This study examined recovery, delirium, and neurocognitive outcome in elderly patients receiving dexmedetomidine or propofol sedation after undergoing cardiac surgery. Open-label randomized trial. Single center. A total of 70 patients older than 75 years without English language limitations and Mini Mental State Examination scores >20. Patients received either propofol (group P) or dexmedetomidine (group D) postoperatively until normothermic and hemodynamically stable. Quality of recovery (QoR) was measured by the QoR-40 questionnaire on postoperative day (POD) three. Secondary outcomes were incidence and duration of delirium, time to extubation, length of hospital stay, hospital mortality rate, postoperative quality of life (QoL; measured by SF-36 performed at baseline and six months postoperatively), and neurocognitive disorder (measured by Minnesota Cognitive Acuity Screen [MCAS] performed at baseline, POD5, and six months postoperatively). A total of sixty-seven patients completed the trial. There was no significant difference in QoR-40 scores (95% confidence interval [CI], -7.6081-to-10.9781; p = 1.000), incidence of delirium (group P, 42%; group D, 24%; p = 0.191), mean hospital stay (95% CI, -5.4838-to-1.5444; p = 0.297), mean time to extubation (95% CI, -19.2513-to-7.5561; p = 0.866), or mean duration of delirium (95% CI, -4.3065-to-1.067; p = 0.206) between groups. No patients died in the hospital. There were no significant differences in changes in SF-36 or MCAS scores over time between groups. There was a decline in MCAS score from preoperatively to POD5 in group P (95% CI, -8.95725-to- -2.61775; p = 0.0005), which was greater than that observed in group D. The authors' findings demonstrated that the use of dexmedetomidine compared with propofol in elderly patients undergoing cardiac surgery was unlikely to improve QoR/postoperative QoL. Although the study was underpowered to detect secondary outcomes, the results suggested no reductions in delirium, time to extubation, and hospital stay, but a potential decrease in delayed neurocognitive recovery.