CrossTalk opposing view: Which technique for controlling resistant hypertension? Carotid chemoreceptor denervation/modulation

Julian Paton obtained his PhD at the University of London. He was supervised by Professor K. M. Spyer and received classical training in in vivo integrative cardiovascular physiology. Subsequently, he worked in industry (DuPont, Wilmington, DE, USA), the University of Washington, Seattle, USA and the University of Göttingen, Germany, where he developed in vitro neurophysiological skills and an interest in the central neural control of respiration. He has been awarded the Sharpey–Schaffer prize, the Carl Ludwig prize lecture and was a recipient of a Royal Society Wolfson research merit award. He is presently the Deputy Editor-in-Chief (Europe) for The Journal of Physiology. Laura Ratcliffe is a Clinical Research Fellow at the University of Bristol and University Hospitals Bristol NHS Foundation Trust (UK). Dr Ratcliffe obtained her MBBS and BSc from Imperial College London (UK) and started specialist training in Renal and General Internal Medicine in 2009. She works in the Specialist Hypertension Clinic at the Bristol Heart Institute and is involved in studying interventional treatments for resistant hypertension. Her current research focuses on understanding the role played by the carotid body in hypertension.

Why do we need interventional approaches to treat hypertension?

The often asymptomatic condition of hypertension can soon become symptomatic with drug treatment. Over a third of patients report unwelcome medication side-effects and the majority would prefer to lower their blood pressure (BP) without medication (Benson & Britten, 2003).

The chronicity of hypertension and the associated treatment side-effects lead to medication non-adherence. Recent data suggest that 25% of patients attending specialist hypertension clinics are non-adherent to medications, and 23% referred for renal denervation (RDN) were completely non-adherent to prescribed anti-hypertensive drugs (Tomaszewski 2014). Medication non-adherence is associated with increased risk of stroke and cardiovascular complications (Herttua et al. 2013).

Despite an armoury of poly-pharmacy there remains an unmet clinical need. Between 12 and 15% of patients with hypertension are drug resistant (BP ≥140/90 mmHg despite ≥3 anti-hypertensive medications; NICE, 2011; Pimenta & Calhoun, 2012). These patients are at risk of significant morbidity and mortality, but there has been a dearth of new anti-hypertensive drugs licensed.

Numerous organ-targeted interventions have arisen over the last 5 years. Herein, we consider the carotid body (CB; peripheral chemoreceptor) and carotid body modulation (CBM) as a potential target for the treatment of drug-resistant hypertension. Our proposal is reliant on historic human and recent pre-clinical animal data.

The CB and its link to hypertension

Many patients with hypertension experience an imbalance in their neuro-humoral homeostatic control of vascular resistance and/or cardiac output (Fink, 2009; Hart et al. 2009). Autonomic imbalance with raised sympathetic nerve activity is typical in patients with resistant hypertension (Schlaich et al. 2004; Fisher & Paton, 2012).

The carotid bodies are strategically placed at the carotid bifurcation to detect blood hypoxia and hypercapnia, and play a critical role in protecting the brain from oxygen deficit (Ponte & Purves, 1974) by promptly increasing minute ventilation and arterial pressure, thereby improving cerebral perfusion. There is a relatively direct reflex pathway from the CBs leading to the excitation of pre-sympathetic neurones in the medulla (Guyenet, 2000) and hypothalamus (King et al. 2012). Given the importance of maintaining brain oxygen, it is not surprising that reflex activation of CB chemoreceptors evokes pronounced increases in arterial pressure and sympathetic activity in many sympathetic outflows (Marshall, 1994; Paton et al. 2006) and muscle sympathetic nerve activity in humans (Somers et al. 1989; Narkiewicz et al. 2006).

