Selecting patients for interventional procedures to treat hypertension

Abstract

Purpose: Interventional approaches to treat hypertension are an emerging option that may be suitable for patients whose BP control cannot be achieved with lifestyle and/or pharmacotherapy and possibly for those who do not wish to take drug therapy.

Materials and Methods: Interventional strategies include renal denervation with radiofrequency, ultrasound and alcohol-mediated platforms as well as baroreflex activation therapy and cardiac neuromodulation therapy. Presently renal denervation is the most advanced of the therapeutic options and is currently being commercialised in the EU.

Results: It is apparent that RDN is effective in both unmedicated patients and patients with more severe hypertension including those with resistant hypertension.

Conclusion: However, at present there is no evidence for the use of RDN in patients with secondary forms of hypertension and thus evaluation to rule these out is necessary before proceeding with a procedure. Furthermore, there are numerous pitfalls in the diagnosis and management of secondary hypertension which need to be taken into consideration. Finally, prior to performing an intervention it is appropriate to document presence/absence of hypertension-mediated organ damage.

PLAIN LANGUAGE SUMMARY

• RDN has emerged as a safe and effective approach to treat hypertension with BP lowering efficacy equivalent to antihypertensive monotherapy albeit with guaranteed ‘adherence’

• Presently populations most likely to respond to RDN are not clearly defined but given the costs of the procedure it is likely to be initially made available to those with resistant hypertension and those at highest cardiovascular risk

• There is no evidence to support the use of RDN in patients with secondary forms of hypertension and thus this should be thoroughly screened for prior to offering the procedure, especially in the setting of resistant hypertension

• Optimisation of lifestyle and drug therapy is key to good hypertension management and should be undertaken prior to an invasive procedure such as RDN being offered

• There are numerous pitfalls in the screening process for secondary hypertension which means that hypertension specialists should be involved in this component of the pathway

• RDN can be offered by interventional radiologist, interventional cardiologists or angiologists who have had appropriate training

• Clinical pathways for RDN must ultimately involve a multidisciplinary team overview with hypertension specialists, interventionists and imaging specialists combining efforts to ensure appropriate patient selection. This without question must involve patients in the decision-making process.

Keywords:

• Renal denervation
• resistant hypertension
• secondary hypertension
• blood pressure monitoring

Introduction

Given the numerous, established methods to treat hypertension with lifestyle measures and the vast array of pharmacological options with numerous drug classes and multiple within-class drug alternatives, one might question the need for an additional strategy with invasive approaches to treat hypertension. Before considering what is on offer, it is important to establish that both lifestyle measures and drug therapy to treat hypertension are both exquisitely dependent on lifelong patient adherence to strategies to treat an asymptomatic condition – and frequently drug therapy can result in adverse effects, thus causing hypertension-related symptoms for the first time. It has already been clearly demonstrated that clinician expectations with respect to long term commitment by patients to lifestyle and drug therapy strategies are rarely met [1]. Noting the global prevalence of hypertension with alarmingly poor rates of hypertension control in both men (18%) and women (23%), it behoves us to examine strategies not only to improve hypertension detection and drug treatment but also to countenance options such as renal denervation (RDN) which are independent from behavioural manipulation [2].

Selecting patients for interventional procedures

Which patients?

In the aftermath of the Symplicity HTN-3 study, a new generation of randomised, controlled, clinical trials has demonstrated the safety and efficacy of RDN with radiofrequency energy [RF-RDN] (in drug naïve and medicated patients) and with ultrasound [US-RDN] (in drug naïve and resistant hypertensive patients), and emerging data support the efficacy of ethanol mediated RDN too [3–8]. In all these trials, patients with established secondary forms of hypertension were excluded and, given that most recruitment came from hypertension centres of excellence, patients were also thoroughly worked up to ensure undiagnosed secondary hypertension was excluded.

Thus there is no data to date on use of RDN in patients with secondary hypertension and there is even a suggestion that patients with primary aldosteronism are unlikely to respond to renal denervation due to reduced muscle sympathetic nerve activity [9,10]. Patients with resistant hypertension appear an attractive population for whom RDN should be considered given they are highest risk of events and have most to gain [11]. On the other hand, patients should not be required to have resistant hypertension to receive treatment with RDN (or other interventional approaches) especially if this offers the best opportunity to attain hypertension control.

