Chapter 9: Hypertension and kidney disease

Hypertension is a strong, independent risk factor for the development and progression of chronic kidney disease (CKD) to end-stage kidney disease (ESKD). Hypertensive kidney disease itself is the second most common cause of ESKD after diabetic kidney disease.

Assessment rests on two pillars: estimating kidney function with eGFR, traditionally calculated using the 2009 CKD-EPI equation, and detecting kidney injury with the urine albumin-to-creatinine ratio (UACR), preferably from a first-morning urine sample. eGFR and UACR independently and additively predict CKD progression and cardiovascular risk.

Albuminuria is classified as normal or mildly increased (UACR<30 mg/g), moderately increased (30–300 mg/g, formerly microalbuminuria) or severely increased (>300 mg/g, formerly macroalbuminuria). CKD is diagnosed when eGFR remains below 60 mL/min/1.73 m² at any albuminuria level, or UACR remains above 30 mg/g at any eGFR, for more than three months.

Serum creatinine, eGFR and UACR should be documented during the initial hypertension assessment when CKD is present and repeated at least annually. A negative dipstick does not exclude low-level albuminuria because it cannot detect all UACR values. Urinalysis remains useful for haematuria, active sediment and other signs of kidney injury.

Treating hypertension in CKD

Hypertension is the most common modifiable factor driving CKD progression. True resistant hypertension, nocturnal hypertension and masked hypertension are common in CKD and are associated with lower eGFR, greater albuminuria and more HMOD.

CKD also amplifies several kidney-derived mechanisms that raise pressure, including sodium sensitivity, sodium and water retention, activation of the renin–angiotensin and sympathetic systems, and impaired endothelium-dependent vasodilation. Secondary hyperparathyroidism, calcium–phosphate loading with vascular calcification, more frequent sleep apnoea, erythropoietin, glucocorticoids and calcineurin inhibitors may contribute in later CKD.

Blood-pressure targets

Available evidence suggests that lowering blood pressure with any major antihypertensive class can provide similar protection against major cardiovascular events and all-cause death in people with CKD. The most protective target has nevertheless been debated for more than a decade, including in CKD with diabetes.

Older observational data linked ESKD risk to systolic pressure above 120 mmHg. A prospective five-year Chinese study in untreated CKD associated pressure above 130/90 mmHg with substantially higher cardiovascular and renal risk, but did not establish the most protective treatment target.

In MDRD, people with non-diabetic CKD randomised to lower or usual mean-pressure targets had no significant difference in three-year GFR decline, ESKD or death. Baseline proteinuria modified the result: in people excreting more than 1 g of protein per day, the lower target reduced protein excretion and slowed GFR decline. AASK similarly found no overall difference between target groups, but a small subgroup with proteinuria above 1 g/day appeared to benefit from the lower target.

Long-term follow-up of MDRD linked the lower target to lower ESKD or ESKD-plus-death risk, mainly in people with baseline proteinuria above 1 g/day. Long-term AASK follow-up found no overall difference in creatinine doubling, ESKD or death, but a benefit of the lower target in people with UPCR above 0.22 g/g, approximately 0.25–0.3 g/day of protein.

A pooled analysis of these trials, involving 1,907 people followed for an average of 14.9 years, associated the lower target with lower ESKD and mortality risk in the overall population; the effect was mainly driven by people with UPCR above 0.44 g/g. Sustained blood-pressure reduction and the degree of proteinuria therefore appear to be major determinants of kidney protection in non-diabetic CKD.

SPRINT randomised 9,361 high-risk people to an intensive systolic target below 120 mmHg or a standard target below 140 mmHg. About 28% had CKD with eGFR 20–60 mL/min/1.73 m², but people with proteinuria above 1 g/day or 1 g/g, diabetes or previous stroke were excluded. The primary cardiovascular outcome was lower with intensive treatment, while kidney outcomes did not differ. The CKD subgroup had lower all-cause mortality with intensive treatment but no significant difference in the primary or prespecified kidney outcome. SPRINT was not designed or powered for kidney outcomes, which were few, so it cannot justify a universal systolic target below 120 mmHg in CKD.

Direct evidence for the optimal target in type 2 diabetes with CKD is limited. UKPDS38 and the diabetic subgroup of HOT informed diastolic targets. ACCORD-BP compared systolic targets below 120 and below 140 mmHg in high-risk type 2 diabetes but excluded serum creatinine above 1.5 mg/dL. Post-hoc RENAAL and IDNT analyses linked higher baseline systolic pressure with higher ESKD or creatinine-doubling risk. A meta-analysis of stage 3–5 CKD suggested mortality benefit with a systolic reduction of about 16 mmHg to an achieved pressure around 132 mmHg, but little benefit below 125 mmHg. A later synthesis of AASK, ACCORD, MDRD and SPRINT suggested a non-significant mortality trend with intensive treatment; after excluding people with higher GFR and those receiving intensive glycaemic control, pressure below 130 mmHg was associated with lower mortality than a target below 140 mmHg.

