**2. The physiological roles of RAAS in DD-CKD patients**

The role of RAAS in hypertensive DD-CKD patients was confirmed by the normalization of blood pressure (BP) upon administration of an angiotensin antagonist, saralasin. Normally, volume overload and elevation of BP result in suppression of RAAS production. Since this feedback is often incomplete in CKD patients, CKD patients often show HT and high or normal RAAS. Weidmann et al. reported that the renin levels of hypertensive hemodialysis-dependent CKD (HDD-CKD) patients were approximately twice as high as those of normal subjects. Parenchymal renal injury and renovascular disease may cause increased renin secretion in end-stage CKD. The prevalence of renal artery stenosis may be as high as 40% in patients starting HD, although the diagnosis was determined in only one-quarter of such a group before entering a dialysis program. Kimura et al. reported that plasma rennin activity (PRA) increased from 2.3±0.5 ng/ml/hr at just before initiation of HD to 6.5±1.3 ng/ml/hr over an 8- to 10-year period in HDD-CKD patients. These data suggested that renin secretion continued even after disuse atrophy of kidney with almost complete deterioration of its excretory function.

Activated RAAS increased inflammatory mediators, which is an independent risk factor for CVD. The mechanism is thought to be as follows. Activated RAAS directly increases pro-

<sup>\*</sup> Corresponding Author

The Renin-Angiotensin-Aldosterone System in Dialysis Patients 115

Fig. 1. (P)RR expression in peripheral blood mononuclear cells in healthy subjects (n=10) and HDD-CKD patients (n=49). Columns represent the mean ± s.e. (P)RR, prorenin receptor;

HDD-CKD, hemodialysis-dependent chronic kidney disease (Morishita *et al.*, 2011).

Fig. 2. Correlations of plasma prorenin level and 8-OHdG in hemodialysis-dependent chronic kidney disease patients. 8-OHdG, 8-hydroxydeoxyguanosine (Morishita *et al.*, 2011).

Tradolapril and captopril have been reported to be effective for HT in HDD-CKD patients. Zheng et al. reported that systolic BP (SBP) was decreased from 122.2±7.1 to 116.4±11.6 mmHg and diastolic BP (DBP) was decreased from 75.3±10.4 to 70.4±11.4 mmHg after 2-10 weeks by the administration of tradopril (2-8 mg/thrice a week (TIW)) after each HD session in 10 HDD-CKD patients. In that study, atenolol (25-50 mg/TIW) and/or amlodipine (10 mg/TIW) were also given if the patients had any member of these classes of drugs as part of their daily regimen. Wauterd et al. reported that captopril was administered

**3. The blocked effect of RAAS in DD-CKD patients** 

**Angiotensin-converting enzyme inhibitors in HDD-CKD patients** 

**3.1 HDD-CKD patients** 

inflammatory gene expression and activates oxidative stress, leading to progressive inflammation of the vascular endothelium. In addition, among RAAS, mainly angiotensin II (AT II) stimulates vascular reactive oxygen species (ROS) production from sources such as NADPH oxidase and uncoupled endothelial nitric oxide (NO) synthesis. Increased ROS down-regulates NO activity that leads to endothelial dysfunction. AT II also directly increases pro-inflammatory gene expressions, such as VCAM-1 and MCP-1. Both processes lead to further recruitment of inflammatory cells and accelerate the vascular inflammatory response. On the other hand, the prognostic value of a high plasma aldosterone concentration (PAC) in HDD-CKD patients is unknown, where PAC is known to be associated with poor outcome in patients with cardiac disease. Kohagura et al. reported that lower PAC was independently predictive of death in hypertensive DD-CKD patients. The adjusted hazard ratio (95% confidence interval) of the lower PAC group was 2.905 (1.187– 7.112, *p*=0.020). The significance of PAC became marginal when normalized with albumin or potassium. These results suggested that higher PAC was not associated with an increase in total and cardiovascular deaths among hypertensive HDD-CKD patients. The association between lower PAC and poor survival may be driven by volume retention and/or lower potassium level. Diskin et al. also reported that HDD-CKD patients with higher aldosterone levels tended to have longer survival.

