Chen J et al. Association of plasma bicarbonate with risk of renal and cardiovascular outcomes in CKD: A report from the chronic renal insufficiency cohort (CRIC) study. Am J Kid Dis 2013; 62(4): Nolan C, Califano JR, Butzin CA. Influence of calcium acetate or calcium citrate on intestinal aluminum absorption. Kid Int 1990; 38: Goraya N, Simoni J, Jo C-H et al. Comparison of treating the metabolic acidosis of CKD stage 4 hypertensive kidney disease with fruits and vegetables or sodium bicarbonate. Clin J Am Soc Nephrol 2013; 8: Goraya N, Simoni J, Jo C-H et al. Dietary acid reduction with fruits and vegetables or sodium bicarbonate reduces kidney injury in subjects with moderately reduced GFR due to hypertensive nephropathy. Kid Int 2012; 81: Wesson DE, Simoni J. Acid retention during kidney failure induces endothelin and aldosterone production which lead to progressive GFR decline, a situation ameliorated by alkali diet. Kid Int 2010; 78: Wesson DE, Jo C-H, Simoni J. Animals with reduced GFR but no metabolic acidosis nevertheless have increased distal nephron acidification that is induced by acid retention and mediated through angiotensin II receptors. Kid Int 2012; 82: Lewis EJ, Hunsicker LC, Bain RP et al. The effect of angiotensionconverting enzyme inhibition on diabetic nephropathy. NEngJMed 1993; 329: Wright JT, Bakris G, Greene T et al. Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease. JAMA 2002; 288: Nath KA, Hostetter MK, Hostetter TH. Pathophysiology of chronic tubulointerstitial disease in rats-interactions of dietary acid load, ammonia, and complement component C3. J Clin Invest 1985; 76: Phisitkul S, Hacker C, Simoni J et al. Dietary protein causes a decline in the glomerular filtration rate of the remnant kidney mediated by metabolic acidosis and endothelin receptors. Kid Int 2008; 73: Ng H-Y, Chen H-C, Tsai Y-C et al. Activation of intrarenal renin-angiotensin system during metabolic acidosis. Am J Nephrol 2011; 34: Nagami G. Role of angiotensin II in the enhancement of ammonia production and secretion by the proximal tubule in metabolic acidosis. Am J Renal Physiol 2008; 294: F874 F Sasser J, Pollock JS, Pollock D. Renal endothelin in chronic angiotensin II hypertension. Am J Physiol Regulatory Integrative Physiol 2002; 283: R243 R Xue C, Siragy HM. Local renal aldosterone system and its regulation by salt, diabetes, and angiotensin type 1 receptor. Hypertension 2005; 46: Henger A, Tutt P, Riesen WF et al. Acid-base and endocrine effects of aldosterone and angiotensin II inhibition in metabolic acidosis in human patients. J Lab Clin Med 2000; 136: Seccia TM, Maniero C, Belloni AS et al. Role of angiotensin II, endothelin-1, and L-type calcium channel in the development of glomerular, tubulointerstitial, and perivascular fibrosis. JHypertens2008; 26: Anderson S, Rennke HG, Brenner BM. Therapeutic advantage of converting enzyme inhibitors in arresting progressive renal disease associated with systemic hypertension in the rat. J Clin Invest 1986; 77: Kidney International 7

8 clinical investigation N Goraya et al.: Dietary alkali and GFR preservation in CKD 23. Kobori H, Harrison-Bernard LM, Navar G. Urinary excretion of angiotensinogen reflects intrarenal angiotensionogen production. Kid Int 2002; 61: van der Heuvel M, Batenburg WW, Jainandunsing S et al. Urinary renin, but not angiotensinogen or aldosterone, reflects the renal renin-angiotensinaldosterone system activity and the efficacy of renin-angiotensinaldosterone system blockade in the kidney. JHypertens2011; 29: Toto RD, Mitchell HC, Smith RD et al. Strict blood pressure control and progression of renal disease in hypertensive nephrosclerosis. Kid Int 1995; 48: Wesson DE, Simoni J. Increased tissue acid mediates progressive GFR decline in animals with reduced nephron mass. Kid Int 2009; 75: Mahajan A, Simoni J, Sheather S et al. Daily oral sodium bicarbonate preserves glomerular filtration rate by slowing its decline in early hypertensive nephropathy. Kid Int 2010; 78: Pascual JM, Rodilla E, Miralles A et al. Determinants of urinary albumin excretion reduction in essential hypertension: A long-term follow-up study. J Hypertens 2006; 24: Costigan M, Rustom R, Shenkin A et al. Origin and significance of urinary N-acetyl b-d-glucosaminidase (NAG) in renal patients with variable function, pathology and proteinuria. Clinica Chimica Acta 1996; 255: D Amico G. Tubulo-interstitial damage in glomerular diseases: its role in the progression of the renal damage. Nephrol Dial Transplant 1998; 13(Suppl 1): Dharnidharka VR, Kwon C, Stevens G. Serum Cystatin C is superior to serum creatinine as a marker of kidney function: a meta-analysis. Am J Kid Dis 2002; 40: Peralta CA, Katz R, Sarnak MJ et al. Cystatin C identifies chronic kidney disease patients at higher risk for complications. J Am Soc Nephrol 2011; 22: Klahr S, Levey AS, Beck GJ et al. The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. Modification of Diet in Renal Disease Study Group. New Eng J Med 1994; 330: Remuzzi G, Bertani T. Pathophysiology of progressive nephropathies. New Eng J Med 1998; 339: Wesson DE. Dietary reduces distal tubule acidification by increasing cellular secretion. Am J Physiol 1996; 271, Renal Fluid and Electrolyte Physiol. 40 F132 F Warmoth L, Simoni J, Wesson DE. Cigarette smoking and urine albumin excretion predict subsequent GFR decline in severe primary hypertension despite improved blood pressure control. Am J Med Sci 2005; 330: Remer T, Manz F. Potential renal acid load of foods and its influence on urine ph. J Am Dietetic Assn 1995; 95: Regalado M, Yang S, Wesson DE. Cigarette smoking is associated with augmented progression of renal insufficiency in severe essential hypertension. Am J Kid Dis 2000; 35: Chobanion AV, Bakris GL, Black HR et al. The seventh report of the Joint National Commission on Detection, Evaluation, and Treatment of High Blood Pressure: The JNC 7 Report. JAMA 2003; 289: Cunarro JA, Weiner MW. A comparison of methods for measuring urinary ammonium. Kid Int 1974; 5: Kidney International