Spinal anesthesia is a common technique for providing anesthesia for knee joint replacement surgery. The investigators wish to demonstrate that using a lower approach to spinal anesthetic (Taylor's approach of L5-S1) causes less low blood pressure while still providing adequate anesthesia for knee joint replacement surgery than a higher approach (L3-L4).

1. Purpose

   To determine if Taylor's approach (paramedian, via L5-S1 interspace) to subarachnoid blockade produces less hypotension than the more commonly used paramedian lumbar approach (via L3-L4 interspace).

2. Hypothesis

   Taylor's approach to subarachnoid blockade for total knee joint arthroplasty will result in less hypotension which requires treatment than a paramedian lumbar approach.

3. Justification

   Taylor described a method of obtaining spinal anesthesia for urological procedures using a paramedian approach to L5-S1. This interspace is large and tends to be less affected by degenerative changes than the lumbar interspaces. Because of these attributes this approach can be used as an alternative to either midline or paramedian approaches to the lumbar vertebrae interspaces.

   Hypotension is a frequent and important complication of spinal anesthesia. As the degree of hypotension correlates with the height of blockade achieved, performing the dural puncture at a lower interspace may result in less hypotension. Several previous studies have compared the degree of hypotension resulting from a subarachnoid block performed via different lumbar interspaces with conflicting results. However a direct comparison of the lower Taylor's approach with a lumbar approach has not been investigated previously.

   Studies in the past used surface anatomy to determine the level of the lumbar interspaces, which has since been shown to be inaccurate. A study comparing identification of lumbar interspaces by palpation and with ultrasound showed that while palpation was correct 30% of the time, correct identification was able to be achieved by ultrasound 71% of the time. Similar rates of correct identification by ultrasound have been shown in subsequent studies.

   Using ultrasound in this study to identify the relevant Lumbosacral anatomy will enable us to make a direct comparison between these two approaches to subarachnoid blockade with rates of subsequent hypotension as well as ease of procedure and other side effects.

4. Objectives

   a) Primary: To compare the degree of hypotension requiring treatment between Taylor's approach (L5-S1) and a paramedian lumbar approach (at L3-L4) to spinal anesthesia for total knee joint arthroplasty.

   Secondary: For each approach to compare:

   The ease of each technique assessed by:

   • The number of needle insertion attempts and number of needle passes (insertion and redirection attempts) required to establish dural puncture.
   • Procedure time including the time taken to establish landmarks with ultrasound.

   Patient acceptability

   • Patient rated pain score immediately after procedure (0 - 10)
   • Patient rated satisfaction immediately after procedure (1 - 5) Success of central neuraxial blockade
   • The maximum block height achieved
   • Time from completion of administration of spinal to block height of L1 dermatome
   • Time from completion of administration of spinal to regression of block height to L3 dermatome
   • Amount of additional analgesia required during the procedure

5. Research method

To fulfill our objectives the investigators plan to perform a randomized trial comparing the two approaches to spinal anaesthesia on patients undergoing primary total knee joint arthroplasty.

1. Recruitment

   Eligible patients presenting for primary knee arthroplasty surgery at University of British Columbia Hospital will be identified by anesthesiologists at the anesthesia pre-assessment clinic before the day of their surgery. At this stage they will be informed of the study and given an information leaflet to read.

   On the day of the surgery a member of the research team will approach the eligible patient and inform the patient again about the option of participating in the study, what each treatment group entails and answer any questions that the patient may have.

   If the patient is interested in participating in the study they will be asked to sign a consent form and informed that they may withdraw from the study at any time without any consequence. There will be no monetary or material compensation to participants.

   Inclusion criteria are male and female volunteers who are over the age of 19 and are able to understand the study protocol and give their informed consent. They need to be presenting for elective primary total knee joint arthroplasty under spinal anesthesia and be American Society of Anesthesiology (ASA) class 1, 2 or 3.

   Exclusion criteria are patients who are ASA 4 or 5, unable to understand the study protocol or give informed consent. Patients who have elected not to have or are deemed clinically inappropriate for spinal anesthesia will be excluded as well as patients who are presenting for a revision of a knee joint arthroplasty.

2. Randomization

   Prior to the procedure, for each participating patient, a sealed unmarked opaque envelope that contains one of the two treatment groups will be opened. Patients will receive their spinal anaesthetic either via a paramedian lumbar approach or Taylor's approach.

