Abstract Background Clinical studies of transcatheter radiofrequency renal denervation for treating hypertension have been hampered by the lack of consistent denervation efficacy. Microwave energy is well suited to renal denervation due to its capacity to spare vascular structures due to cooling from adjacent blood flow while enabling deep perivascular heating. Purpose We aimed to: 1) develop a transcatheter microwave system capable of safely delivering deep and circumferential perivascular renal nerve ablation, and 2) demonstrate the feasibility, short-term efficacy and safety of transcatheter microwave renal denervation. Method A novel 7F transcatheter microwave denervation system was designed, built, and iteratively prototyped in vitro and in 15 sheep. A histological grading system for microwave induced renal arterial and renal nerve injury was devised. The microwave denervation system was validated in an additional 9 sheep, which underwent unilateral renal denervation. Up to 2 microwave ablations were delivered to each artery with maximum power at 100–110W for 480s. Sheep were euthanised at 2–3 weeks post procedure. Gross microscopic histological examination as well as renal tissue norepinephrine content was analysed. Results Catheter deployment and ablation was successful in all 19 targeted vessel segments and ablation produced substantial circumferential perivascular injury; median ablation lesion area >395 (IQR 251–437) mm2, depth 17.1 (IQR 15.8–18.4) mm, length 16 (IQR 12–20) mm, without collateral visceral injury. Limiting power to 100W minimised arterial injury, while maintaining a deep circumferential perivascular ablation. At microwave ablation sites, a total of 292 nerve fascicles were identified, median distance from the renal artery of 4.2mm (IQR 2.1–8.8mm), of which 249 (85%) had sustained thermal injury with 128/249 (51%) showing grade 3–4 (moderate to severe) injury. Microwave denervation reduced median functional sympathetic nerve surface area at the renal hilum on anti-tyrosine hydroxylase staining by 100% (IQR 87%-100%), p=0.0039, and median renal cortical norepinephrine content by 83% (IQR 76%–92%), p=0.0078, compared to the paired control kidney. Conclusion Transcatheter microwave ablation can produce deep circumferential perivascular ablations over a long segment of the renal artery without significant arterial or collateral visceral injury to provide effective renal denervation. Clinical translation may enable more consistent and complete transcatheter renal denervation and antihypertensive efficacy. Acknowledgement/Funding University of Sydney; Western Sydney Local health District; National Health and Medical Research Council of Australia; National Heart Foundation (Au)