Recently, it was found that RAAS components are expressed in many tissues, such as the heart, kidney, placenta, testis, eye, and lymphocytes. These local tissue RAAS components are suggested to contribute to tissue damage. Prorenin, which is a biosynthetic precursor of renin, is secreted not only by the kidney but also by many other tissues, whereas circulating renin is derived exclusively from juxta-glomerular cells of the kidney. Prorenin does not have enzymatic activity itself; however, several studies have reported that circulating prorenin could be taken up by tissues and contribute to activating local RAAS by binding (pro)renin receptor [(P)RR] and then that inactive prorenin is converted to the active form, which obtains enzymatic activity by conformational change. Recent studies have demonstrated that prorenin-(P)RR interaction activated tissue RAAS and contributed to the pathogenesis of organ damage in several diseases, such as HT and diabetes end organ damage; however, few studies have reported the role of tissue RAAS and prorenin in DD-CKD patients. Takemitsu et al. reported that arterial (P)RR may contribute to activate arterial tissue RAAS in HDD-CKD patients since arterial (P)RR mRNA expression was correlated with arterial angiotensin-converting enzyme (ACE) mRNA expression. Takemitsu et al. also reported that plasma prorenin concentration was correlated with PRA, plasma AT I level, plasma AT II level, and PAC level in HDD-CKD patients. Recently, we reported that the plasma prorenin level increased in HDD-CKD patients [147.1 +/- 118.9 pg/ml (standard value <100 pg/ml)]. The (P)RR mRNA expression level in peripheral mononuclear cells (PBMCs) also increased 1.41 +/- 0.39-fold in HDD-CKD patients compared with that in healthy control subjects (p < 0.001) (Figure 1). Plasma prorenin significantly correlated with plasma 8-hydroxydeoxyguanosine (8-OHdG) level (r = 0.535, p < 0.001), which is a useful marker for assessment of oxidative DNA damage in reactive oxygen species (ROS) including in HDD-CKD patients (Figure 2). These results suggested that tissue RAAS activated by circulating prorenin contributes to the regulation of plasma 8- OHdG level in HDD-CKD patients.

inflammatory gene expression and activates oxidative stress, leading to progressive inflammation of the vascular endothelium. In addition, among RAAS, mainly angiotensin II (AT II) stimulates vascular reactive oxygen species (ROS) production from sources such as NADPH oxidase and uncoupled endothelial nitric oxide (NO) synthesis. Increased ROS down-regulates NO activity that leads to endothelial dysfunction. AT II also directly increases pro-inflammatory gene expressions, such as VCAM-1 and MCP-1. Both processes lead to further recruitment of inflammatory cells and accelerate the vascular inflammatory response. On the other hand, the prognostic value of a high plasma aldosterone concentration (PAC) in HDD-CKD patients is unknown, where PAC is known to be associated with poor outcome in patients with cardiac disease. Kohagura et al. reported that lower PAC was independently predictive of death in hypertensive DD-CKD patients. The adjusted hazard ratio (95% confidence interval) of the lower PAC group was 2.905 (1.187– 7.112, *p*=0.020). The significance of PAC became marginal when normalized with albumin or potassium. These results suggested that higher PAC was not associated with an increase in total and cardiovascular deaths among hypertensive HDD-CKD patients. The association between lower PAC and poor survival may be driven by volume retention and/or lower potassium level. Diskin et al. also reported that HDD-CKD patients with higher aldosterone

