In Reply

We thank Dr. Schwenk and Dr. Gandhi for their insightful comments to our article.¹  We agree that the dynamometer readings do seem variable in the adductor canal block group. It possibly could be due to the block techniques. Although we standardized the approach to the adductor canal block (mid-thigh, depositing 15 ml of the local anesthetic to spread around the lateral, anterior, and medial portions of the superficial femoral artery), having different anesthesiologists perform the block can lead to some interoperator variability. However, more likely, we think it is due to variability in the patient strength and effort in performing the dynamometer examination. It would have been interesting to present patients' change from baseline values to help eliminate the effect of baseline variability between patients’ strength. By recording percentage of difference of strength within each patient, we would have been able to examine the blocks (true) ability to preserve or weaken quadriceps strength. Reasons why there is less variability in the femoral nerve block group may be that significant weakness among most of the group prevented stronger patients (e.g., younger male patients) from performing the examination better in comparison to others (e.g., older female patients). In other words, the femoral nerve block effectively shut down the quadriceps muscles to a point that it no longer depended on patients' efforts to perform the examination.

We do agree that an alternative study design for our study would have been to enroll patients with an intravenous patient-controlled analgesia and/or to include a control (no block) group. But given that our institutional standard of practice comprise of an epidural patient-controlled analgesia and a block, which are included in our hospital preoperative education classes and the patient handbook, we decided to proceed with the study with an epidural patient-controlled analgesia and a block (adductor canal block [ACB] or femoral nerve block [FNB]). We did not think that it would be ethical to intentionally give patients worse analgesia, to have a study that would be easier to interpret. Nine years ago, at our institution, we conducted a study of whether an epidural patient-controlled analgesia would be enough to provide effective analgesia after total knee arthroplasty with or without a femoral nerve block.²  The study demonstrated that the addition of a FNB significantly (P = 0.045) improved visual analog scale pain scores. Our institution promotes early ambulation and regardless if a patient remains in the recovery room, our team of physical therapists visit them on postoperative day (POD) 0. Our objective in performing the adductor canal block is to enable patients to ambulate as early as possible. We agree that further studies need to be done to demonstrate the clinical significance of quadriceps sparing and early ambulation with adductor canal blocks. As for continuous catheters, we are opting to use additives instead, such as dexamethasone, which probably prolong nerve blockade. Our department is currently conducting a study involving dexamethasone with bupivacaine for adductor canal blocks in anterior cruciate ligament surgeries (currently in enrollment).

We would like to thank Dr. Deloach and Dr. Boezaart for their pertinent comments on our article. Effective multimodal analgesia should be the mainstay at any hospital in treating postoperative pain for joint arthroplasty. We, however, do not believe that with an epidural infusion and oral analgesics, any or no block would suffice in providing adequate analgesia for total knee arthroplasties. We do not agree that “neither single-injection FNB nor ACB could be expected to last to 24 h.” In the setting of total knee arthroplasty with epidural analgesia, our randomized-controlled trial showed that analgesia is improved by FNB for 2 days.²  Contrary to the assertions that the buckling was a “coincidental finding after the single-injection nerve blocks have long worn off,” it is probable that the nerve block contributed to buckling in the FNB group. Among patients receiving epidural analgesia after total knee arthroplasty, buckling occurred on POD1 among 34% of patients who received FNB (vs. 3% without FNB).² 

As for “the question of whether volume, concentrations or total dose has a more significant effect on analgesia, motor function, and spread of LA [local anesthesia] has not been conclusively answered,” we agree that this question should be further investigated, but it is not a relevant confounder to this study. Before enrollment, the literature was scarce on the location, volume and concentration of local anesthesia to use for an adductor canal block. Before we designed the study, a group of our anesthesiologists gained clinical experience with different distances (from the knee), volumes (10, 15, 20, and 30 ml), and concentrations (0.25, 0.5, 0.75% bupivacaine). It seemed that the most effective regimen was 0.5% bupivacaine, 15 ml, at mid-thigh. We found volumes of 20–30 ml led to significant quadriceps weakness, most likely due to proximal spread. Our institutional practice is to use 30 ml of 0.25% bupivacaine for our femoral nerve blocks. We agree that effective postoperative analgesia is a major enabler of early ambulation. However, the FNB has been demonstrated to significantly weaken the quadriceps to preclude any ambulation on POD0.^3,4  We agree that it is not just the saphenous nerve that is being blocked but several that ensure adequate analgesia. It is the limited volume that prevents spread to the most proximal muscular branches, allowing the motor sparing.^4,5 

Finally, we would like to thank Dr. Ziemann-Gimmel for his thoughtful comments on our article.

We agree that ideally a more conservative smaller M and a larger sample size n would have strengthened the conclusions of noninferiority. However, when the study design was being finalized in 2010, there was very little literature on adductor canal blocks. At the time, another study was being conducted by our group, comparing subsartorial saphenous nerve block to femoral nerve block for advanced cardiac life reconstruction.⁶  The study’s preliminary data demonstrated equivalent postoperative analgesia, with no difference in opioid use. We decided to design a study that focused more on the strength sparing benefit of the adductor canal block. Moreover, other studies have since been published that tout the adductor canal block’s analgesic efficacy.⁷  Although fall-risk has not been attributed to regional anesthesia as shown by Dr. Memtsoudis,⁸  previous ambulation, quicker recovery, and shorter hospital stays are more pertinent reasons to perform this block. The purpose of our study was to demonstrate that adductor canal blocks preserve motor strength with adequate analgesia. We agree that future studies need to be done to demonstrate ACB’s probable superiority to FNB in enabling a quicker recovery, leading to even day zero ambulation.