Other than those with secondary hypertension, patients in whom RDN is (relatively) contraindicated include those with reduced renal function (eGFR < 40 ml/min) and those with renovascular disorder (either renal artery stenosis or fibromuscular dysplasia) and small renal arteries (diameter < 3 mm) and structural renal abnormalities. Furthermore RDN should be avoided in those patients with valvular heart disease and pregnant females. Presently it is thought that patient with isolated systolic hypertension are less likely to benefit from RDN and these patients have been excluded from the recent randomised clinical trials [12].

Which procedure?

To date, US-RDN (Paradise™, Recor Medical, Palo Alto, California, USA) and RF-RDN (SPYRAL™, Medtronic, Minneapolis, Minnesota, United States) are advanced in both their clinical evaluation programmes and with commercialisation efforts. Some geographies already have reimbursement codes for the procedures (e.g. Germany; The Netherlands) and thus availability of RDN as a therapeutic option will be initially governed largely by location and in any particular site, interventionalists will have a choice of using one or both platforms. The advantages of having both are to provide a greater range of anatomical suitability given that US-RDN with the Paradise system presently requires renal artery dimensions of 4–8 mm diameter and >25 mm in length. Multielectrode RF-RDN with the SPYRAL catheter can access vessels 3–8 mm in diameter and can navigate complex anatomy.

Presently there is data to suggest equivalent efficacy (with no safety concerns) using these two catheters from a single small study and therefore choice of catheter will ultimately come down to operator preference given that costs are comparable and procedural duration is similar between the technologies [13].

Work up of patients for intervention

Reversible secondary causes of hypertension should be established before offering a costly invasive procedure to treat hypertension as their presence will alter the management pathway. Patients with uncontrolled hypertension can be selected for interventional therapy following the pathway outlined in Table 1. A reasonable time frame for such a work-up would be 3–6 months given constraints in socialised medicine systems. The findings from the work up should be discussed in a multidisciplinary team meeting with appropriate specialists (interventionists and hypertension specialists) with documentation of the outcome and decision of whether or not to proceed to interventional therapy of hypertension.

Table 1. Work up pathway for selection of hypertensive patients for interventional therapy.

Steps in patient selection	Rationale for approach
1. Identify patients for whom device therapy of hypertension seems appropriate
Patient with uncontrolled HTNCould be unmedicatedMay be uncontrolled taking 1-2 drugsMay have resistant hypertension	Interventional therapy should be available to the following:Not keen to take medsPrepared to take meds & have RDNUnable to take meds due to drug intolerances
2. BP measurement to confirm hypertension Uncontrolled hypertension should be confirmed in the out of office setting before considering RDN
Office BPABPMHome BP	Important starting point: does not exclude white coat or masked HTN GOLD STANDARD To confirm daytime and night-time BP levels Where ABPM not available, critical to confirm out of office hypertension
3. Lifestyle & Medication optimisation
High fruit and vegetable (Mediterranean) dietSalt restrictionIncreased dynamic cardiovascular exerciseWithdraw drugs that cause hypertensionImprove adherence to antihypertensive medication	BP lowering effects similar to monotherapy [14] Aim for <5-6 g per day, (see also section 3A)Prevents hypertension & lowers BP, reduces CV morbidity & mortality [15 ](See paper by Sudano et al) Depends on access to resource such as SPC therapy. (See section 3B)
4. Screen for secondary hypertension(see Figure 1)(can be done in conjunction with step 2 to minimise delay in offering procedure and de-risking patient)
Renal parenchymal or renovascular hypertensionEndocrine hypertensionObstructive sleep apnoeaAortic coarctation	Common in both children and adults [16]. (See also section 4A)Includes common and rare causes of hypertension (See also Section 4B)Common in patients with resistant hypertension (see also Section 4C)Second most common cause of hypertension in children Screen with echocardiogram, also MRA or CTA aorta can be used [17]
Uncontrolled hypertension confirmed and lifestyle & medication optimised,no reversible secondary cause of hypertension identifiedDocument HMOD MDT Discussion - PROCEED TO RDN

BP: blood pressure; ABPM: Ambulatory BP monitoring; RDN: renal denervation; SPC: single pill combination; CV: cardiovascular; MRA: magnetic resonance angiogram; CTA: computed tomography angiogram; HMOD: hypertension-mediated organ damage; MDT: multidisciplinary team.

