Effect of a Single Shot Sciatic Nerve Block Combined with a Continuous Femoral Block on Pain Scores After Knee Arthroplasty. A Randomized Controlled Trial ()
1. Introduction
Postoperative pain is a concern for patients submitted to knee arthroplasty and it has been reported as being moderate in up to 30% and severe in 60% of the patients [1]. There is an ongoing debate on which type of anesthesia is better for these patients but there is no clear evidence of any advantage of regional versus general anesthesia on long-term outcomes such as morbidity, mortality or rehabilitation [2]. Regional anesthesia provides better postoperative analgesia, and may improve early rehabilitation with fewer side effects when compared to intravenous opiates [2,3], but there are no clear conclusions on what type of regional anesthesia is better.
Peripheral nerve blocks have less side effects compared to neuraxial blocks and femoral nerve blocks are well established as an option for postoperative analgesia in knee replacement surgery [4] and is now recommended by the Procedure Specific Postoperative Pain Management Working Group [5]. There is uncertainty on whether sciatic or obturator nerve blocks are needed as adjuncts to a femoral block to improve postoperative analgesia, and conflicting results have been published [6-11]. A review by Hogan et al. states the benefits of peripheral nerve blocks in analgesia for hip and knee analgesia, but fails to recognize which are the most effective blocks [12]. Two meta-analysis recently published showed inconclusive evidence to define the effect of adding sciatic nerve block (SNB) to femoral nerve block (FNB) on acute pain and related outcomes due to the lack of good randomized studies comparing both techniques [13]. An editorial by Ilfeld & Madison in 2011 recognized the lack of consensus and the need for additional research [14].
The aim of this study is to determine if a combination of a single shot SNB and continuous femoral nerve block (CFNB) provide better postoperative analgesia when compared to isolated CFNB in patients undergoing total knee arthroplasty (TKA).
Our hypothesis is that the association of a single shot SNB to a CFNB can reduce mean pain scores at least 15% when compared to CFNB alone, as we considered this to be a clinical significant difference.
2. Material and Methods
The study protocol was approved by the Ethics Committee of Centro Hospitalar do Porto and received institutional approval. It was registered in Clinicaltrials.gov under the identifier NCT01337115.
2.1. Type of Study
Interventional, prospective, randomized, controlled, parallel group, single blinded, treatment efficacy study.
2.2. Selection of Participants
Eligible participants were drawn from 167 consecutive patients undergoing unilateral TKA from April 2011- February 2012, in the Orthopedic Department of Centro Hospitalar do Porto, Portugal.
Patients had to be anesthetized by one of 5 senior anesthesiologists of the orthopedic anesthesia group of the department. Exclusion criteria were refusal to give informed consent, contraindication to general anesthesia, infection at needle insertion site, coagulation disorders, preexisting neurologic disorders, known allergies to local anesthetics, diclofenac or tramadol, severe dyspepsia, ASA status 4 or 5, weigh less than 50 kg, body mass index > 40, inability to understand and use the visual analogue scale (VAS) pain score.
Group allocation was made by a computer generated random number list (Office Excel 2011®, Microsoft Corporation, Redmond, WA, USA) and codes stored in opaque sealed envelops.
Written consent was obtained immediately prior to envelope opening and group allocation.
2.3. Intervention
We used the Stimuplex® HNS12 nerve stimulator (B. Braun Melsungen AG, Germany) to perform the nerve blocks. For CFNB, the paravascular approach [15] was used to identify the femoral nerve. We considered a positive location when quadriceps contraction (patellar elevation) was elicited with a current of 0.46 mA or less [16]. Ten milliliters of ropivacaine 0.375% were injected through the needle (Contiplex® Tuohy 18 G 50 mm length needle, with 20 G, 100 mm catheter, B. Braun Melsungen AG, Germany), a catheter was then inserted and 20 ml of ropivacaine 0.375% were administered through the catheter.
For the single shot sciatic block, we used the anterior approach described by Chelly [17]. A Stimuplex® insulated needle 21 G × 100 mm or 20 G × 150 mm was used (B. Braun Melsungen, Germany) and 25 ml of ropivacaine 0.375% were administered when either the common peroneal or tibial nerve (dorsiflexion or plantar flexion of the foot) were identified with a stimulation of 0.46 mA or less [16]. Blocks success was assessed by absence of thermal sensitivity with an alcohol swab on the anterior region of the thigh and dorsum of the foot.
