Chapter 5: Assessment of hypertension-mediated organ damage

Hypertension-mediated organ damage (HMOD)

HMOD means structural or functional changes in large or small arteries or in target organs—the brain, heart, kidneys and eyes—caused by elevated blood pressure. It is a marker of cardiovascular risk and of preclinical or silent kidney and cardiovascular disease.

HMOD is common in severe or long-standing hypertension, but it can also occur with apparently lower-risk hypertension. As imaging has become more widely used, silent HMOD is detected more often. Cardiovascular risk increases when HMOD is present and rises further when damage involves several organs or functions.

Some forms of HMOD can regress with antihypertensive treatment, particularly when treatment starts early. With long-standing hypertension, however, damage may become irreversible even when pressure is controlled. Treatment remains important because it can delay further progression and counter the increase in cardiovascular risk.

Technical limitations and cost may restrict HMOD assessment in some countries. At least a basic screen should nevertheless be considered in every person with hypertension, with more detailed assessment when the result will influence treatment decisions.

HMOD in the heart

Hypertension exposes the heart to increased pressure load and causes structural and functional changes. Early changes may be silent but are powerful predictors of later cardiovascular events, including HFpEF, HFrEF, atrial fibrillation, coronary artery disease, sudden death and stroke.

Preclinical hypertensive heart disease includes left-ventricular hypertrophy or altered ventricular geometry, impaired diastolic or systolic function, left-atrial enlargement and more arrhythmias. A comprehensive assessment should use electrocardiography and the available imaging techniques to evaluate most or all relevant indicators.

Left-ventricular mass and geometry

As noted elsewhere, left-ventricular hypertrophy diagnosed by two-dimensional transthoracic echocardiography is more sensitive than ECG. LVH is also a major predictor of morbidity and mortality in hypertension and in the general population. Adding LVH to conventional cardiovascular risk factors substantially reclassifies risk in many, although not all, studies. In a population sample, adding LVH to the commonly used SCORE risk model significantly improved risk reclassification in people with hypertension.

Among several echocardiographic indices, left-ventricular mass index has been the most consistent and perhaps the most important predictor of adverse events in hypertension. Antihypertensive treatment can cause LVH regression, but only a minority of people with this HMOD show substantial regression; when regression occurs it is associated with a better prognosis.

When LVH is detected, periodic follow-up is recommended, preferably with echocardiography, to monitor ventricular hypertrophy and other structural and functional changes. The goal is regression of LVH. This may take years and is not always achieved; regression is less frequent in women and in people with diabetes, obesity or African ancestry.

HMOD in the arteries

Carotid intima–media thickness and plaque

Carotid intima–media thickness (cIMT) combines the thickness of the intimal and medial layers and is measured by ultrasound. It can be viewed as an early marker of atherosclerosis. IMT at the carotid bifurcation is thought to reflect atherosclerosis more closely, whereas common-carotid IMT may reflect hypertensive thickening. A carotid IMT above 0.9 mm is generally considered abnormal, although the upper limit of normal varies with age.

The value of cIMT for quantitative cardiovascular-risk assessment is debated. In some studies, adding cIMT did not improve risk prediction or improved it only slightly. The prognostic meaning of treatment-related IMT change is also uncertain. A meta-analysis of 119 RCTs involving 100,667 patients with a mean age of 62 years found that progression of cIMT over a mean 3.7-year follow-up was reduced by combinations of antihypertensive, lipid-lowering and diabetes treatments as well as dietary and other interventions. Across all interventions, each 10-mm-per-year reduction in IMT progression was associated with a 9% reduction in cardiovascular risk. However, cIMT changes analysed separately for antihypertensive, lipid-lowering or diabetes treatment did not show a consistent independent relationship with cardiovascular risk.

Carotid plaque is defined by an IMT above 1.5 mm, a focal increase in thickness of more than 0.5 mm, or a focal thickness more than 50% greater than the surrounding carotid IMT. Carotid plaque predicts stroke and myocardial infarction independently of traditional risk factors and risk scores, and predicts future myocardial infarction better than IMT alone. It raises cardiovascular risk across all baseline risk categories.

Carotid imaging is recommended when a bruit is present, after a transient ischaemic attack or cerebrovascular disease, or as part of the assessment of people with evidence of vascular disease to identify more than mild carotid stenosis, above 50% of the lumen.

Pulse-wave velocity

Increased stiffness of the large arteries is a major mechanism of age-related high systolic and low diastolic pressure, isolated systolic hypertension and increased pulse pressure in older people. Uncontrolled hypertension accelerates age-related stiffening. Recent data suggest that increased stiffness may also participate in the early stages of hypertension and precede its development.

