BP in humans is determined both by the cardiac output and by peripheral vascular resistance. Cardiac output is dependent on the intravascular component of the extracellular space as well as the heart rate; and peripheral resistance is dependent on functions of the vascular endothelial and smooth muscle cells in response to the actions of various vasoactive mediators (both vasoconstrictors and vasodilators). Although alteration of either cardiac output or vascular resistance would initially be expected to affect BP in a concordant direction, a healthy kidney can adapt to short-term changes in BP to restore normotension. An increase in renal sodium excretion is the expected response to increased BP in healthy individuals. This pressure natriuresis allows for the restoration of extracellular volume and BP following an increase in cardiac output. However, in CKD, this homeostatic mechanism is impaired, and the kidney fails to sufficiently excrete sodium loads. In CKD and ESRD, it is hypothesized that an increase in cardiac output initiates the increase in BP, but ultimately, an increase in vascular resistance sustains the BP elevation.³³,³⁴ Accordingly, addressing both volume status and the degree of peripheral vasoconstriction are necessary to control BP in these patient populations.

Sodium balance is an important aspect in BP control in CKD and ESRD. Because of the kidney’s ability to excrete sodium, healthy individuals can tolerate large amounts of sodium intake without significant increases in BP.³⁵ However, in the presence of kidney disease, BP is highly dependent on extracellular volume. Rats that have undergone 70% renal ablation develop severe hypertension on a high sodium diet, but hypertension is completely prevented in these animals while on a low sodium diet.³⁶ A lower hematocrit found in the animals on a high sodium diet suggested that extracellular volume overload was present and likely the mechanism responsible for the increased BP. Another experiment with a renal ablation model demonstrated that increased cardiac output is responsible for the initial increase in BP during acute salt loading, but increased peripheral resistance (which occurs after BP has already become increased) is responsible for the maintenance of elevated BP even after cardiac output has returned to normal.³⁷

Because CKD patients are limited in the amount of renal sodium excretion, BP and sodium balance are interrelated. Among CKD patients, those with more severe renal impairment experience greater increases in BP in response to a sodium load. Although the pressure natriuresis curve serves to maintain sodium balance in healthy individuals, CKD patients require much higher BP increases than healthy individuals to augment their renal sodium excretion. Similarly, the degree of blood volume expansion required to induce a natriuresis is higher in patients with severe renal impairment compared to those with moderate renal impairment. Overall, these findings show that the effects on BP related to sodium intake (“salt sensitivity”) are augmented with the progression of renal disease.³⁸ During the early stages of CKD, despite suboptimal renal sodium excretion, some patients remain normotensive despite an increase in cardiac output because of a reduction in peripheral resistance.³³ Although BP is sustained in the normotensive range, these patients, who are already limited in their sodium excretion because of reductions in GFR, fail to induce the necessary pressure natriuresis to re-establish sodium balance. Consequently, further increases in sodium loading result in increased cardiac output, peripheral resistance, and BP.

When subjects with CKD increase dietary sodium ingestion from 20 to 120 mEq per day, they experience an increase in BP that is not seen even in healthy subjects who increase their dietary sodium ingestion to levels as high as 1200 mEq per day.³⁹ Although both groups have a similar suppression of RAAS at moderate amounts of sodium ingestion, plasma renin activity (PRA) and angiotensin (Ang) II decreased drastically in healthy subjects during periods of high sodium ingestion (1200 mEq per day).³⁹ Although vascular resistance could not be measured directly in this study, increased vasoconstriction in the CKD subjects appeared to be the likely mechanism responsible for the increased BP and the reduced distribution of fluid into the interstitial space. Even under clinical conditions of pharmacologic RAAS blockade, salt balance has an important role in BP. In nondiabetic proteinuric subjects with creatinine clearance ranging from 33 to 110 mL per minute ingesting a low sodium diet, a reduction in BP and proteinuria during the chronic administration of fixed doses of angiotensin converting enzyme (ACE) inhibitors is reversed during periods of increased dietary sodium ingestion.⁴⁰ As further evidence of the importance of sodium intake in relation to BP and proteinuria in CKD patients, the coadministration of hydrochlorothiazide during a high salt intake period reduces both BP and proteinuria back to values seen during the low sodium diet.