Identifying patients for device therapy of hypertension

A reasonable initial criterion for the practising clinician is to ask whether or not my patient can achieve hypertension control without the need for a high-cost invasive procedure. With increasing emphasis on the role of patient choice, an alternative view would be to offer RDN to those who express a preference for it as long as this provides the most pragmatic route to achieving hypertension control. Notably studies demonstrate willingness of patients to have device therapy of hypertension in preference to lifelong medication and this is unrelated to their BP levels or number of prescribed medications [18,19].

BP measurement

It is critical to ensure that the patient is truly hypertensive when measured according to appropriate standards both in the office and out of office (see Table 2). This requires correct technique of measurement as well as trustworthy technology. Presently only oscillometric monitors are validated for home BP measurement although an emerging class of cuffless, wearable BP monitors will at some point in the foreseeable future also provide valuable out of office BP metrics [21,22].

Table 2. Thresholds for diagnosis & treatment of hypertension with out of office monitoring.

Category	Office	Ambulatory	Home	Office	Ambulatory	Home
	Threshold for hypertension diagnosis			Treatment targets
SBP (mmHg)	≥ 140	≥135 (awake) ≥120 (asleep) ≥120 (24 H mean)	≥ 135	Age 18–64. < 130Age 65-79. <140 < 130 where toleratedAge 65-79 with ISH. < 140-150 < 130–139 if tolerated Age ≥80 < 140-150	Not stated	Not stated
DBP (mmHg)	≥ 90	≥ 85 (awake) ≥ 70 (asleep) ≥120 (24 H mean)	≥ 85	< 80	Not stated	Not stated

Adapted from the 2023 ESC/ESH Hypertension guideline [20].

SBP: systolic blood pressure; DBP: diastolic blood pressure 24 H, 24 h; ISH: isolated systolic hypertension.

The safety and efficacy of RDN has not been demonstrated in individuals with masked hypertension (although it is likely to be effective) or white coat hypertension to date. Whilst office measurement of BP is a starting point in this process, ambulatory BP monitoring (ABPM) and standardised home BP monitoring are critical adjuncts to assist with understanding BP on a continuous basis.

Home BP monitoring

Patients should be educated how measure BP at home using validated, digital, oscillometric equipment with the appropriate cuff size and with the correct technique [23]. This includes measurement in either arm (or the arm with higher BP if there is an interarm difference), whilst seated with their back supported, after resting for five minutes. They should not have consumed caffeine or smoked or have exercised within 30 min of measurement. Blood pressure should be measured at least twice, 1 min apart, in the mornings and evenings in order to evaluate BP levels, and/or capture treatment effects in advance of the next clinic visit. Home BP can be used to both diagnose and guide treatment of hypertension and provides superior prognostic information compared to office BP (see Table 2) [24].

ABPM

ABPM is proven to be cost effective and is recommended by international guidelines as a key component of the diagnostic pathway in patients who are being considered for medication to treat their hypertension [23,25,26]. In multiple studies ABPM has been proven superior to office BP in predicting cardiovascular morbidity and mortality in both patients with hypertension and population cohorts [27].

ABPM can confirm the diagnosis of white coat hypertension (prevalence up to 20% above the age of 70 years) as well as masked hypertension (prevalence 15%) [28,29]. In addition to provide diagnostic information on BP levels, ABPM is increasingly used to guide treatment decisions and monitor response to therapy though there is presently insufficient data for firm guidelines’ recommendation regarding optimal targets (see Table 2) [27]. There are significant limitations to ABPM use as outlined in Table 3 along with proposed strategies to get around them.

Table 3. Pitfalls in the use of ABPM and options to solve them.