Stiffness in superficial arteries can be estimated from the relationship between pulse waves and changes in arterial diameter. It is now usually measured with carotid–femoral pulse-wave velocity (cfPWV) or brachial–ankle PWV (baPWV). cfPWV is the European reference standard for large-artery stiffness, with reference values in healthy and cardiovascular-risk populations. Reference values for baPWV are more established in Asian populations, with European data emerging.

Arterial stiffness increases with pressure and is variable, with a greater effect during systole than diastole. This reflects pressure-dependent recruitment first of more distensible elastic tissue and then of less distensible collagen. Ageing and disease also change the arterial wall: collagen deposition and fibrotic tissue increase while elastin decreases. Functional factors can rapidly alter stiffness without an immediate change in pressure, probably through contraction of vascular smooth muscle. This may explain the acute stiffening caused by increased sympathetic activity in medium and large arteries.

The traumatic effect of pulsatile pressure is greater in stiff arteries and promotes atherosclerosis. Framingham and European data show that increased large-artery stiffness is common in hypertension. cfPWV is higher in MH than in true normotension, so increased PWV in a person with elevated office pressure may identify someone who needs out-of-office monitoring and may have high risk.

Two meta-analyses found that cfPWV or baPWV can improve cardiovascular-risk classification compared with conventional risk models, particularly in young and middle-aged people at low or intermediate risk. Adding cfPWV to traditional Framingham risk factors improved the net reclassification index for cardiovascular mortality by up to 27%; adding baPWV to a Framingham model improved it by 24.7%. Higher cfPWV and baPWV also predict later hypertension in apparently healthy adolescents, young adults and middle-aged adults. Because PWV is age-related, it is a central measure of vascular ageing.

All antihypertensive drugs reduce stiffness passively by lowering pressure and unloading the stiffer collagen component of the arterial wall. Some studies suggest that certain drugs may reduce stiffness more effectively and potentially improve arterial structure, but the evidence is not conclusive because stiffness must be compared at equivalent blood pressures. Reports of reduced stiffness without a pressure reduction exist, but no RCT has proven that treatment-induced reduction in stiffness lowers cardiovascular events.

In SPARTE, a PWV-guided strategy was compared with conventional blood-pressure-targeted treatment. Despite limitations, there was no significant difference in the primary cardiovascular endpoint, although age-related PWV increase was smaller in the PWV-guided group. A post-hoc SPRINT analysis also found less PWV increase with more intensive pressure reduction, and the PWV effect was associated with better outcomes. Improved PWV has similarly been linked with better prognosis in resistant hypertension and dialysis populations.

Ankle–brachial index (ABI)

ABI is the ratio of ankle systolic pressure to brachial systolic pressure, measured with the patient supine using continuous-wave Doppler, a blood-pressure cuff and preferably a Doppler or automated oscillometric device. An ABI below 0.90 suggests lower-extremity arterial disease (LEAD), while a value above 1.40 suggests medial arterial calcification. Both low and high ABI are independently associated with cardiovascular events.

One meta-analysis found that 10-year cardiovascular mortality was 4.2 times higher in men and 3.5 times higher in women with low ABI than in people with normal ABI. Adding ABI to the Framingham score reclassified risk in 19% of men and 36% of women. ABI is relatively easy to obtain and requires only brief training. It should be measured in people with symptoms or signs of LEAD.

Renal HMOD

Hypertension is the second most important cause of CKD after diabetes and can also result from primary kidney disease. Kidney-function decline can be detected with routine laboratory tests and equations estimating eGFR from serum creatinine. Serum creatinine alone is an insensitive marker because a substantial fall in kidney function can occur before it rises.

CKD is classified by eGFR, calculated with the 2009 CKD-EPI equation, and by albuminuria measured as the urine albumin-to-creatinine ratio (ACR), preferably from an early-morning spot urine sample. Renal HMOD is diagnosed by reduced kidney function or albuminuria, although in hypertensive kidney disease albuminuria may appear only after eGFR has fallen.

Lower eGFR and higher urinary albumin indicate kidney injury and independently predict cardiovascular risk and kidney-function decline. In a study of more than three million participants, eGFR and ACR improved prediction of myocardial infarction, stroke and cardiovascular mortality when added to SCORE2 and SCORE2-OP, with a net reclassification improvement of approximately 10%. Whether treatment-related changes in eGFR or ACR predict later renal or cardiovascular outcomes is not yet fully established.

Long-term treatment-related changes in GFR have been associated with kidney failure and cardiovascular events, while treatment-related changes in proteinuria or albuminuria have shown inconsistent relationships with mortality. In ONTARGET, a two-year change in albuminuria was assessed in more than 20,000 high-risk patients. Changes were associated with total mortality and cardiovascular and renal outcomes in people with and without diabetes: a 50% reduction was associated with lower mortality, whereas a 100% increase was associated with adverse cardiovascular and renal outcomes and higher mortality.

