All participants enrolled in the olfactory enrichment group will receive twice sessions per day with an interval of at least 6 hours on preoperative day 1 and postoperative day 1-3. Each session will last 30 minutes. Each participant will be equipped with an electrical odor dispenser which allowed to distribute 8 odors (grapefruit, lavender, lemon, peppermint, tangerine, green tea, cloves and eucalyptus). When pushing a button, 4 odors will be released in turn and participants will sniff each odor (approximately 4 ml odorized air) for approximately 30 seconds by a nasal catheter. Considering of the adaption of odors, 4 odors will be used in the morning session and another 4 odors will be used in the afternoon session.

The sham group will receive sham olfactory enrichment sessions, which is designed with the same pattern (twice a day, 30 min for each session, on preoperative day 1 and on postoperative day 1-3. However, the containers provided to the sham group will be odorless. The odor containers and labels will appear identical to those used in the olfactory enrichment group.

Each participant will be equipped with an electrical odor dispenser which allowed to distribute 8 odors (grapefruit, lavender, lemon, peppermint, menthol, tangerine, green tea, and bergamot). The choice of odors was guided by (1) pleasantness of the odors, (2) presence of slight trigeminal activation in some of the odors, e.g., peppermint, (3) evaporation characteristics, so that the odors would last for the duration of the experiment, (4) technical issues mostly in terms of compatibility with the odor cartridge, (5) availability, and (6) inspiration from the previous studies.

The presence of postoperative delirium will be defined according to 3D-CAM, which includes four entries: acute onset and fluctuating course, inattention, disorganized thinking and altered level of consciousness. To diagnose delirium, both the first and second criteria have to be present plus the third and/or fourth criteria. Every participant will be assessed twice daily after anesthesia/surgery from postoperative day 1 to postoperative day 3, with the first assessment between 8:00 and 10:00. The second assessment will be between 15:00 and 17:00.

Participants' cognitive function will be assessed at baseline, on postoperative day 21, and at postoperative month 6 using a battery of neuropsychological tests, including the Montreal Cognitive Assessment (MoCA), Hopkins Verbal Learning Test Retention (HVLTRet), Hopkins Verbal Learning Test Total Recall (HVLTTR), Brief Visuospatial Memory Test Total Recall (BVMTTR), Brief Visuospatial Memory Test Delayed Recall (BVMTDR), Benton Judgment of Line Orientation (JLO), Digit Span Test (DST), Verbal Fluency Test (VFT) and Trail Making Test (TMT). Cognitive Change Index will be calculated by subtracting baseline scores from postoperative scores in each of the neuropsychological tests. Delayed neurocognitive recovery or postoperative neurocognitive disorder will be defined as by a decrease by one or more standard deviations in at least two of the neuropsychological tests from the baseline to 21 days or 6 months after anesthesia/surgery.

Severity of delirium will be assessed by the CAM-S. CAM-S includes 10 items: acute onset or symptom fluctuation, inattention, disorganized thinking, altered level of consciousness, disorientation, memory impairment, perceptual disturbances, psychomotor agitation, psychomotor retardation and sleep-wake cycle disturbance. Each item is scored as absent (0), mild (1) or marked (2), except acute onset or fluctuation, which is scored as absent (0) or present (1). The sum of CAM-S ranges from 0 (none) to 19 (most severe). The Delirium Symptom Interview (DSI) Daily will be used to determine CAM-S grades. The incidence of postoperative delirium will also be referred to the records by nurses on duty.