Pitfalls	Solution
ABPM is not always easily accessible. Global uptake has been variable and often limited through lack of reimbursement and in poorer geographies there may be no access at all.	Home BP monitoring is an acceptable alternative where ABPM not available
ABPM is obtrusive during the day Cuff inflation optimally occurs every 20 minutes during the day - patients undergoing a study during a working day may be inconvenienced. This may result in cuff alerting or even removal by the patient.	Consider setting daytime measurement interval to every 20-30 minutes. Permit wearer to remove monitor temporarily for unavoidable reasons (e.g. important meeting)
ABPM is poorly tolerated overnight Nocturnal cuff inflation ideally takes place every 20 minutes and, in some individuals, this is a cause of sleep disruption and nocturnal hypertension and loss of prognostic significance of the night time BP [30].	Consider setting nocturnal measurement interval to every 60 minutes Be aware that elevated nocturnal BP may represent alerting to cuff inflation although sleep disordered breathing (obstructive sleep apnoea) is also a cause
Correct programming for day/night is critical For instance, a night shift worker will have a very different day/night pattern compared to someone who sleeps overnight and this must be taken into account in the acquisition and reporting of their data.	Ensure a standardised operating procedure (SOP) for ABPM fitting that mandates programming according to each individual’s sleep/wake cycle
Intra-individual short-term reproducibility of ABPM is poor For example, an ABP monitor fitted on a busy working day may show higher values than one fitted on a weekend but the five weekdays are more representative of the total BP load than two weekend days.	ABPM daytime mean should be compared to home BP average where possible. Monitor should be fitted on a day that is representative of a normal pattern for the individual

ABPM: ambulatory blood pressure monitoring.

Lifestyle & medication optimisation

Importance of sodium

Primary hypertension and age-related increase in BP does not occur in populations consuming < 1 g of salt daily [31]. Accordingly, high sodium intake is recognised to contribute to hypertension and increased cardiovascular morbidity and mortality [32,33]. Many clinicians overlook, and most patients are not always aware of, this fact and in general patients are unaware of the extent of their own salt intake. Globally average intake of sodium in adults is reckoned to be approximately 4 g per day (salt 10 g per day) [34]. International and country-specific guidelines recommend a reduced salt intake of < 6 g daily in individuals with hypertension [23,26,35–38]. This requires that patients be counselled regarding the sources of salt in daily life, both apparent (e.g. salt-shakers) and hidden (e.g. processed foods, condiments).

Reduction in salt intake is recommended to reduce BP levels and save lives in population cohorts: it is estimated that a modest 30% reduction in sodium intake, could save 40 million lives globally within 25 years [39]. Furthermore, a recent metanalysis has demonstrated that interventions to reduce salt intake result in clinically meaningful blood pressure reduction (4 mmHg per 6 g salt) in the short term [40].

Salt restriction is thus a reasonable approach to improving hypertension control in individuals prior to selecting them for renal denervation and is more likely to demonstrate larger effects on BP levels in salt sensitive individuals (older patients, black and Asian ethnicity) [41]. Methods to evaluate salt intake and their drawbacks are outlined in Table 4.

Table 4. Pitfalls in the assessment of salt intake.

Pitfalls	Solution
Commonly used methods to estimate sodium intake (24-hour dietary recall, food frequency questionnaire and food diary), have limited utility and tend to underestimate salt consumption [42]	Biochemical assay of urine is the most accurate way to record salt intake as up to 90% of consumed sodium is excreted in urine [43]
Spot urine electrolytes is not a reliable guide to salt ingestion	a 24-hour urine electrolyte assay is the gold standard method to estimate salt intake
24-hour urine collection underestimates salt intake insensible losses in sweat which will increase in hotter climates [44].	Assume salt loss of up to 1g a day due to sweating and more in hot climates

Optimising drug adherence

It would be reasonable to ensure that efforts to improve hypertension control through timely uptitration of medication although constraints upon primary and secondary healthcare provision in different geographies may mean that patients wait months for follow up appointments to have medication review. As such telemedicine programmes have been demonstrated to result in improved hypertension care but are not in widespread use [45].