Putting these indirect data together, a systolic target below 130 mmHg and diastolic target below 80 mmHg, if well tolerated, may protect people with CKD and albuminuria above 30 mg/g, including proteinuric diabetic kidney disease. A similar target may reduce mortality in many people with CKD. In advanced stage 4–5 CKD, eGFR should be monitored because a further functional but reversible fall can occur when pressure is lowered.

A clinic target below 120/70 mmHg cannot be recommended because evidence is lacking. Important limitations include the exclusion of diabetic CKD from trials comparing targets, the use of mean-pressure rather than directly transferable systolic/diastolic targets in MDRD and AASK, the relatively young trial populations, and the fact that long-term benefits of lower targets were clearest in people with proteinuria.

Antihypertensive medication

People with CKD should receive lifestyle measures, especially sodium restriction, because a low-sodium diet reduces protein excretion in proteinuric CKD. Achieving the recommended target usually requires combination therapy: a renin–angiotensin-system blocker plus a calcium-channel blocker or thiazide/thiazide-like diuretic when eGFR is around 45 mL/min/1.73 m² or higher. Below 30 mL/min/1.73 m², a loop diuretic generally replaces a thiazide; the transition between 30 and 45 should be individualised.

Trials in diabetic and non-diabetic CKD support an ACE inhibitor or ARB as the first treatment choice, particularly with moderate or severe albuminuria, because these drugs reduce albuminuria, slow GFR decline and reduce creatinine doubling or ESKD. They should be used at the highest tolerated dose for kidney protection. Dual ACE-inhibitor/ARB blockade should be avoided because trials were stopped early for excess adverse events.

With normal-range albuminuria, ACE inhibitors and ARBs can delay progression to severe albuminuria, but it is not established that they preserve kidney function better than other major classes. In a trial randomising people to stop or continue RAS blockade, stopping was not associated with a significant long-term difference in eGFR decline.

Special treatment challenges

ACE inhibitors and ARBs dilate the efferent arteriole and reduce intraglomerular pressure. An initial eGFR fall of 10–15% during the first weeks is common, as can occur after a large pressure reduction with any drug. eGFR and electrolytes should be checked repeatedly during the first 4–8 weeks, according to baseline function. A modest early fall is not necessarily harmful; a persistent or greater-than-30% decline should prompt stopping RAS blockade and evaluation for renovascular disease.

RAS blockade also increases hyperkalaemia risk. Hyperkalaemia is associated with higher mortality and is a common reason for reducing or stopping ACE inhibitors or ARBs, but reducing or stopping them may increase cardiovascular risk. The newer potassium binders patiromer and sodium zirconium cyclosilicate can normalise and chronically maintain potassium in people with CKD taking ACE inhibitors, ARBs or spironolactone. They may help maintain potassium below 5.5 mmol/L.

Most people with CKD will not reach target pressure with ACE-inhibitor or ARB monotherapy. A dihydropyridine calcium-channel blocker or diuretic, and often both, should be included. Dihydropyridine blockers can increase proteinuria when used without RAS blockade in proteinuric CKD, but in the wider hypertensive population, where albuminuria is absent or mild, their kidney outcomes are broadly similar to RAS blockers or diuretics. A study with 19% moderate and 5% severe albuminuria found RAS blockade plus a dihydropyridine calcium-channel blocker superior to RAS blockade plus a thiazide for kidney outcomes.

Diuretics are particularly useful in CKD because sodium sensitivity and resistant hypertension are common. They can also reduce proteinuria when added to RAS blockade. Below eGFR 45 mL/min/1.73 m², thiazides may become less effective, although a randomised study of 160 people with stage 4 CKD found that chlorthalidone, at a mean dose of 23 mg/day, reduced 24-hour systolic pressure by 10.5 mmHg.

In stage G3b CKD, with eGFR 30–44, diuretic choice and dose should be individualised. In stage 4 CKD with eGFR below 30, a loop diuretic is generally preferred; torsemide may be more convenient than furosemide because its longer half-life permits less frequent dosing.

