Effects of renal sympathetic denervation on the stellate ganglion and brain stem in dogs.

BACKGROUND: Renal sympathetic denervation (RD) is a promising method of neuromodulation for the management of cardiac arrhythmia. OBJECTIVE: We tested the hypothesis that RD is antiarrhythmic in ambulatory dogs because it reduces the stellate ganglion nerve activity (SGNA) by remodeling the stellate ganglion (SG) and brain stem. METHODS: We implanted a radiotransmitter to record SGNA and electrocardiogram in 9 ambulatory dogs for 2 weeks, followed by a second surgery for RD and 2 months SGNA recording. Cell death was probed by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay. RESULTS: Integrated SGNA at baseline and 1 and 2 months after RD were 14.0 ± 4.0, 9.3 ± 2.8, and 9.6 ± 2.0 µV, respectively (P = .042). The SG from RD but not normal control dogs (n = 5) showed confluent damage. An average of 41% ± 10% and 40% ± 16% of ganglion cells in the left and right SG, respectively, were TUNEL positive in RD dogs compared with 0% in controls dogs (P = .005 for both). The left and right SG from RD dogs had more tyrosine hydroxylase-negative ganglion cells than did the left SG of control dogs (P = .028 and P = .047, respectively). Extensive TUNEL-positive neurons and glial cells were also noted in the medulla, associated with strongly positive glial fibrillary acidic protein staining. The distribution was heterogeneous, with more cell death in the medial than lateral aspects of the medulla. CONCLUSION: Bilateral RD caused significant central and peripheral sympathetic nerve remodeling and reduced SGNA in ambulatory dogs. These findings may in part explain the antiarrhythmic effects of RD.

Sympathetic nerve fibers and ganglia in canine cervical vagus nerves: localization and quantitation.

BACKGROUND: Cervical vagal nerve (CVN) stimulation may improve left ventricular ejection fraction in patients with heart failure. OBJECTIVES: To test the hypothesis that sympathetic structures are present in the CVN and to describe the location and quantitate these sympathetic components of the CVN. METHODS: We performed immunohistochemical studies of the CVN from 11 normal dogs and simultaneously recorded stellate ganglion nerve activity, left thoracic vagal nerve activity, and subcutaneous electrocardiogram in 2 additional dogs. RESULTS: A total of 28 individual nerve bundles were present in the CVNs of the first 11 dogs, with an average of 1.87±1.06 per dog. All CVNs contain tyrosine hydroxylase-positive (sympathetic) nerves, with a total cross-sectional area of 0.97±0.38 mm(2). The sympathetic nerves were nonmyelinated, typically located at the periphery of the nerve bundles and occupied 0.03%-2.80% of the CVN cross-sectional area. Cholineacetyltransferase-positive nerve fibers occupied 12.90%-42.86% of the CVN cross-sectional areas. Ten of 11 CVNs showed tyrosine hydroxylase and cholineacetyltransferase colocalization. In 2 dogs with nerve recordings, we documented heart rate acceleration during spontaneous vagal nerve activity in the absence of stellate ganglion nerve activity. CONCLUSIONS: Sympathetic nerve fibers are invariably present in the CVNs of normal dogs and occupy in average up to 2.8% of the cross-sectional area. Because sympathetic nerve fibers are present in the periphery of the CVNs, they may be susceptible to activation by electrical stimulation. Spontaneous activation of the sympathetic component of the vagal nerve may accelerate the heart rate.