In the case of adherence to drug therapy of hypertension, there are two key obstacles to therapeutic success:

1. Covert non-adherence: the patient does not take antihypertensive medication either from the outset (initiation failure) or does not take them as prescribed with missed doses (implementation failure) but does not admit this to physician – this is exceedingly common and more of an issue with a greater burden of prescribed antihypertensive medication [46].

2. Overt non-adherence: the patient complains of intolerance to antihypertensive medication and states this to physician – adverse effects can be explicable (e.g. cough with ACE-inhibitor, ankle swelling with calcium channel blocker) or unpredictable (feeling awful, fogging). This is less common but clinically very challenging as strategies to manage this are time consuming and costly although efficacy has been proven [47].

Improving covert non-adherence in hypertensive patients is challenging and whilst therapeutic drug monitoring assists in the identification of this problem (See Table 5), these are no data to support long term cost effectiveness or efficacy of adherence strategies although they should not be overlooked entirely [46]. These include approaches such as pill reminders and e-health monitoring systems coupled with self-monitoring of BP. A highly useful strategy would be to simplify medication regimens where possible through use of once daily single pill combination (SPC) formulations which are known to both improve adherence and reduce cardiovascular outcomes in hypertensives compared to monotherapy regimens [52–54]. Importantly the use of SPC for hypertension is now endorsed by major recent guidelines in the US and Europe although availability of two- and three-drug SPC varies widely across Europe and the UK [23,55].

Table 5. Pitfalls in Assessing & managing adherence.

Pitfalls	Solution
Assessing drug adherence can be time consuming and is costly and current non-biochemical methods (e.g. questionnaires, pill counting, prescription refill data, electronic diaries and directly observed therapy) all have significant drawbacks [48]	Mass Spectrometric analysis of plasma/urine samples is the current gold standard to assess adherence [49]
Measurement of drugs/metabolites in plasma/urine gives only limited information due to the dynamic nature of adherence [50]	Increased research utilising mass spectrometry techniques is required and will inform future use of these technologies to evaluate adherence and plan treatment approaches
Current methods of improving medication adherence for chronic health problems are mostly complex and not very effective [51]	Simplify antihypertensive medication regimen as much as possible: use a SPC formulation where possible with once daily dosing

SPC: single pill combination.

Screening for secondary hypertension

This should be limited to those patients in whom a secondary cause of hypertension is more likely as it would not be cost effective to screen all comers, especially those with milder forms for hypertension. Clinical features that should prompt investigation for a secondary form of hypertension are outlined in Table 6 and summarised in Figure 1.

Figure 1. Summary of causes of secondary hypertension and key screening investigations. A:CR: albumin:creatinine ratio; Ca: calcium; CKD: chronic kidney disease; CTA: computed tomography angiogram; DUS: duplex ultrasound; eGFR: estimated glomerular filtration rate; FHx: family history; LDDS: low dose dexamethasone suppression test; OSA: obstructive sleep apnoea; pMets: plasma metanephrines; PTH: parathyroid hormone; TSH: thyroid stimulating hormone; TTE: transthoracic echocardiogram; US: ultrasound.

Table 6. Clinical features suggestive of secondary hypertension.

Clinical feature	Description
Hypertension in childhood	Hypertension of any grade in children must always be investigated
Younger patients (<40 years) with grade 2 hypertension	Raises suspicion of a secondary cause especially in the absence of risk factors (family history/obesity/recreational drug use)
Acute worsening hypertension in patients with long standing stable normotension	Be mindful that other causes of BP increase should also be considered such as stress and medication non-adherence
More severe forms of acute or chronic hypertension	Resistant hypertension, Severe (grade 3] hypertension Hypertensive emergency (aortic dissection, hypertensive heart failure, stage IV hypertensive retinopathy, acute renal failure, eclampsia)
Presence of extensive HMOD	Left ventricular hypertrophy, hypertensive retinopathy, renal impairment, Arterial stiffening (increased pulse pressure and/or pulse wave velocity)
Clinical or biochemical features suggestive of secondary causes of hypertension (see also section 4A, 4B, 4C)	Coarctation of the aorta: differential BP between upper and lower limbs or between upper limbs (≥ 20/10 mmHg) Obstructive sleep apnoea: snoring, daytime somnolence, nocturia, obesity and increased neck circumference CKD: pallor, peripheral oedema Renovascular disorder: abdominal bruit, peripheral arterial disease Phaeochromocytoma: paroxysmal hypertension, palpitations, anxiety, or family history of phaeochromocytoma Cushing’s Disease: obesity, hirsutism, buffalo hump, diabetes, striae Hyperaldosteronism - hypokalaemia

HMOD: hypertension-mediated organ damage.