In ESRD patients with little or no renal sodium excretion, sodium removal through ultrafiltration during HD is the primary means to maintain extracellular volume status because a pressure natriuresis is not possible. Consequently, compared to healthy controls, HD patients have a significantly higher cardiac output.⁴¹ However, the presence of hypertension among HD patients is also highly dependent on elevations in vascular resistance.⁴¹ Extracellular volume potentially impacts both of these parameters to influence BP. When subjected to sodium loading, HD patients typically respond with an increase in both BP and cardiac output. The pattern of vascular resistance following sodium loading is more variable, but increases typically do not occur until after the elevated cardiac output has already led to an
increase in BP.³⁴,⁴² In between HD treatments, during the interdialytic period, HD patients gradually gain weight with their routine dietary intake of sodium and water. The BP patterns during a typical interdialytic period show rhythmic oscillations superimposed on a general linear increase in BP over time.⁴³ This pattern is modified by interdialytic weight gain such that greater weight gain is associated with an increase in the slope of BP rise.⁴⁴ Subsequently, greater increases in interdialytic weight gain have been associated with increased pre-HD systolic BP at the next HD treatment.⁴⁵ However, that increased interdialytic weight gain is also associated with a greater reduction in BP during the course of that treatment (likely as a response to the ultrafiltration required to remove the interdialytic fluid gain).⁴⁵ Such evidence supports the hypothesis that extracellular volume (through either increases in cardiac output or vascular resistance) is primarily responsible for hypertension in this population. Consequently, one HD center has described a 98% success rate in withdrawing antihypertensive medications while using the ultrafiltration that can be achieved during a cumulative weekly dialysis time of 24 hours.⁴⁶ Thus, recognition and successful attainment of a patient’s dry weight can facilitate the initial BP management in most, but not all, HD patients.

Therefore, HD treatment with adequate ultrafiltration should normalize BP in many HD patients. However, the basis for this practice assumes that interdialytic sodium and fluid intake is not excessive. An insufficient time on HD needed to completely restore normal extracellular fluid volume and achieve dry weight without inducing symptomatic hypotension limits this practice. A cross-sectional study found that major differences between ultrafiltration-sensitive and ultrafiltration-resistant HD patients were the pre-and post-HD atrial natriuretic peptide (ANP) levels between groups.⁴⁷ The ultrafiltration-sensitive group had reductions in ANP during HD (as well as lower pre-HD ANP), whereas the higher pre-HD ANP levels in the ultrafiltration-resistant group persisted despite similar amounts of ultrafiltration, suggesting that this group had remaining extracellular volume contributing to the elevated BP. It should be recognized, however, that extracellular volume overload may not always manifest itself overtly. In some cases, patients may have additional extracellular volume despite appearing euvolemic on clinical exam. The Dry Weight Reduction in Hypertensive Hemodialysis Patients (DRIP) study was a randomized clinical trial in which hypertensive HD subjects were randomized to either continue their current HD and ultrafiltration prescription or have their dry weight challenged during each HD treatment over several weeks by 0.1 kg per 10 kg dry weight until symptoms developed. The subjects randomized to additional ultrafiltration demonstrated significant decreases in ambulatory systolic BP after 4 weeks and 8 weeks.⁴⁸

The dialysate used during HD is another important factor that may contribute to hypertension in ESRD patients. Although ultrafiltration effectively removes water and sodium concurrently through convection, there may be an additional exchange of sodium from the dialysate to the patient’s plasma depending on the sodium concentration gradient between the two compartments. Directly programmable ultrafiltration enables greater convective sodium removal in shorter periods of time, but increases the risk of intradialytic hypotension related to the abrupt hemodynamic changes. Although the use of higher dialysate sodium concentrations may reduce the risk of intradialytic hypotension, it may increase the risk of hypertension.⁴⁹ In contrast, the use of individualized dialysate sodium concentrations (as opposed to standardized concentrations, which may exceed the pre-HD plasma sodium of the patient) have been associated with lower pre-HD systolic BP in the context of decreased thirst and interdialytic weight gain.⁵⁰

The RAAS has local and systemic effects that control BP by altering renal sodium reabsorption and vascular resistance. Renin released from juxtaglomerular cells cleaves angiotensinogen to Ang I, which is ultimately converted to Ang II by the ACE. Ang II causes vasoconstriction upon binding to angiotensin type 1 (AT1) receptors in vascular smooth muscle cells (VSMCs). Ang II increases proximal tubular reabsorption of sodium and stimulates aldosterone release from the adrenal gland, which is responsible for further sodium reabsorption in the distal nephron. Although this proposed sequence applies to the levels and activity of RAAS components in the plasma, there is also local RAAS activity. Consequently, measurements of RAAS mediators in the plasma may not always identify a disruption of the axis at the tissue level.⁵¹

