Limitations of renal arterial hemodynamic measures as candidate biomarkers of renal sympathetic innervation

Introduction: The key role of maladaptive renal sympathetic activation in hypertension has led to the development of catheter-based renal denervation therapies. Unfortunately, the lack of practical physiological biomarkers impairs patient selection and prevents intraprocedural feedback for renal denervation. Methods: We tested whether renal arterial hemodynamic measures and the beat-to-beat variability of these measures reflect renal sympathetic innervation in two tightly controlled preclinical models of renal denervation. Bilateral renal hemodynamic data were obtained from ten conscious unilaterally denervated rabbits and ten anesthetized unilaterally denervated pigs that subsequently underwent stepwise catheter-based radiofrequency denervation of the contralateral kidney. Renal arterial mechanics were assessed by quantification of wave speed and input impedance modulus and phase shift. Variability was quantified as the within-recording standard deviation of each metric when measured on a beat-to-beat basis. Results: Wave speed trended down after surgical denervation in rabbits (P = 0.054) but not swine (P = 0.57); wave speed variability was unchanged in both models. Renal arterial input impedance modulus was not significantly affected by surgical denervation. Surgical renal denervation increased input impedance phase shift in rabbits (P = 0.0048) but not in swine (P = 0.85). The beat-to-beat variability of modulus did not differ significantly between innervated and surgically denervated kidneys in either species. Surgical renal denervation decreased phase shift variability in swine (P = 0.0048) but not rabbits. The difference in phase shift variability between kidneys in swine was eliminated after a single round of branch-vessel ablation (P = 0.42). Discussion: Renal arterial wave speed, input impedance, and their beat-to-beat variability did not consistently or dose-dependently reflect renal sympathetic innervation in these models. These findings argue against their use as direct physiological surrogates for renal sympathetic outflow.