Renal parenchymal and renovascular disorder

Renal parenchymal disease is the most common cause of hypertension in children and adolescents and up to 10–15% of all adults [38]. It is generally asymptomatic and best detected with biochemical assay (estimated glomerular filtration rate, urinary albumin:creatinine ratio), urine dipstick for blood/protein and renal tract ultrasound imaging.

Renovascular disorder due to either atherosclerotic renal artery disease or fibromuscular dysplasia is recognised as an important cause of secondary hypertension and investigation to rule this out is an important component in the work-up of resistant hypertension [56].

Assessment of the renal arteries in the work up of renovascular hypertension has the added benefit on providing valuable information on anatomical suitability for the RDN procedure and presence/absence of accessory renal arteries. Imaging modalities include doppler ultrasound, Computed Tomographic and Magnetic Resonance angiography (CTA & MRA) and access to these is dependent on geography [57]. There are caveats to assessment of renovascular disorder as outlined in Table 7.

Table 7. Pitfalls in the assessment of renovascular disorder.

Pitfalls	Solution
DUS is highly operator dependent and has low sensitivity for detection of renal artery stenosis, especially in obese patients [58]	Where there is a high index of suspicion for RVD, cross sectional imaging techniques (with MRA/CTA) are preferred to DUS.
Appearances of renovascular lesion(s) using duplex US or cross-sectional imaging does not necessarily denote a functional stenosis	Further assessment with functional imaging such as Mag3 renography may be warranted Avoid use of plasma renin activity which is not of clinical value in RVD [58]
The diagnosis of FMD of the renal arteries should raise suspicion of disease in other vascular beds	Screening is recommended from brain to pelvis with CTA/MRA to document additional vascular tree involvement (e.g. cerebral aneurysm, Carotid artery dissection, visceral artery FMD) [59] Interventional treatment of FMD is with renal artery angioplasty (stent not required) and warrants input from a multidisciplinary team with vascular surgeon, interventional radiologist and nephrologist/hypertension specialists. The outcome may result in hypertension cure and the procedure may need to be repeated.
Atherosclerotic renal artery stenosis is a complex disorder	Interventional treatment of atherosclerotic renovascular disorder with renal artery angioplasty and stenting requires multidisciplinary team collaboration: interventionists and hypertension specialists +/- nephrologists. This should only be undertaken in centres of excellence Patients with isolated systolic HTN may be less likely to demonstrate a BP lowering response to renal stenting. Patients with resistant hypertension and atherosclerotic renovascular disease have increased arterial stiffness compared to those with primary hypertension [60]. In both the STAR and ASTRAL studies, grossly elevated Baseline pulse pressures (81 and 73 mmHg respectively, suggestive of severe arterial stiffness) may have been an important factor in the reduced BP lowering response to renal artery stenting [61,62]

DUS: duplex ultrasonography; RVD: renovascular disorder; MRA: magnetic resonance angiogram; CTA: computed tomography angiogram; MAG3 mercaptoacetyltriglycine-3 (a radionuclide isotope).

Currently there is no evidence to suggest the efficacy of RDN or other interventional therapies in treated (i.e. post balloon angioplasty or renal artery stenting) renovascular disease and it is largely considered a contra-indication for RDN in the main artery.

Renal accessory arteries are common and stenoses of these vessels are largely non-functional though whether RDN is appropriate and effective in this setting is unclear as these patients were largely excluded from rennet clinical trials.

In contrast, presence of chronic kidney disease is not a contraindication to RDN or other interventional approaches until the patient reaches end-stage kidney disease and trials are ongoing to ascertain whether even in this group there is benefit from RDN for BP control) [63].

