The kidney-heart connection during electrical storm: from bedside back to bench.

Abstract

One of the most challenging clinical pre-sentations for cardiologists is the ‘electrical storm’ of incessant ventricular tachycardia (VT). This often occurs in patients with previous myocardial infarctions or non-ischaemic cardiomyopathies, who are already on the maximally tolerated doses of cardiac medications (including β-blockers). Such patients may have already been implanted with a cardiac defibrillator (ICD) and are therefore likely to receive multiple therapies from their device. The cardiologist corrects plasma electrolytes, administers additional antiarrhythmic drugs (such as amiodarone and lidocaine) and reprograms the ICD in an attempt to suppress the storm, but ultimately many patients will require general anaesthesia and emergency ablation of their VT. This is a perilous ‘last-ditch’ procedure with limited long-term success. It may, however, stabilize the patient long enough for consideration of heart transplantation. Case reports and small case series have highlighted novel procedures aimed at reducing cardiac sympathetic drive as possible adjunct therapies in such patients. This builds on over 50 years of intense research linking cardiac sympathetic drive to arrhythmogenicity (Shen & Zipes, 2014). Even sedation or general anaesthesia in itself can be antiarrhythmic, because sympathetic hyperactivity during the electrical storm can be reinforced by chest pain from VT and ICD shocks received whilst still conscious. Recent publications have advocated stellatectomy (Schwartz, 2014), thoracic epidural anaesthesia and renal sympathetic nerve denervation (Bradfield et al. 2014) as potential approaches. In this context, the paper by Huang et al. (2014) in this issue of Experimental Physiology is of particular interest. In anaesthetized dogs following renal sympathetic denervation or a sham procedure, Huang et al. (2014) performed catheter-based measurements of ventricular refractory period and action potential duration (APD) restitution at multiple epicardial sites. This is the first publication to my knowledge to characterize ventricular electrophysiological changes in response to renal denervation. Interestingly, the flattening in the slope of the APD restitution curve, prolongation of the refractory period and reduced ability to induce APD alternans mirrors previous reports of the action of selective cardiac sympathetic stimulation on these parameters. Moreover, renal denervation reduced the incidence of ventricular arrhythmias in the hour following ligation of the left anterior descending coronary artery. The paper by Huang et al. (2014) also follows the recently published failure of renal sympathetic nerve denervation to produce a significant reduction in blood pressure in patients with resistant hypertension when compared with a sham procedure in the Symplicity HTN-3 trial (Bhatt et al. 2014). There are a number of interesting methodological differences in the approach to renal denervation taken by Huang et al. (2014) compared with that used in Symplicity HTN-3. In human subjects, renal denervation is usually undertaken via an endovascular approach using a radiofrequency ablation catheter introduced via the femoral artery. Discrete lesions are made circumferentially in a spiral pattern along both renal arteries, but unfortunately there is no way of determining that successful denervation has been achieved during the procedure. Huang et al. (2014) performed epivascular ablation via small bilateral retroperitoneal flank incisions. Furthermore, they demonstrated as evidence of successful denervation that high-frequency electrical stimulation of the adventitia no longer caused hypertension following ablation. This is a useful and novel approach, and a similar clinical end-point of success in patients is clearly needed. The fact that afferent and efferent nerves innervate the kidney through a plexus surrounding the renal arteries has long been appreciated. Renal afferent nerves are able to stimulate the hypothalamus in response to renal hypoxia and ischaemia and thereby increase global sympathetic activity. Likewise, sympathetic efferent nerves cause renin release, sodium and water retention, and through angiotensin II, reinforce sympathetic drive both centrally and by increasing catecholamine release peripherally (Singh et al. 2014). Renal denervation as a method to reduce cardiac sympathetic drive therefore appears logical, although many questions regarding this approach remain unanswered. Can a reduction in cardiac sympathetic activity following renal denervation be measured directly? Do the electrophysiological changes in APD and refractory period observed in the normal canine heart also occur in ischaemic and non-ischaemic cardiomyopathy? How long do these electrophysiological changes persist for and what are the underlying mechanisms by which they occur? In the clinical case reports of renal denervation for electrical storm, patients were already maximally β-blocked. Is the antiarrhythmic action of this procedure therefore independent of β-receptors, and if so, which other receptors or sympathetic cotransmitters play a role? This promises to be an active area of research, which may help us to identify new antiarrhythmic targets as well as refine our approach to renal denervation in patients. Readers are invited to give their opinion on this article. To submit a comment, go to: http://ep.physoc.org/letters/submit/expphysiol;99/11/1451 None declared. N.H. is a British Heart Foundation CRE Intermediate Fellow at the University of Oxford.