All patients were then induced to general anesthesia with propofol (1.2 - 2 mg/kg) and fentanyl (0.15 - 0.2 μg/kg) and maintained with short action halogenated anesthetics (Desflurane or Sevoflurane). Airway was maintained with laryngeal mask airway and pressure support ventilation. Additional fentanyl was given during the procedure if attending anesthesiologist considered necessary. Total intraoperative fentanyl dose was registered.
Thirty minutes before the end of the procedure, IV ketorolac 30 mg and paracetamol 1000 mg was administered.
On arrival to the post-anesthesia unit (PACU), neurologic function was assessed by an independent observer and an infusion of ropivacaine 0.2% was started on the femoral catheter at a rate of 8 ml/h using an Easypump® C-bloc RA 400-8 (B. Braun Melsungen AG, Germany).
Both groups got diclofenac 50 mg q12h PO and paracetamol 1000 mg q6h PO and as rescue analgesia, tramadol 100 mg q6h prn IV. Thromboprophylaxis with LMWH was started in the first 24 h and maintained for 5 weeks.
2.4. Data Collection
We assessed pain using a standard 100 mm VAS (VAS = 0 no pain; VAS = 100 worst pain).
At 15 - 30 minutes after arrival to PACU, with the patient awake and fully collaborating, there was a first pain evaluation. If VAS > 30, and the patient asked for treatment, morphine 2 mg IV was given every 15 minutes until VAS 30 or less and patient was comfortable.
VAS pain scores, morphine consumption, and time of discharge from PACU were recorded. Presence of toe movements was confirmed before PACU discharge.
Additional measurements of VAS pain scores (VASr: at rest; VASm: with movement) at 12 h and 24 h (+/– 2 h) after surgery were made.
Rescue tramadol usage and complications (falls, foot paralysis, paresthesias, nausea and vomiting) were registered up to 24 h.
One month after surgery, subjects were contacted by telephone and asked to rate their satisfaction with postoperative analgesia in a categorical scale of 4 groups (bad; average; good; excellent).
Investigators blinded for the allocation group of the subjects made all assessments.
2.5. Variables
Pain scores on VAS is a continuous variable with a range of 0 - 100.
The primary outcome of the trial is the difference in mean scores from VAS pain assessment in both groups (0 = no pain; 100 mm = worst pain) at three predefined moments on the first 24 h after surgery: 15 - 30 min (in PACU); 12 h and 24 h (+/– 2 h).
Tramadol consumption and patient satisfaction are ordinal variables.
Side effects and complications (as previously defined) are reported and incidence calculated.
2.6. Statistical Analysis
Fifty subjects were randomized into Control (CFNB; n = 25) and Intervention (CFNB + SNB; n = 25) groups.
Power analysis for sample size was calculated using a web free application [18] and was designed to detect minimum differences of 15%, with type I error 0.05 and 80% power.
A two-tailed t-test for independent variables was used to analyze differences in mean VAS scores in intentionto-treat manner.
The normal distribution of the continuous variables was assessed by the Kolmogorov-Smirnov test.
Our null hypothesis states that there is no difference between the mean pain scores of control and intervention groups. Statistical significance is assumed if P < 0.05.
Ordinal variables were analyzed with Fisher’s exact test [19].
Continuous variables analysis was done with IBM® PASWSatistics 18.0.3 for MAC (IBM Corporation, Somers, NY, USA).
3. Results
A total of 167 subjects were assessed for eligibility from April 2011 until February 2012. A high number of excluded patients were due to the fact that only 5 senior anesthesiologists were involved in the study.
Figure 1 shows a flowchart of patient selection and participation in the study.
Two patients were excluded before the 12 h assessment because of femoral catheter displacement. The drop out of 2 patients in the control group (CFNB) did not affect the statistical power because the sample needed only 23 patients in each group.
Table 1 shows the patient demographic data. There were no differences between groups. The higher number of female subjects is consistent with the reported higher risk of osteoarthritis in women and with the average caseload of our department.