Dexmedetomidine for Sedation in Total Knee Replacements

The investigators plan a randomized control trial.

Sample. Based on a literature and a previous study by the investigators team (RQHR #13-80/USask Bio #13-232), average reduction in opioid consumption for dexmedetomidine bolus and infusion is 50%. Less data exists for single dose. However two randomized control trials (Kaya et al., 2010; Hong et al., 2012) reduced consumption frequency by 55% and 45%, using 0.5ug/kg and 1.0ug/kg respectively. A meta-analysis demonstrated a mean difference of -10mg vs -21mg in opioid consumption for bolus studies vs bolus and infusion studies at 24 hours (Schnabel et. al 2013). A reduction of 50% was recently seen in a study conducted earlier by the investigators center at 24hours with a bolus and infusion (RQHR #13-80/USask Bio #13-232). An expected Patient Controlled Analgesia (PCA) consumption at 24hours for total knee arthroplasty is 27mg with a standard deviation of 19mg (Paul et. al, 2013). Using this information, it was decided 40% reduction in opioid consumption at 24hours could be expected and represent a clinically significant result. Subsequent power calculation required a sample size of 50 (25 per arm) to demonstrate a 40% reduction in opioid consumption at 24 hours, with a study power of 80% and P-value under 0.05.

Ethics. Ethics approval was obtained from the University of Saskatchewan Biomedical Research Ethics Board, and from the Regina Qu'Appelle Health Region Research Ethics Board.

Patients will be identified by one of the investigators through the daily surgical slate to which they already have access for patient care.

Informed written consent will be obtained from 50 patients on the Same-Day Admission Unit by one of the investigators not involved in the patient's care.

Randomization and Allocation. Following informed consent by one of the investigators, patients will be randomized into one of two groups based on a 1:1 ratio. Sealed envelopes will be prepared by the Department of Anesthesia's Research Coordinator using a web-based random number generator, and opened by a researcher (Dr. Maslany or Dr. Vipulananthan) independent of the clinical team. The drug will be prepared by a resident or nurse not involved in the patient's case; this staff member will vary depending on who is available when the need arises. Surgeons, anesthetists, ancillary staff and patients will be blinded to patient allocation.

Study Drug. The study drug (which does not require refrigeration) will be kept locked in the Anesthesia office at the Regina General Hospital. It has been arranged that Pharmacy will deliver the study drug in a batch, and the investigators will use a tracking sheet to account for the 25 vials that are required for the study. Both Dexmedetomidine and Normal saline are colorless and thus unrecognizable by either the patient or the anesthetist. There is no risk to the patient of delay in receiving midazolam at an appropriate time once the case is started. If the patient asks for more sedation at any time, it will be provided.

Procedure. Prior to entering the operating room, all patients will be familiarized with the Numerical Rating Scale (NRS) and instructed how to use patient controlled analgesia and instructions to press the PCA demand button if their NRS pain is 4 or greater. All patients will receive a 500cc bolus of lactated ringer's solution intravascular volume loading during spinal anesthetic delivery. Monitors include electrocardiography, non-invasive blood pressure measurement, pulse oximetry, and end-tidal carbon dioxide concentration with nasal prongs for monitoring respiration. Oxygen will be delivered at 3 litres per min.

Using a computer-generated randomization table by blinded staff, patients will be randomly allocated 1:1 to receive 0.5ug/kg of dexmedetomidine (experimental group) or same volume of normal saline (control group) over 10 min. The current standard of care, though variable, entails patients receiving a midazolam bolus for sedation during the operation, if the patient requests sedation. For patients who request sedation, a syringe of dexmedetomidine or saline will be run as a single dose infused over 10 minutes. However, a midazolam bolus (0-4 mg IV) will be available at the anesthetist's discretion to achieve a moderate sedation score as defined by the American Society of Anesthesiology.

Bupivicaine 0.75% 1.7cc (12.75mg) and fentanyl 10 micrograms will be administered intrathecally for analgesia 5 min after infusion has been completed.

The level of sensory block will be assessed, as per standard care, with pinpricks and ice cubes. Motor block will be assessed with a modified Bromage scale (0=no paralysis; 1=unable to raise extended leg; 2=unable to flex knee; 3=unable to flex ankle) (Bromage et al. 1964).

Rescue phenylephrine and ephedrine will be available to the anesthesiologist to use at their discretion for hypotension.

Patients will be discharged from post-anesthetic care unit, as per standard practice, once discharge criteria met as per modified Aldrete scoring system. For Total Knee Arthroplasty, the average PACU stay would be 45-60min. Patients will be kept until they meet the requirements of the modified Aldrete score and no longer, as per standard care.

Data collection. Baseline characteristics will be collected by Dr. Vipulananthan from the patients' charts, including: age, weight, height, sex, duration of surgery, baseline heart rate and blood pressure. Primary and secondary outcome measures will be collected, including: total morphine consumption at 6, 12, and 24hours, time to first morphine request, pain scores (Numerical Rating Scale) at 6, 12, and 24 hours at rest, intraoperative midazolam use, time of readiness of discharge from the post-anesthetic care unit, intraoperative and post-anesthetic care unit hemodynamics, duration of sensory and motor blockade by two-dermatome sensory regression, recovery of L2 dermatome sensation and knee flexion, adverse opioid effects of nausea, vomiting, pruritis, urinary retention, post-operative shivering, and patient satisfaction.

Analysis. Statistical methods will include multivariate analysis of variance and Mann-Whitney U-test to compare groups. There will be no crossover between groups and results will be analysed on an intention to treat basis. Sub-group analyses of Type II diabetic patients will be analyzed for insulin resistance. Statistical significance will be accepted at a probability level of under 0.05.