Antihypertensive treatment often causes an early rise in serum creatinine, up to 20–30%, especially with renin–angiotensin-system blockers. This is usually a haemodynamic effect caused by lowering pressure-dependent filtration and does not necessarily represent kidney injury. The long-term clinical meaning of the early change remains uncertain.

Serum creatinine, eGFR and ACR should be documented in every patient with hypertension. If CKD is diagnosed, they should be reassessed at least annually. A negative urine dipstick does not exclude albuminuria; a laboratory ACR from a spot urine sample should be obtained. The renal resistive index (RRI) can be measured by Doppler. It is non-invasive and reproducible and reflects renal and systemic arterial compliance or resistance. In healthy people, RRI has been reported around 0.58–0.64; a value below 0.70 has traditionally been considered evidence of increased resistance to renal blood flow, although substantial heterogeneity exists. Higher RRI is associated with subclinical kidney damage and predicts cardiovascular and renal outcomes in hypertension, CKD and other cardiovascular diseases, independently of conventional factors, albuminuria and eGFR.

Cerebral HMOD

Hypertension is a major risk factor not only for acute cerebrovascular events such as stroke, intracranial haemorrhage and transient ischaemic attack, but also for chronic brain injury with or without symptoms that may ultimately lead to dementia. Long-standing hypertension has cumulative effects, including atherosclerosis, white-matter lesions, silent cerebral infarcts, microbleeds and brain atrophy, particularly hippocampal atrophy.

Hypertension damages small cerebral vessels and their network, contributing to microbleeds and lacunar infarcts. Lacunes are small, usually 2–20 mm, infarcts in deep white matter, basal ganglia or pons, thought to result from occlusion of a single small penetrating artery. White-matter lesions appear as T2 and FLAIR hyperintensities. Increasing white-matter burden and silent infarcts are associated with stroke and cognitive decline, including dementia.

Aortic stiffening transmits excessive pulsatile energy to the high-flow, low-resistance cerebral circulation, changing and damaging small vessels. Retinal arterioles can provide a window on this remodelling, although advanced retinal imaging is not widely available. Cross-sectional studies in middle-aged and older adults link aortic stiffness and excessive pulsatile transmission with cerebral microvascular damage and lower scores in several cognitive domains. Longitudinal data in people around 60 years also link higher pulsatile transmission with accelerated cognitive decline over the following decade. PWV is associated with cognitive decline in people over 80 independently of blood-pressure level. A meta-analysis of 29 cross-sectional and nine longitudinal studies confirmed an inverse association between large-artery stiffness and cognition, particularly executive function, memory and global cognition. Blood-pressure variability and orthostatic hypotension in older adults may also contribute.

Long-term systolic and pulse pressure in cognitively healthy adults over 50 have been associated with later cognitive decline and dementia. White-matter lesions, microbleeds and MRI hyperintensities are among the commonest forms of cerebral HMOD. MRI is not widely available for routine hypertension assessment, but, where possible, white-matter disease and silent infarcts should be sought in people with neurological symptoms, memory loss or cognitive decline.

Short screening tools such as the Mini-Mental State Examination (MMSE) and the newer Montreal Cognitive Assessment (MoCA) can be considered in people with hypertension over 65 years. Their simplicity allows use in primary care or a hypertension clinic. A person with an MMSE below 24, a MoCA below 26 or a subjective memory complaint should be referred for neurological or geriatric assessment.

Ocular HMOD

Hypertensive retinopathy is classified by fundus examination, which can detect haemorrhage, microaneurysms, hard exudates and cotton-wool spots (grade 3), as well as papilloedema and/or macular oedema (grade 4). These changes indicate severe retinopathy, are relatively specific and reproducible, and predict all-cause mortality. Grade 1 and 2 findings, such as focal or generalised arteriolar narrowing and arteriovenous nicking, are less specific and reproducible and have much weaker prognostic value.

Hypertension is also a major risk factor for retinal vascular disease, including retinal vein or artery occlusion and ischaemic optic neuropathy. Fundus examination should be targeted, particularly in people with hypertensive emergencies, suspected malignant hypertension or diabetes. Smartphone-based fundus imaging may help extend assessment of hypertensive retinopathy to more patients.

Retinal arteries may provide a useful marker of microvascular remodelling in other organs. Laser Doppler and adaptive-optics techniques are increasingly used to estimate the wall-to-lumen ratio of retinal arterioles, which correlates with pressure load and other HMOD markers. Microscopic assessment remains the invasive reference method for microvascular structure. Although subcutaneous small-artery changes have prognostic value, the ability of retinal wall-to-lumen ratio and its treatment-related change to predict cardiovascular events still requires confirmation.