Patients undergoing elective total knee arthroplasty are randomized to either intrathecal anesthesia, wound infiltration with local anesthetics and standardized postoperative care (NA = neuraxial anesthesia) or general anesthesia, intraoperative glucocorticoids with accelerated postoperative care principles (GA). The study hypothesis is that GA would reduce length-of-hospital stay. Outcome variables are postoperative nausea and vomiting, mobilization rate, pain, requirement of analgesics, and length-of-hospital stay.

Study design. The study design is consecutive and randomized. Patients with osteoarthritis scheduled for TKA at the department of orthopedic surgery, Hässleholm Hospital, Sweden, are eligible for participation in the study. Exclusion criteria are body mass index (BMI) > 35 m/kg2, prior major knee surgery to the ipsilateral knee, ongoing infection, known immunological deficiency or ASA physical status category > IV.

Randomization and blinding procedure Randomization is performed by an employee, not involved in the study, who prepared non-transparent, sealed envelopes each containing a slip of paper with descriptions of whether the patient should receive GA or NA. The randomization is computerized. On the study day a nurse, likewise not involved in the study, opens the appropriate envelope and prepared accordingly the procedures. From the point where anaesthesia is delivered both patients and staff in the operating theatre and in the recovery unit are, for obvious reasons, aware of the method of anaesthesia being used. However, once the patients leaves the recovery unit staff members assessing home readiness are blinded to group allocation.

Assessments All patients are familiarized with a horizontal visual analogue scale (VAS, [100 mm]) used for assessment of pain and PONV (0 = no symptom, 100 = worst symptom possible). Assessments of pain are made pre-operatively, upon arrival to the post-anesthesia care unit (PACU), every second hour the first day, twice a day during the remaining hospital stay and at follow-up visits 3 and 12 months after surgery. Plasma levels of troponin T (TNT) are measured three times during the first 24 postoperative hours in order to evaluate signs of myocardial ischemic events. The Short Portable Mental State Questionnaire (SPMSQ) is used to evaluate the cognitive function pre-operatively, at the discharge and at follow-up visits 3 and 12 months after surgery. Global satisfaction score is assessed with a VAS (0 = best possible satisfaction, 100 = least possible satisfaction) at follow-up visits 3 and 12 months after surgery.

Anesthesia Oral premedication, administered 1 hour before surgery, is with midazolam 2.5 mg, paracetamol 2 g, meclozine 10 mg, celecoxib 200 mg and oxycodone 10 mg. A low-volume fluid regimen is used with 1000 mL of Ringer's solution (Fresenius-Kabi AB, 751 74 Uppsala, Stockholm) and 1000 mL of glucose 2.5% during the first 24 hrs.

In the NA group an indwelling urethral catheter is inserted prior to surgery. Intrathecal anesthesia is with bupivacaine 12.5-15.0 mg, morphine 0.2 mg and clonidine 30 µg (total volume (3.2-3.7 mL). During surgery an infusion of propofol (0.8-2.5 mg/min) is used to induce a light level of sedation. At the end of surgery a mixture of ropivacaine 110 mg, epinephrine 0.5 mg and morphine 10 mg (total volume 21 mL) is injected into the peri-surgical area. The mixture is injected using a systematic technique ensuring uniform delivery of the local anaesthetic to all tissues incised, handled or instrumented during the procedure. The first 7 mL of the mixture are injected into the posterior joint capsule and both collateral ligaments after the bone cuts had been performed. After insertion of the prosthesis another 7 mL are to be injected along the borders of and into the capsule and cut quadriceps tendon, infra-patellar ligament, possible remnants of the fat pad, cruciate ligaments and the soft tissues surrounding the joint. The last 7 mL is infiltrated into the subcutaneous tissues before wound closure.

In the GA group patients are asked to void before the transfer to the OR. The patients are draped before induction of anesthesia. Induction of anesthesia is with propofol and remifentanil, and endotracheal intubation is facilitated by succinylcholine. Maintenance of anesthesia is with target-controlled infusion (Marsh and Minto algorithm) with propofol (Braun Medical, Germany) and remifentanil (GlaxoSmithKline, Great Britain) aiming at initial concentrations of 5 µg/ml and 5 ng/ml, respectively. Ventilation is mechanical with oxygen in air (FIO2 = 0.45) and aiming at ETCO2 4.5 kPa. Betamethasone (Swedish Orphan Biovitrum, Sweden) 4 mg i.v. is given during surgery. At the end of surgery a mixture of ropivacaine 250 mg and epinephrine 0.3 mg (100 mL) is injected in the tissues in the same way as described above. Twenty min before the end of anesthesia an i.v. bolus dose of oxycodone 7.5-10 mg is given.

Surgery In the NA-group a tourniquet is applied around the thigh before the start of surgery. The GA-group does not receive a tourniquet. The surgeries are performed via a ventral incision with a parapatellar medial entrance to the joint. The patella is everted. A cemented single radius cruciate retaining (CR) total knee is used (the TriathlonTM Knee System (Stryker, Mahwah, New Jersey, USA)) for all patients. Appropriate guide instruments are used according to the surgical-technique manual supplied with the knee system.

PACU In the PACU intermittent doses of oxycodone 2-8 mg i.v. is given as rescue medication. Patients complaining of PONV are given ondansetron 4 mg i.v.

In the GA group bladder scans are done every second hour during the first 24 post-op hours. If the bladder contains > 400 ml an intermittent catheterization is performed. Mobilisation is started within 2 hrs of arrival to the PACU. A physiotherapist does passive bending of the knee to 90˚ and walks 10 meters with the patient on the floor.

In the NA group mobilization is not started until the day after surgery, due to residual motor and sensory blockade.

Ward In the NA group the indwelling urethral catheter is removed the day after surgery. Patients are considered ready to be discharged from the hospital when they are able to get in and out of bed, get dressed, sit down and get out of a chair, able to walk 50 m with or without walking aid, flex the knee to at least 70 º, fit for staircase climbing under supervision of a physiotherapist and accepting to be discharged. LOS is measured from end of surgery to time of discharge.

Inclusion Criteria:

• Patients with osteoarthritis scheduled for TKA at the department of orthopedic surgery, Hässleholm Hospital, Sweden, are eligible for participation in the study.

Exclusion Criteria:

• body mass index (BMI) > 35 m/kg2,
• prior major knee surgery to the ipsilateral knee,
• ongoing infection, known immunological deficiency or ASA physical status category > IV.