Endocrine hypertension

Whilst some forms of adrenal hypertension (phaeochromocytoma, Cushing’s disease) are rarely encountered (Table 8), it is believed that primary aldosteronism (PA) accounts for 5–10% of all hypertension and is generally underdiagnosed [67,68]. Recent Endocrine Society guidelines to improve case detection and management of primary hyperaldosteronism have been published which standardise the approach to diagnosis [69]. Complexities surrounding the diagnosis and management of hypertension of adrenal origin are outlined in Table 9.

Table 8. Endocrine causes of hypertension.

Diagnosis	Prevalence in hypertension	Description	Work up
Primary Aldosteronism	5-15%	The commonest form of secondary hypertension and largely asymptomatic. Numerous subtypes identified (e.g. Familial Hyperaldosteronism Types I-IV) and increasing genetic basis identified [64]	Plasma electrolytes may show hypokalaemiaPlasma renin activity/mass, aldosterone levels24-hour urine for Na, K & aldosteroneConfirmatory tests (e.g. fludrocortisone suppression test or saline infusion test) CT or MRI adrenals
Cushing’s Disease	< 1%	Mostly arises from pituitary tumours but adrenal and ectopic sources also recognised [65]	Low dose dexamethasone suppression testing
Phaeochromocytoma	< 1%	Can be benign or malignant tumours of mostly adrenal origin but extra-adrenal origin recognised as well as hereditary forms (Multiple Endocrine Neoplasia, Von- Hippel Lindau, Hereditary paraganglioma) [66]	Plasma/urine fractionated metanephrinesCT or MRI abdomen & pelvis
Thyroid Disease	1-2%	Both overactive and underactive thyroid function are associated with hypertension	Thyroid function testsThyroid autoantibodies
Hyperparathyroidism	< 1%	Hypercalcemia is often asymptomatic unless very elevated Ca levels are present	Plasma calcium & parathyroid hormone levels CT neck

CT: computed tomography imaging; MRI: magnetic resonance imaging.

Table 9. Pitfalls in the assessment of hypertension of adrenal origin.

Pitfalls	Solution
Hyperaldosteronism is common but easily missed and complex to investigate and manage [70]	It is crucial to avoid interfering drugs to assay renin/aldosterone. This requires treatment with only verapamil and/or hydralazine and/or alpha blockers for 4 weeks prior to assay
ARR ratio is sometimes misleading as a screen	Up to 50% of ARR are falsely positive requiring confirmatory testing with saline infusion or fludrocortisone suppression testing False negative ARR arise in pregnancy and with hypokalaemia [71]
Adrenal venous sampling is required to prove lateralising autonomous aldosterone secretion if a surgical approach (adrenalectomy) is desired	This should only be done in specialised settings such as Hypertension centres of excellence or tertiary endocrine units
Cross sectional imaging (CT or MRI) will reveal adrenal nodules in 3-10% of all individuals:	mostly these are likely to be non-secretory (incidentalomas) but will require work up once detected [72]
Plasma/urine metanephrines can be falsely elevated [66]	Causes include anxiety & disease states (OSA, heart failure) Beware drugs that interfere with metanephrine assay – tricyclic antidepressants, monoamine oxidase inhibitors, stimulants (caffeine, nicotine)

ARR: aldosterone:renin ratio; CT: computed tomography imaging; MRI: magnetic resonance imaging; OSA: obstructive sleep apnoea.

Whilst adequate reversal of underlying endocrine drivers to hypertension are largely not a barrier to interventional BP therapies, this is not the case with medically treated (i.e. not suitable for unilateral adrenalectomy) primary aldosteronism, which is still considered to be a contra-indication for RDN and other modalities in the absence of dedicated data.

Obstructive sleep apnoea (OSA)

OSA is a common cause of hypertension in young adults (18–35 years) in whom other secondary cause are excluded and is also seen commonly in adults with resistant hypertension [73,74]. It should be screened for with sleep surveys (Epworth Sleep Scale, Berlin Questionnaire or STOP-BANG questionnaire) with higher scores correlating with increasing severity of OSA [75]. The gold standard investigation is overnight polysomnography (PSG) which requires a hospital stay. Alternatively home-based sleep tests are a reasonable alternative which have equivalent diagnostic sensitivity to PSG and are popular with patients [76]. Pitfalls in screening for OSA are outlined in Table 10.