Recently, it was found that RAAS components are expressed in many tissues, such as the heart, kidney, placenta, testis, eye, and lymphocytes. These local tissue RAAS components are suggested to contribute to tissue damage. Prorenin, which is a biosynthetic precursor of renin, is secreted not only by the kidney but also by many other tissues, whereas circulating renin is derived exclusively from juxta-glomerular cells of the kidney. Prorenin does not have enzymatic activity itself; however, several studies have reported that circulating prorenin could be taken up by tissues and contribute to activating local RAAS by binding (pro)renin receptor [(P)RR] and then that inactive prorenin is converted to the active form, which obtains enzymatic activity by conformational change. Recent studies have demonstrated that prorenin-(P)RR interaction activated tissue RAAS and contributed to the pathogenesis of organ damage in several diseases, such as HT and diabetes end organ damage; however, few studies have reported the role of tissue RAAS and prorenin in DD-CKD patients. Takemitsu et al. reported that arterial (P)RR may contribute to activate arterial tissue RAAS in HDD-CKD patients since arterial (P)RR mRNA expression was correlated with arterial angiotensin-converting enzyme (ACE) mRNA expression. Takemitsu et al. also reported that plasma prorenin concentration was correlated with PRA, plasma AT I level, plasma AT II level, and PAC level in HDD-CKD patients. Recently, we reported that the plasma prorenin level increased in HDD-CKD patients [147.1 +/- 118.9 pg/ml (standard value <100 pg/ml)]. The (P)RR mRNA expression level in peripheral mononuclear cells (PBMCs) also increased 1.41 +/- 0.39-fold in HDD-CKD patients compared with that in healthy control subjects (p < 0.001) (Figure 1). Plasma prorenin significantly correlated with plasma 8-hydroxydeoxyguanosine (8-OHdG) level (r = 0.535, p < 0.001), which is a useful marker for assessment of oxidative DNA damage in reactive oxygen species (ROS) including in HDD-CKD patients (Figure 2). These results suggested that tissue RAAS activated by circulating prorenin contributes to the regulation of plasma 8-

levels tended to have longer survival.

OHdG level in HDD-CKD patients.

Fig. 1. (P)RR expression in peripheral blood mononuclear cells in healthy subjects (n=10) and HDD-CKD patients (n=49). Columns represent the mean ± s.e. (P)RR, prorenin receptor; HDD-CKD, hemodialysis-dependent chronic kidney disease (Morishita *et al.*, 2011).

Fig. 2. Correlations of plasma prorenin level and 8-OHdG in hemodialysis-dependent chronic kidney disease patients. 8-OHdG, 8-hydroxydeoxyguanosine (Morishita *et al.*, 2011).

#### **3. The blocked effect of RAAS in DD-CKD patients**

#### **3.1 HDD-CKD patients**

#### **Angiotensin-converting enzyme inhibitors in HDD-CKD patients**

Tradolapril and captopril have been reported to be effective for HT in HDD-CKD patients. Zheng et al. reported that systolic BP (SBP) was decreased from 122.2±7.1 to 116.4±11.6 mmHg and diastolic BP (DBP) was decreased from 75.3±10.4 to 70.4±11.4 mmHg after 2-10 weeks by the administration of tradopril (2-8 mg/thrice a week (TIW)) after each HD session in 10 HDD-CKD patients. In that study, atenolol (25-50 mg/TIW) and/or amlodipine (10 mg/TIW) were also given if the patients had any member of these classes of drugs as part of their daily regimen. Wauterd et al. reported that captopril was administered

The Renin-Angiotensin-Aldosterone System in Dialysis Patients 117

benidipine (4-12 mg/day), and an α-blocker, doxazosin (2-4 mg/day). SPB was significantly decreased in both placebo and losartan groups after 12 months.These results suggested that this beneficial effect of losartan for left ventricular hypertrophy was independent of a BP lowering effect. Takahashi et al. stated that candesartan (4-8 mg/day) reduced cardiovascular events and mortality compared with placebo (16.3% vs. 45.9 and 0.0 vs. 18.9%, respectively) after 19.4±1.2 months in HDD-CKD patients in a stable condition and with no clinical evidence of cardiac disorder. In this study, the brain natriuretic peptide (BNP) level did not differ between the two groups at enrolment, whereas in patients who had not experienced a cardiovascular event at 12 months, the BNP levels were significantly increased in the control group but not in the candesartan group. Suzuki et al. reported that angiotensin receptor blockers (ARBs) (valsartan: 160 mg/day, candesartan: 12 mg/day, or losartan: 100 mg/day) treatment was independently associated with reduced cardiovascular events (hazard ratio, 0.51; 95% confidence interval, 0.33 to 0.79; p=0.002) compared with no ARB treatment group for 180 HDD-CKD patients in each group during a 3 year observation period. There were 34 (19%) fatal or nonfatal CVD events in the ARBs group and 59 (33%) in the no ARB group. BP did not differ between the ARBs group and the no ARB group during the observation period. After adjustment for age, sex, diabetes, and SBP, treatment with an ARB was independently associated with reduced fatal and nonfatal CVD events (hazard ratio, 0.51; 95% confidence interval, 0.33 to 0.79; p=0.002). These results demonstrated that ARBs are effective in controlling BP and preventing CVD in HDD-CKD patients. We predict that the cardioprotective effects of ARBs may be independent of their BP lowering effect in