There is compelling evidence that in hypertension the sympathoexcitatory reflex response is augmented in both rats (McBryde et al. 2013; Tan et al. 2010) and humans (Anderson et al. 1989; Kara et al. 2003). Pertinent to this discussion is our finding that in spontaneously hypertensive (SH) rats the CBs are pivotal for both initiation (Abdala et al. 2012) and maintenance of hypertension and elevated sympathetic activity (McBryde et al. 2013), and its augmented respiratory modulation (Moraes et al. 2014). This has led to the concept of ‘CB tone’, which appears to be driving sympathetic vasomotor activity chronically (Paton et al. 2013) but not cardiac autonomic activity or ventilation (McBryde et al. 2013). Moreover, CB tone is specific to the hypertensive condition (Fletcher, 2001; Abdala et al. 2012; McBryde et al. 2013; Peng et al. 2014) and is ‘functionally’ absent in normotensive rats. Parallel studies in humans with hypertension have also revealed CB tonicity driving sympathetic vasomotor tone (Sinski et al. 2012). In these studies CB tone was reversibly inactivated using 100% oxygen which reduced sympathetic activity, an effect not seen in subjects with normotension.

Studies in the 1940s–1960s attempted to alleviate dyspnoea in patients with obstructive airway disease by surgically resecting one or both carotid bodies. Bilateral CB resection was noted to produce an acute reduction in blood pressure (Winter & Whipp, 2004), which was observed to persist for at least 6 months in a sub-group of hypertensive patients (Nakayama, 1961). Of note, there was no reduction in blood pressure in the normotensive group (Nakayama, 1961). Collectively, these data support the notion of targeting the CB for treatment of hypertension.

Hypothetical ideals for an interventional approach to hypertension

Any new interventional therapy must be safe and should:

1. be free from significant adverse effects;

2. allow pre-procedure identification of patients likely to respond;

3. have a measure of procedural success at time of intervention and at follow-up;

4. produce BP falls comparable to those achieved with pharmacological therapy;

5. cause permanent BP lowering towards target levels;

6. permit reduction of anti-hypertensive medication, especially those affecting quality of life;

7. be reversible.

How well might CBM map onto these hypothetical ideals?

Surgical CB resection in >5600 patients with asthma or chronic obstructive pulmonary disease (COPD) appeared relatively safe (Paton et al. 2013). CB resection requires manipulation of the common carotid artery and its bifurcation. This could cause embolisation of unstable atherosclerotic lesions, resulting in cerebrovascular complications. The resistant-hypertensive population may be at greater risk of atherosclerosis than those with asthma or COPD, but non-invasive imaging can be used to exclude those at highest risk.

CB resection has the potential for loss or attenuation of the hypoxic ventilatory drive. This could be significant in situations of chronic hypoxia (e.g. high altitude) or intermittent hypoxia (e.g. severe sleep apnoea or free diving), particularly in patients with COPD or poor physiological reserve. Studies in patients with severe COPD and asthma showed altered ventilatory and arterial blood gas responses to exercise (Stulbarg et al. 1989; Whipp & Ward, 1992) and acute hypoxia (Holton & Wood, 1965; Vermeire et al. 1987) post bilateral CB resection. Animal studies suggest that aortic body chemoreceptors compensate for the lack of chemosensitivity post bilateral CB resection, but this was not seen in humans (Holton & Wood, 1965). There was, however, no evidence of increased nocturnal desaturation in four patients after bilateral CB resection compared to other patients with chronic airflow limitation (Vermeire et al. 1987). It would seem prudent to consider only unilateral CB resection in any clinical trial for hypertension, with the aim of maintaining afferent sensitivity to hypoxia.

Unlike RDN, CBM therapy may allow a priori identification of patients most likely to benefit from the procedure. It is possible to non-invasively demonstrate the presence of CB tone and its relationship with hypertension and autonomic imbalance prior to intervention. Both hyperoxia (Sinski et al. 2012) and intravenous dopamine have been used to reversibly inactivate the CB whilst monitoring cardiovascular parameters, including sympathetic activity (Stickland et al. 2007; van de Borne et al. 1998).

