Background Referral rates and outcomes of cardiac rehabilitation have not been evaluated in patients with transcatheter aortic valve replacement or compared with surgical aortic valve replacement. Method A retrospective cohort study was conducted in 488 patients who underwent transcatheter aortic valve replacement ( n = 199) and surgical aortic valve replacement ( n = 289) from a university-based statewide transcatheter aortic valve replacement/surgical aortic valve replacement program during 2015-2017. Cardiac rehabilitation consisted of supervised exercise, diet education, and stress and depression management. We compared changes from baseline in exercise duration and intensity during cardiac rehabilitation sessions, quality-of-life (36-Item Short-Form Health Survey), and psychosocial measures (anxiety, depression, mood, social support, and diet) between transcatheter aortic valve replacement and surgical aortic valve replacement patients using t-test and chi-square analyses. Results Of 488 patients, cardiac rehabilitation referral rates were similar at 41% (transcatheter aortic valve replacement 81/199 versus surgical aortic valve replacement 117/289), but enrollment rates were lower in transcatheter aortic valve replacement (27/199, 14%) versus surgical aortic valve replacement (102/289, 35%, p < 0.01). Among eligible patients, cardiac rehabilitation completion rates were lower in transcatheter aortic valve replacement (12%) than surgical aortic valve replacement (32%). Exercise intensity during cardiac rehabilitation improved in both groups in a similar fashion (transcatheter aortic valve replacement 1.03 ± 1.09 versus surgical aortic valve replacement 1.34 ± 1.15 metabolic equivalents), but increase in exercise duration was higher in transcatheter aortic valve replacement patients versus surgical aortic valve replacement patients (14.52 ± 6.42 versus 10.67 ± 8.38 min, p = 0.02). Improvement in physical composite score was higher in surgical aortic valve replacement versus transcatheter aortic valve replacement (8.72 ± 7.87 versus 2.36 ± 7.6, p = 0.02) while improvement in mental composite score was higher in transcatheter aortic valve replacement (8.19 ± 8.50) versus surgical aortic valve replacement (1.18 ± 7.23, p = 0.02). There was no significant difference between the two groups in improvement in psychosocial measures. Conclusion Cardiac rehabilitation enrollment was low in transcatheter aortic valve replacement patients versus surgical aortic valve replacement patients despite similar referral rates. Improvement in functional and quality-of-life performance was achieved in both transcatheter aortic valve replacement and surgical aortic valve replacement. Future studies should address obstacles for enrollment of transcatheter aortic valve replacement patients.