3. Intervention

   For both treatment groups the patients will have the usual preparation for spinal anesthesia, which involves an intravenous line with intravenous fluids running, standard monitoring, and being appropriately positioned on the operating table. The patient will receive premedication of 0.02 - 0.05mg/kg of midazolam. The blood pressure obtained in the anesthesia pre-assessment clinic or in the preoperative area will be taken as the patient baseline blood pressure. An experienced anesthesiologist will then perform an ultrasound scan on the patient with a low-frequency (2 - 5 MHz) curved-array probe, identifying the relevant anatomy, prior to performing the spinal anesthetic.

   Taylor's approach:

   Patients randomised to the Taylor's approach group will first have the sacrum identified with the ultrasound probe in a transverse position in the patients' midline. The transducer will then be rotated 90 degrees into the paramedian axis and the L5 - S1 interspace identified. The position of the transducer will be adjusted to obtain the best image of the interspace with it centred on the ultrasound screen. A marking pen will then be used to mark the midpoints of the probe's long and short edges. The point where these marks would intersect will be used as the initial needle insertion point for spinal anesthesia.

   The lumbosacral area will then be prepared in a sterile manner and with appropriate sterile technique spinal anaesthesia will be performed with a 25-gauge, 90 mm, pencil-point needle inserted through a 20-gauge introducer needle. Once dural puncture has been achieved and cerebral spinal fluid is obtained in the hub of the spinal needle 12.5 mg of 0.5% bupivacaine (preservative free) will be injected.

   Paramedian lumbar approach:

   Patients randomised to the Paramedian lumbar approach group will be scanned in a similar manner to the Taylor's approach group. The sacrum will be identified first, then with the transducer in the paramedian axis the interspaces will be identified and counted up from L5 - S1 to L3 - L4. At the L3 - L4 interspace the image will be optimised and the skin marked as described above. Spinal anesthesia will then proceed in an identical manner with the same dose of bupivacaine.

   After spinal anaesthesia is performed in both groups the patient will remain in the sitting position for 5 minutes and then will be positioned supine for surgery. The patient will have their blood pressure and heart rate measured at 5 minute intervals (or more regularly at the discretion of the anesthesiologist). If a decrease of mean arterial blood pressure occurs, which is greater than 20% from the patients' pre-recorded baseline, or if there is a decrease in heart rate to less than 50 bpm then 5 to 10mg increments of ephedrine will be administered. Sensory block will be assessed by bilateral application of ice at 5 minute intervals. Intravenous fentanyl in 25 - 50 mcg increments will be administered if the patient complains of any pain during the surgery.

4. Data Collection

   The patient's age, height and weight will be recorded, also any significant past medical problems, in particular if the patient has a history of hypertension. Baseline observations including blood pressure will be taken from the anesthetic assessment clinic visit prior to the day of surgery.

   An independent observer will record the time taken to establish landmarks with ultrasound imaging and the time taken to complete the spinal anesthesia. The number of needle insertions and redirections required to achieve dural puncture will also be recorded (as previously defined by Chin et al.). Immediately after the spinal anaesthesia has been administered, the patient will be asked by the observer to rate the pain they experienced during the procedure (from 0 - 10) and their satisfaction with the procedure (from 1 - 5).

   The sensory block height will be measured and recorded at 5 minute intervals. The number of 5 minute increments where the mean arterial blood pressure is measured as being greater than 20% below baseline will be collected as well as the amount of ephedrine required to treat each patient. The amount of intravenous fluid given during the procedure will be recorded. Other details related to the surgery will also be collected including time of first incision, tourniquet inflation and deflation times, estimated total blood loss and time of surgical completion.

   Any additional analgesia requirements during the procedure will be recorded as well as any other unexpected complications, including the need to convert to a general anesthetic.

   After the procedure the block height will be measured and recorded in 15 minute intervals until the block height has regressed to L3.

5. Statistical Analysis

Power calculation is based on the primary outcome of a reduction in hypotension. From clinical experience and a previous study by Carpenter et al. the investigators anticipate a 35 - 40% reduction in the incidence of hypotension in the Taylor's approach group. Based on a p-value of < 0.05 the sample size calculation is 12 patients in each group. The investigators will aim to recruit 15 patients in each group.

Inclusion Criteria:

• Able to understand study protocol and give informed consent
• Presenting for primary total knee joint arthroplasty at University of British Columbia Hospital
• American Society of Anesthesiology (ASA) Class 1, 2 or 3

Exclusion Criteria:

• American Society of Anesthesiology (ASA) Class 4 or 45
• Patients who have elected not to have spinal anesthesia for their procedure
• Patients deemed clinically inappropriate for spinal anesthesia by the attending anesthesiologist
• Patients who are allergic to Local Anesthetic or Ultrasound gel.