Early evidence for systemic RAAS activation in CKD stems from studies in patients with autosomal dominant polycystic kidney disease (ADPKD). It has been shown that PRA and plasma aldosterone were higher in hypertensive ADPKD patients compared to patients with essential hypertension, and plasma aldosterone is higher in normotensive ADPKD patients compared to healthy controls (despite similar creatinine clearance between the groups).⁵² Autosomal dominant polycystic kidney disease patients also manifest an accentuated response to ACE inhibitors compared to unaffected family members, further supporting the role of RAAS in the hypertension seen in ADPKD even prior to the onset of significant renal impairment.⁵³ However, the findings from these studies may not be entirely generalizable to more heterogeneous groups of CKD patients because the proposed renal ischemia induced by large cyst formation in ADPKD does not necessarily apply to CKD from other etiologies. There is evidence that PRA and aldosterone are higher in hypertensive CKD patients compared to healthy controls, essential hypertension patients, or even normotensive CKD patients.⁵⁴,⁵⁵ However, there is also evidence that CKD patients have lower PRA compared to healthy
controls, although the values were similar to subjects with essential hypertension and normal renal function.⁵⁶ Despite these conflicting findings, increased intrarenal RAAS activity has been described in patients with hypertension and varying etiologies of CKD, including immunoglobulin A (IgA) nephropathy, membranous nephropathy, and diabetic nephropathy.⁵⁷,⁵⁸,⁵⁹

As mentioned previously, success in managing BP can be achieved in many ESRD patients by appropriately using ultrafiltration for volume removal.⁴⁶ However, there are HD patients who remain quite hypertensive despite achieving their estimated dry weight. Previous reports have shown both increased or normal PRA in groups of HD patients compared to controls, and PRA did not consistently correlate with BP.⁶⁰,⁶¹ However, increased PRA has been demonstrated in HD patients with hypertension that is ultrafiltration resistant compared to those whose BP responds to ultrafiltration.⁶¹ Bilateral nephrectomy, a procedure previously used for ultrafiltration-resistant hypertension in ESRD patients, has been shown to reduce PRA, Ang I and II, along with reductions in BP in these ultrafiltration-resistant hypertensive patients.⁶²

The SNS has been studied extensively as a possible contributor to hypertension in CKD and ESRD patients. In CKD patients, there is evidence for multiple pathways that the SNS may affect to increase BP. In support of the previous discussion of the effects of salt balance on BP, renal sympathetic nerve stimulation increases proximal tubular reabsorption of sodium and water.⁷⁰ Additionally, the systemic effects of the SNS include increased cardiac output and vasoconstriction. Although evidence of activated SNS activity based on elevated catecholamine levels in CKD is inconsistent and possibly confounded by decreased renal clearance of these compounds, other estimates of SNS, such as muscle sympathetic nerve activity (MSNA), confirm the hyperactive state in CKD and ESRD. The mechanisms responsible for hyperactive SNS are thought to be related to increased renal nerve activity. The kidney possesses baroreceptors and chemoreceptors that increase renal nerve firing secondary to pressure changes or metabolites produced in response to ischemia
or uremia. Animal models of renal artery stenosis, arterial ligation causing partial renal ablation, or intrarenal phenol injection all reveal increased nerve activity.⁷¹,⁷²,⁷³ Activation of renal nerves results in a centrally mediated hypertension via activation of the SNS, which can ultimately be interrupted by a blockade of neural signals. For example, Sprague-Dawley rats that have undergone a 5/6 nephrectomy experience an attenuation in hypertension following a dorsal rhizotomy with a concurrent reduction in the turnover of norepinephrine in the hypothalamic nuclei and locus coeruleus.⁷² In humans, there is evidence for SNS activation even prior to overt renal impairment if a potential cause for renal ischemia is present. Hypertensive ADPKD patients have increased MSNA compared to controls despite preserved renal function in both groups.⁷⁴ Muscle sympathetic nerve activity was even higher in ADPKD patients with decreased GFR, but was not different between normotensive ADPKD patients and controls. Given the experimental association with bilateral renal ischemia, the possibility exists that the increased MSNA is related to increases in RAAS activation. One study confirmed the presence of increased MSNA in CKD patients, and showed that the administration of the ACE inhibitor enalapril caused a reduction in MSNA (relative to the decrease in BP it caused), whereas amlodipine increased MSNA.⁷⁵ In a small study including patients with various etiologies of nondiabetic CKD and matched control, MSNA changed similarly in the patients and controls with variability in volume status, but MSNA was persistently higher in the CKD patients.⁷⁶

The evidence for the role of increased SNS activity in the etiology of hypertension in renal disease is best supported by the findings of Converse et al.⁷⁷ Twenty years ago it was established that ESRD patients had increased MSNA compared to healthy controls and other ESRD patients that had undergone bilateral nephrectomies. Higher mean arterial BP was also seen in the ESRD patients whose native kidneys were still present, and the hypothesis was that afferent signaling from the ischemic kidneys modulated an overall SNS response best managed by surgical removal of the kidneys. This is further supported by the fact that reversing the uremic state of ESRD patients with renal transplantation does not decrease MSNA, but a native kidney nephrectomy in the transplant recipient does.⁷⁸ The fact that nondiabetic HD patients have a normal arterial and cardiopulmonary baroreflex response weakens any suggestion that uremia-induced impairments in these reflexes might further contribute to abnormally elevated SNS activity in this population.⁷⁹