The goal of this observational study is to explore early cognitive changes in patients aged 70 years or older undergoing same-day discharge after hip or knee replacement surgery. The main question it aims to answer is:

Does patients experience cognitive changes in cognitive test performance from before surgery to the first day after surgery?

Participants will:

• Complete a cognitive test battery virtually about 14 days before surgery, and on the first and seventh day after their operation.
• Record pain levels and pain medicine use during the first week after surgery

Introduction Background Postoperative cognitive dysfunction (POCD) is a prevalent complication in elderly patients and has significant clinical and socioeconomic consequences, delayed recovery, decreased quality of life, persistent neurocognitive disorders and increased morbidity and mortality within the first year of surgery . The pathogenesis remains poorly understood, partly due to varying diagnostic approaches and methodologies. Nevertheless, there is consensus that POCD is multifactorial influenced by factors such as advanced age, sleep disturbances, and inadequate perioperative pain management .

Sleep deprivation, in particular, can exacerbate hyperalgesia, impair cognition, and lead to increased respiratory and cardiac risks due to REM sleep rebound. Fast-track surgical protocols, which emphasize multimodal rehabilitation-including regional anesthesia, opioid-sparing analgesia, and early mobilization-have been shown to reduce postoperative pain, sleep disturbances, and the risk of delirium. These protocols have reduced the average length of hospital stays worldwide. In Denmark length of hospital stay after total hip arthroplasty (THA), total knee arthroplasty (TKA) and unicompartmental knee (UKA) is now only one day and an increasing number of patients are discharged on the day of surgery. Same-day discharge enable earlier return to a familiar environment and potentially enhanced pathophysiological recovery.

No studies have described early postoperative cognitive test performance in elderly patients undergoing same-day discharge hip and knee arthroplasty. This study therefore aims to address this knowledge gap by examining early cognitive changes in patients aged ≥70 years after same-day discharge hip and knee arthroplasty.

Aim The primary aim of this study is to report early cognitive changes measured with Symbol Digit Modality Test (SDMT) in patients over 70 years undergoing same-day discharge THA, TKA and UKA. Secondary, to report cognitive changes using the Stroop Color Test and verbal fluency test as well as patient's perceived cognitive function. As patients return home on the day of surgery in this contemporary same-day discharge arthroplasty setting, we hypothesize that early postoperative changes in cognitive performance will be limited compared with earlier fast-track studies with a longer median length of stay (LOS).

Methods

The cognitive tests:

The cognitive tests will be administered by research assistants with prior research experience.

Eligible participants will be recruited from the outpatient clinic. Testing will be performed virtually with the patients being at home. The patients will receive instructions to complete all sessions in the same quiet room and on the same device, to keep environmental stimuli (lighting, background noise, screen distance, seating, posture, and internet conditions) constant and thereby reduce within-subject variance and practice effects. The baseline test will be administered 14 days +- 3 days prior to surgery to avoid preoperative anxiety effects. Postoperatively, testing will be carried out on postoperative day 1 and day 7. More frequent testing is avoided to reduce practice effects.

Only patients receiving spinal anesthesia will be included to minimize confounding from general anesthesia and to reflect current fast-track practice. Spinal anesthesia is administered using a relatively short-acting blockade bupivacaine 5 mg/mL, 2.5 mL (12.5 mg) in THA, while for primary TKA and UKA, bupivacaine 5 mg/mL, 2 mL (10 mg) . Standard local infiltration analgesia (LIA) will be used with 150 mL ropivacaine 2 mg/mL in knee arthroplasties but not in hip arthroplasties . Peripheral nerve blocks are not used. Patients will also receive cloxacillin 2 gr, methylprednisolone 125 mg and 1 gr tranexamic acid intravenously immediately prior to skin incision.

Postoperative pain management and other medication Postoperative pain management will follow a standardized multimodal, opioid-sparing regimen. All patients will receive paracetamol (1 g × 4) in combination with ibuprofen 400 mg × 3 (7 days) for THA patients in and 400 mg × 4 (14 days) for TKA and UKA patients, unless contraindicated.

Opioids will be reserved as rescue medication, administered only on an "as needed" basis. Morphine 10 mg tablets will be used as first-line rescue analgesia, with oxycodone 5 mg as an alternative for patients who do not tolerate morphine.

Data Source:

A REDCap database has been established in collaboration with the Open Patient Data Explorative Network (OPEN) at Odense University Hospital (25). Patients scheduled for hip and knee arthroplasty at Svendborg hospital and included in the planned study will be enrolled in the database following informed consent. Patient's baseline database will be registered on comorbidities and medication use entered by trained research staff.

Statistics and sample size Although this is an exploratory study, the sample size was guided by considerations from both the literature and our pilot planning. Because the design is paired, each participant serves as their own control (baseline vs. postoperative assessment). For the SDMT we assumed a minimal clinically important difference (MCID) of 4 points , a within-subject correlation of 0.85 between baseline and postoperative scores, and a standard deviation of 7.5 . With a two-sided alpha of 0.05 and 80% power, this corresponds to approximately 14 participants with complete paired measurements per subgroup (THA, TKA, and UKA) if analyses are performed separately. To account for anticipated dropout/non-same-day discharge, we will aim to enroll 16 patients per subgroup to obtain 14 participants with complete paired measurements.

Continuous variables will be summarized as mean with standard deviation (SD) and range if approximately normally distributed; otherwise as median with interquartile range (IQR) and range. Categorical variables will be described using absolute and relative frequencies (n, %).

Patient involvement Patients who have previously undergone hip or knee arthroplasty were involved in the design process. Their feedback informed the research questions and outcome measures, ensuring relevance and patient-centeredness. During pilot testing, participating patients also provided input on the planned assessment procedures and overall feasibility, which further informed methodological adjustments.

Inclusion Criteria:

• Unilateral elective primary Total Hip Arthroplasty, Total Knee Arthroplasty, or Unicompartmental Knee
• Native Danish speaker
• Spinal anesthesia

Exclusion Criteria:

• Inability or unwillingness to provide informed consent
• Inability to comply with virtual study procedures or follow-up sessions, including because of visual or hearing impairment
• Known dementia or other diagnosed cognitive disorders or Brief Assessment of Impaired Cognition (BASIC)≤20
• Parkinson's disease or other neurological conditions causing functional impairment
• History of alcohol abuse (≥35 units per week)
• Daily use of anxiolytics or hypnotics
• Not discharged on the day of surgery

Patients scheduled for hip or knee arthroplasty at Svendborg hospital, Denmark.

Individual participant data will not be publicly shared due to Danish data protection regulations and the nature of the clinical dataset. De-identified data may be made available from the investigators upon reasonable request and subject to applicable approvals.