An oral direct renin inhibitor, aliskiren, is effective against essential HT by reducing PRA, resulting in suppression of RAAS; however, little was known about the effects of aliskiren in HDD-CKD patients. Recently, we reported on antihypertensive and potential cardiovascular protective effects of aliskiren, in hypertensive HDD-CKD patients. Aliskiren (150 mg/day) significantly reduced SBP and DBP after 2 month in hypertensive HDD-CKD patients (Figure 3). RAAS was suppressed by aliskiren treatment (PRA: 3.6±4.0 to 1.0 ±1.5 ng/ml/hr, p=0.004; AT I: 1704.0±2580.9 to 233. 7±181.0 pg/ml, p=0.009; AT II: 70.2±121.5 to 12.4±11.5 pg/ml, p=0.022) (Figure 4). Surrogate markers for cardiovascular disease such as BNP, highsensitivity CRP (hs-CRP), and an oxidative stress marker, diacron-reactive oxygen metabolite (d-ROM), were inhibited by aliskiren (BNP: 362.5±262.1 to 300.0±232.0 pg/ml, p=0.043; hs-CRP: 6.2±8.1 to 3.5±3.7 mg/l, p=0.022; d-ROM: 367.0±89.8 to 328.3±70.9 U.CARR, p=0.022) (Figure 5). The levels of inhibition of these surrogate markers for CVD by aliskiren did not correlate with the decreased levels of BP. Two treatments were discontinued owing to an adverse event and symptomatic hypotension by aliskiren. The adverse event was eyebrow alopecia (1 patient). A possible connection of this event to aliskiren treatment could not be excluded. The symptomatic hypotension recovered to the basal level after aliskiren withdrawal. Increased serum potassium was not observed in any patients. Several studies reported that mean trough plasma aliskiren concentrations were increased by renal impairment; however, an increase in exposure did not correlate with the severity of renal impairment. Moreover, renal clearance of aliskiren was found to occur for only a small fraction (0.1-1.0%). These data suggest that adjustment of the aliskiren dose is

HDD-CKD patients. **Direct renin inhibitor** 

orally in 2 daily doses of 25 to 200 mg in 8 HDD-CKD patients. These patients showed HT despite intensive ultrafiltration and conventional antihypertensive therapy. For 4 patients with the highest PRA, their BP was normalized by captopril alone. For the 4 remaining patients, captopril therapy was complemented by salt subtraction, which consisted of replacement of 1-2 liters of ultrafiltrate by an equal volume of 5% dextrose until BP was controlled. After an average treatment period of 5 months, BP of all 8 patients was reduced from 179/105 ± 6/3 (mean ±SEM) to 134/76 ± 7/5 mmHg (p <0.001) without a significant change in body weight. These clinical studies demonstrated that ACE inhibitor has beneficial effects to control BP in HDD-CKD patients. In addition, some studies reported that ACE inhibitors showed cardiovascular protective effects in HDD-CKD patients as follows. London et al. reported that perindopril (2-4 mg after each HD session) significantly reduced left ventricular mass (317±18 to 247±21 g, p=0.036) after 12 months in 14 HDD-CKD patients whereas a calcium channel antagonist, nitredipine (20-40 mg/day), did not when perindopril and nitredipine showed similar reductions of BP. Matsumoto et al. reported that imidapril (2.5 mg/day) significantly reduced left ventricular mass index (132±10 to 109±6 g/m2, p<0.05) after 6 months; however, placebo did not produce a change in HDD-CKD patients. SBP and DBP were not significantly changed in either imidapril or placebo group. In that study, ACE was reduced (12±1 to 5±2 U/I, p<0.01) and PRA was increased (3.3±0.8 to 8.1±3.2 ng/ml/h, p<0.01) but plasma AT II and aldosterone (Ald) were not significantly changed (13±3 to 17±3 pg/ml and 365±125 to 312±132 pg/ml, respectively). These results suggested that the beneficial effect of imidapril against left ventricular mass was independent of BP lowering effect. Zannad et al. reported that no significant benefit was found with fosinopril (5-20 mg/day) for the prevention of cardiovascular events such as cardiovascular death, resuscitated death, nonfatal stroke, heart failure, myocardial infarction, or revascularization in HDD-CKD patients; however, there was a trend that fosinopril treatment may be associated with a lower risk of cardiovascular events after adjustment for risk factors. These lines of evidence suggested that ACE inhibitors are effective in controlling BP and preventing CVD in HDD-CKD patients. It is highly likely that the cardioprotective effects of ACE inhibitors are independent of their BP lowering effect in HDD-CKD patients.