Table 10. Pitfalls in screening for OSA.

Pitfalls	Solution
Not all hypertensive patients with OSA are obese [77]	Be alert to the possibility of OSA in young patients with no risk factors for hypertension and in any patient with loss of nocturnal dipping or reversed nocturnal dipping on ABPM [78]
Sleep scoring surveys have moderate diagnostic sensitivity and limited specificity for the diagnosis of OSA	When there is a high index of suspicion clinicians should proceed with HBST or PSG following a negative sleep survey and a positive survey should be followed up with confirmatory tests.

OSA: obstructive sleep apnoea; ABPM: ambulatory BP monitoring; HBST: home-based sleep test; PSA: polysomnography.

Notably, RDN has been shown to improve both office and ABP as well as sleep apnoea severity in a small randomised controlled study in 60 patients with resistant hypertension and moderate to severe OSA [79]. At present there is insufficient data to recommend treatment with RDN in this setting and more research is needed and OSA should be diagnosed and treated conventionally with lifestyle measures and usual interventions (weight loss, continuous positive airways pressure, mandibular advancement devices etc). In the clinical trial setting, adequate treatment of OSA (with improvement in hyponoea/apnoea index) for at least 3 months has not been a contra-indication for inclusion and thus this is a reasonable approach in clinical practice.

Conclusion

It is important to exclude secondary forms of hypertension prior to selecting patients for device-based therapy as there is no evidence to support their use in this setting. The work up can be complicated, and it is best left to high volume specialised units to fully investigate adrenal hypertension and also to make management decisions regarding interventional therapy of renovascular hypertension. In addition, there is no guarantee that a patient entering a work up pathway for RDN will necessarily be suitable for the procedure either due to ambulatory or home BP levels or not meeting anatomical criteria. Indeed in the recent randomised controlled clinical trials of RDN, only 10–20% of enrolled patients met full inclusion criteria meaning that many patients who are keen to have RDN will ultimately not be suitable if centres offering the treatment adhere strictly to protocol for delivering the therapy [3–6,8].

Given that access to specialised care is frequently limited, hypertensive patients then run the risk of uncontrolled hypertension during work up for interventional therapy and this should be avoided where possible for 2 important reasons. Firstly, it has been demonstrated that delays in intensifying BP control when SBP levels are >150 mmHg leaves patients at substantially increased risk of CV events and death [80]. This mandates clinicians to treat hypertension proactively and not delay initiation/uptitration of antihypertensive medications (i.e. avoid therapeutic inertia) during initial patient assessment. Secondly increasing evidence support the importance of durable BP control and that SBP time in target range is an independent predictor of major adverse cardiovascular events [81]. This warrants close follow up of patients with uncontrolled hypertension which has major resource implications and may best be served with e-health monitoring programmes to ensure ongoing appropriate medicines management to maintain hypertension control.

Thus, patients who are being evaluated for a device-therapy programme should be informed that they may not be suitable for the procedure for anatomical or BP reasons or that a secondary form of hypertension may be discovered. Furthermore, they should be notified that there is a 30% chance of non-response to RDN. Finally, whilst patients are being worked up, clinicians must continue to try to improve BP control with the aforementioned approaches and not delay initiation or uptitration of antihypertensive medication and leave their patient at increased cardiovascular risk.

Disclosure statement

Melvin D LOBO: Personal consultancy with Ablative Solutions, Medtronic, Recor Medical and Aktiia. Grant funding from Medtronic & Recor Medical. Manish SAXENA: Dr Saxena reports personal consultancy with Recor Medical Inc., Novartis, Astra Zeneca, Alnylam, Vifor Pharma, C4 Research. He has received Institutional grant from Recor Medical Inc., Ablative Solutions Inc., Applied Therapeutics, MSD. Gurvinder RULL & Vikas KAPIL: None to report.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.