When triple therapy is insufficient, beta-blockers and alpha-1 blockers may help because sympathetic activity is increased in CKD, although their kidney effects have not been tested in dedicated renal-outcome trials. In PATHWAY-2, spironolactone was particularly effective as a fourth drug, but people with eGFR below 45 or potassium above 4.5 mmol/L were excluded. In AMBER, people with resistant hypertension and eGFR 25–45 achieved blood-pressure reduction with spironolactone; hyperkalaemia at 12 weeks occurred in about 60% with placebo and 35% with patiromer. Spironolactone is therefore not generally recommended as a fourth drug from stage 3b CKD onward except in selected circumstances, with potassium binders considered to maintain potassium below 5.5 mmol/L.

Additional kidney- and cardiovascular-protective treatment

Beyond blood-pressure control, SGLT2 inhibitors and the non-steroidal mineralocorticoid-receptor antagonist finerenone can slow CKD progression and reduce cardiovascular risk, although they are not licensed simply as antihypertensive drugs.

SGLT2 inhibitors reduce proximal tubular glucose reabsorption. Early diabetes studies showed a typical 3–5/1–2 mmHg reduction in systolic/diastolic pressure in hypertension, with reductions also seen on ambulatory monitoring and a larger systolic effect, about 7 mmHg, in stage 4 CKD. Mild natriuresis, diuresis and osmotic diuresis are likely mechanisms. Albuminuria falls by approximately 25–40%, depending on baseline level, and plasma uric acid may decrease.

CREDENCE included 4,401 people with type 2 diabetes, CKD and severely increased albuminuria; DAPA-CKD included 4,304 people with diabetic or non-diabetic CKD and UACR above 200 mg/g; EMPA-KIDNEY included 6,609 people with diabetic or non-diabetic CKD across eGFR 20 to below 45, or eGFR 45 to below 90 with UACR above 200 mg/g. All three trials were stopped early because kidney composite and individual outcomes were substantially reduced compared with placebo.

In EMPA-KIDNEY, benefit was present across the eGFR range and most pronounced with severe albuminuria; eGFR loss was slower with empagliflozin across UACR subgroups. The early small eGFR fall is managed similarly to the initial fall with RAS blockade. CREDENCE and DAPA-CKD also reduced some cardiovascular risks, and DAPA-CKD reduced mortality, a benefit not previously clear with RAS blockade alone.

Steroidal mineralocorticoid antagonists reduce albuminuria when added to an ACE inhibitor or ARB in proteinuric diabetic CKD, through anti-inflammatory and antifibrotic effects, but their use is limited by hyperkalaemia and a lack of dedicated kidney-outcome evidence. Finerenone has different duration and tissue distribution, with less blood-pressure reduction and less potassium increase than steroidal agents.

In FIDELIO-DKD, involving 5,734 people with type 2 diabetes, CKD and moderate or severe albuminuria, finerenone added to ACE-inhibitor or ARB treatment reduced kidney failure, a sustained eGFR decline of more than 40%, renal death and cardiovascular outcomes. The between-group systolic/diastolic difference was 2.7/1.0 mmHg. Hyperkalaemia causing discontinuation occurred in 2.3% with finerenone and 0.9% with placebo, with no fatal hyperkalaemia events reported.

In FIGARO-DKD, involving 7,437 similar participants, finerenone reduced the composite of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke or hospitalisation for heart failure by 13%, with consistent kidney benefit and tolerability. An on-treatment analysis combining both trials found an 18% reduction in mortality. Esaxerenone and apararenone also reduce albuminuria in phase 2 CKD studies, but have not yet been tested in pivotal kidney-outcome trials.

On current evidence, SGLT2 inhibitors or finerenone should be considered in addition to lifestyle measures and antihypertensive treatment. SGLT2 inhibitors are relevant in diabetes and in non-diabetic CKD with moderate or severe albuminuria; finerenone is recommended for diabetic kidney disease with moderate or severe albuminuria. The order of adding them was not tested in trials and should depend on individual needs, including glycaemic control, potassium and persistent albuminuria.

Renal-artery stenosis and renovascular disease

Renovascular hypertension accounts for approximately 2–5% of all hypertension and up to 30% of secondary hypertension. Atherosclerotic renal-artery disease and fibromuscular dysplasia traditionally account for about 90% and 10%, respectively, although prevalence is higher in selected groups such as resistant hypertension.

Atherosclerotic renal-artery disease has been reported in 6.8% of people over 65, 10–12% of people with ESKD, 15–30% with coronary disease and up to 50% with heart failure. Fibromuscular dysplasia ranges from below 1% to 6% in observational studies. Prognosis is poor because cardiovascular-event rates after diagnosis can exceed those of the general population by three- to six-fold. Non-dialysis CKD patients with atherosclerotic renovascular disease have about 1.5-fold higher mortality than people with other CKD causes, and mortality may be threefold higher in those receiving dialysis.