#### **Angiotensin receptor blockers in HDD-CKD patients**

Saracho et al. reported on a multicenter, open 6 month study designed to test the tolerability and efficacy of losartan as an antihypertensive in 406 hypertensive HDD-CKD patients who were previously untreated, treated but uncontrolled, or treated with poor tolerability. There were significant reductions in pre- and postdialysis SBP and DBP at 3 months (pre SBP/DBP: 155 ± 15/84± 9 mmHg, post SBP/DBP: 140 ± 19/78± 10 mmHg) and 6 months (pre SBP/DBP: 152 ± 16/83± 9 mmHg, post SBP/DBP: 139± 18/77± 9 mm Hg) compared with those at baseline (pre SBP/DBP: 163 ± 16/88± 10 mmHg, post SBP/DBP: 148± 18/82± 9 mmHg). Shibasaki et al. reported that losartan (50 mg/day) reduced left ventricular mass index (-24.7±3.2%) after 6 months more than a calcium channel antagonist, amlodipine (5 mg/day: -10.5±5.2%), or an ACE inhibitor, enalapril (5 mg/day: -11.2±4.1%), although all three groups had similar decreases in mean BP. Kanno et al. reported that losartan (100 mg/TIW) reduced left ventricular hypertrophy (145±5 to 122±3 g/m2, p<0.001) after 12 months in 24 diabetic patients on HD therapy whereas a placebo group showed no change. In this study, SBP and DBP were controlled below 140/90 mmHg with a calcium antagonist, benidipine (4-12 mg/day), and an α-blocker, doxazosin (2-4 mg/day). SPB was significantly decreased in both placebo and losartan groups after 12 months.These results suggested that this beneficial effect of losartan for left ventricular hypertrophy was independent of a BP lowering effect. Takahashi et al. stated that candesartan (4-8 mg/day) reduced cardiovascular events and mortality compared with placebo (16.3% vs. 45.9 and 0.0 vs. 18.9%, respectively) after 19.4±1.2 months in HDD-CKD patients in a stable condition and with no clinical evidence of cardiac disorder. In this study, the brain natriuretic peptide (BNP) level did not differ between the two groups at enrolment, whereas in patients who had not experienced a cardiovascular event at 12 months, the BNP levels were significantly increased in the control group but not in the candesartan group. Suzuki et al. reported that angiotensin receptor blockers (ARBs) (valsartan: 160 mg/day, candesartan: 12 mg/day, or losartan: 100 mg/day) treatment was independently associated with reduced cardiovascular events (hazard ratio, 0.51; 95% confidence interval, 0.33 to 0.79; p=0.002) compared with no ARB treatment group for 180 HDD-CKD patients in each group during a 3 year observation period. There were 34 (19%) fatal or nonfatal CVD events in the ARBs group and 59 (33%) in the no ARB group. BP did not differ between the ARBs group and the no ARB group during the observation period. After adjustment for age, sex, diabetes, and SBP, treatment with an ARB was independently associated with reduced fatal and nonfatal CVD events (hazard ratio, 0.51; 95% confidence interval, 0.33 to 0.79; p=0.002). These results demonstrated that ARBs are effective in controlling BP and preventing CVD in HDD-CKD patients. We predict that the cardioprotective effects of ARBs may be independent of their BP lowering effect in HDD-CKD patients.