The procedural success of CBM by surgical resection could be assessed post-operatively by the reduction or absence of cardiovascular/respiratory responses to hypoxia. This would allow any re-growth or re-innervation of glomus tissue within the CB to be identified. In contrast, there is no comparable technique to measure procedural success of RDN, which complicates the interpretation of RDN studies including the recent SYMPLICITY HTN-3 (Bhatt et al. 2014). Carotid sinus stimulation (CSS) benefits from confirmation of a functional implant at the time of surgery, but potential issues surrounding device failure (i.e. lead fracture, poor electrical contact due to fibrosis/movement, co-activation of carotid body afferents) might affect its success.

Nakayama reported substantial blood pressure lowering (systolic BP ∼170 mmHg fell to ∼140 mmHg) after CB resection, persisting for 6 months until study completion (Nakayama, 1961). These data are on par with early reports following RDN (Krum et al. 2009) and CSS (Bakris et al. 2012). Patients recruited into RDN trials have generally had only a very small reduction in number of medications. Similarly, for CSS, reductions in medications have not been substantial (medication reduced from 5.3 ± 1.9 to 5.0 ± 2.0 medications at last follow-up visit; Bakris et al. 2012). In the SH rat following bilateral carotid sinus denervation (disrupting both carotid chemoreceptor and baroreceptor input to the brainstem), BP was reduced substantially but not normalised suggesting that pharmacological therapy may still be required to achieve BP targets (Abdala et al. 2012). Whether it will be possible to reduce medications after CBM remains to be seen.

CSS is a reversible procedure as the device can be explanted. RDN aims to interrupt the afferent and efferent nerves to the kidneys, although it is possible that re-innervation with nerve regrowth may occur. CB resection is non-reversible and the potential loss of hypoxia sensing is undesirable. Unilateral CB resection appeared to dramatically relieve symptoms of dyspnoea in some patients with asthma, although similar improvement was noted in sham-operated patients (O'Rourke & O'Rourke, 1964). Although ineffective in SH rats, unilateral CB resection was recently reported to be beneficial in a patient with heart failure (Niewiński et al. 2013). With the aim of maintaining afferent sensitivity to hypoxia, unilateral CB resection would be a sensible starting point for any clinical trial in hypertensive patients.

Conclusions

Given the multi-factorial mechanisms driving hypertension, a silver bullet interventional cure is unlikely. Interventional therapies may offer an organ-specific approach, aiming to avoid side-effects due to anti-hypertensive medications. The important question remains for both clinician and patient: which interventional approach should they choose? It is now essential to understand which hypertensive patients have raised CB tone and whether its modulation (surgical or pharmacological) can provide a sustained anti-hypertensive response. CBM has the advantage that aberrant CB tone associated with hypertension can be determined, potentially allowing identification of patients most likely to respond and acting as a measure of procedural success. A reversible treatment that could suppress aberrant CB tonicity whilst maintaining the hypoxia-sensing ability would be ideal. The combination of CBM with other interventions (e.g. RDN, CSS or deep brain stimulation; Patel et al. 2011) could also be considered for patients with the most difficult-to-treat hypertension.

Call for comments

Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘Last Word’. Please email your comment to journals@physoc.org.

Glossary

BP

blood pressure

CB

carotid body

CBM

carotid body modulation

CSS

carotid sinus stimulation

COPD

chronic obstructive pulmonary disease

RDN

renal denervation

SH rat

spontaneously hypertensive rat

Additional information

Competing interests

None declared.

Funding

The authors would like to acknowledge support from the British Heart Foundation (IBSRF FS/11/1/28400; E.C.H.), Cibiem Inc. (L.E.K.R.), National Institutes of Health (R01NS069220 and R01HL11162; A.P.A., W.P.) and the International Rett Syndrome Foundation (A.P.A.).