Prognosis varies with phenotype. People with acute pulmonary oedema have two- to threefold higher cardiovascular-event and mortality risk than lower-risk phenotypes without pulmonary oedema, resistant hypertension or acute kidney injury.

Balloon angioplasty without stenting is preferred for haemodynamically important fibromuscular dysplasia. Trials comparing standard medical treatment with medical treatment plus renal angioplasty in atherosclerotic disease found no major difference in pressure, cardiovascular risk or kidney outcomes, but were criticised for non-standard inclusion, weak stenosis assessment, enrolment delays, protocol changes, crossover and low event rates. They mainly included mild or asymptomatic disease and excluded patients with pulmonary oedema, resistant hypertension or rapidly falling kidney function.

Observational studies in severe, well-documented disease, often stenosis above 70%, and high-risk clinical presentations report substantial benefit from revascularisation in blood-pressure control, kidney preservation and cardiovascular outcomes. Current consensus is to add revascularisation to medical treatment in documented secondary hypertension or a high-risk presentation with a confirmed stenosis above 70%.

Medical treatment alone is appropriate for asymptomatic atherosclerotic stenosis below 70%, mild or moderate hypertension easily controlled with medication, low-grade stenosis or a non-viable kidney. If ACE-inhibitor or ARB initiation produces a 30% eGFR fall, the patient requires careful reassessment.

Hypertension after kidney transplantation

Kidney transplantation is the preferred form of kidney replacement therapy for ESKD because it improves survival and quality of life compared with dialysis. Residual cardiovascular risk nevertheless remains much higher than in the general population, and cardiovascular disease is the leading cause of death in the decade after transplantation.

Transplantation itself lowers blood pressure by approximately 8/5 mmHg on 24-hour monitoring in the short and medium term and reduces the need for antihypertensive drugs. Despite this, hypertension is the most common comorbidity in recipients; ambulatory studies find it in more than 95%.

Hypertension is associated with poorer kidney function, HMOD, cardiovascular events, reduced graft function and lower patient survival. Office measurement misclassifies control frequently: masked hypertension occurs in 20–40%, reduced dipping in about 50% and nocturnal hypertension in up to 70–80%. Ambulatory pressure predicts graft dysfunction and HMOD better than office pressure and is recommended for management.

Causes are multifactorial, including conventional risks, CKD-related sodium retention and activation of RAS and the sympathetic system, and transplant-related treatment. Purine-pathway inhibitors such as mycophenolate and azathioprine and mTOR inhibitors such as everolimus and sirolimus do not generally affect pressure. Steroids raise pressure, and steroid-free or low-dose protocols are associated with better control.

Calcineurin inhibitors, cyclosporine and tacrolimus, raise pressure through sodium reabsorption, vasoconstrictor release, increased peripheral resistance, afferent arteriolar constriction and, particularly with cyclosporine, sympathetic activation. Tacrolimus appears to have less pronounced effects than cyclosporine.

Because no transplant-specific randomised target trials exist, targets are extrapolated from CKD. A systolic target below 130 mmHg is reasonable. Lifestyle measures should follow CKD recommendations and most recipients require combinations of major antihypertensive classes.

Specific ACE-inhibitor or ARB benefits remain uncertain. A meta-analysis found a 38% lower risk of graft loss with these drugs across nine studies and 1,246 participants, without a significant effect on non-fatal cardiovascular events or death but with more hyperkalaemia. Dihydropyridine calcium-channel blockers have consistently been associated with improved graft survival and may counteract calcineurin-inhibitor vasoconstriction. In the same meta-analysis, calcium-channel blockers reduced graft-loss risk by 42% across 16 studies and 1,327 participants and increased GFR by 11.11 mL/min compared with ACE inhibitors or ARBs, so they may be preferred early after transplantation.

Thiazide or thiazide-like diuretics can also be useful because they counter cyclosporine-mediated sodium retention. Long-term effects of antihypertensive classes on graft loss remain incompletely defined. Transplant renal-artery stenosis is not uncommon and should be actively sought in sudden or uncontrolled hypertension; percutaneous transluminal renal angioplasty has high technical success in this setting.

Guidelines and studies

• The KDIGO 2021 BP guideline for CKD is the principal kidney-specific reference.
• SPRINT included a CKD subgroup, but it was not designed primarily to answer every kidney-outcome question.
• ACCORD-BP informs the discussion of intensive treatment in diabetes and kidney risk; its results should be interpreted in the population studied.
• PATHWAY-2 supports spironolactone for resistant hypertension, with extra caution in CKD.