Discussion
As we all know, the success of peripheral nerve blocks depends on the accuracy with which the nerves are localized and impregnated. However, it has reported that other relevant factors can affect the success rate and quality of nerve blocks, including the concentration and volume of local anesthetic injected in proximity of the nerves[20, 21]. In this concentration-finding study, we found that the median EC50 was 0.113%(95%CI, 0.108% to 0.343%).
There are various approaches for sciatic nerve blocks, including popliteal, subgluteal or mid-femoral approaches performed under US guidance[22, 23]. Among these approaches, the subgluteal approach showed following advantages. First of all, the depth of the puncture needle through the subgluteal approach is shallow, and ultrasound can more clearly show the course and structural characteristics of the sciatic nerve. Secondly, it was shorter that the onset time of subgluteal sciatic nerve block compared with that of popliteal sciatic nerve block, but the effect of block was basically comparable among the three groups[24].
Circumferential injection is beneficial for US-guided sciatic nerve block[25]. Compared with the multiple injections of LA used to achieve circumferential spreading[26], a single-injection technique was used in our study. It was found that, after injection of 20mL of LA, all patients could achieve circumferential spreading around the sciatic nerve. This is one of the reasons why the EC50 of ropivacaine concentration is so low. Meanwhile, multiple injections for circumferential spreading of LA should be used with caution, since it might cause patient discomfort.
A previous study by Frost et al.[27] found that femoral nerve blockade did not reduce postoperative analgesic requirements in patients undergoing ACL reconstruction when a hamstring graft was harvested. This is anatomically meaningful because femoral blockade only covers the anterior thigh and knee, while hamstring graft harvesting would likely lead to posterior thigh and knee pain, which is the sciatic nerve sensory distribution. A retrospective study demonstrated that the pain scores of patients with hamstring autograft was higher than those patients who received allograft[11]. However, in the study of Jansen et al.[14], it showed no association between hamstring autograft and higher PACU pain scores or increased opioid consumption. In our study, the patients who were conducted femoral nerve block combined with sciatic nerve block had low PACU pain scores and low opioid consumption perioperatively. The posterior knee pain during ACL reconstruction surgery is possibly unrelated to graft harvest site but rather may be due to surgical factors (manipulation, drilling a hole in the tibia), posterior knee edema, tourniquet pain, or a combination of factors.
This study used a relatively high volume of local anesthetic (20mL of ropivacaine) for the nerve blocks. Although lower amounts of local anesthetic can be used, these volumes are used in our practice for prolonging duration. There may be some adverse effects on motor function and early mobilization due to the dose of local anesthetic used; however, our surgeons do not want any knee mobilization on the day of surgery. Despite the amount of local anesthetic used and use of FNB over a more motor-sparing adductor canal block, we saw no decline in our study. This is in accordance with a previous study by Memtsoudis et al[28], which suggests that peripheral nerve blocks are not the major culprit of falls after knee surgery.
This study has limitations. UDM allows determination of an EC50 for a clinical variable with a binary outcome[29], while in a smaller sample size. As we all know that the UDM is unreliable when calculating small or large percentage points, such as the EC95[30], which is a more relevant indicator for clinical application. Although the EC95 level may be more clinically useful, our simulation calculations results in 29 small samples were significantly less accurate.
When calculating EC50 with UDM, the premise is that the concentration-effect relationship is the traditional s-shaped curve, which may be incorrect. It is not accurate to speculate the EC95.
Thus, we conclude that the median EC50 is 0.113%. Further concentration-comparative studies are needed for other volume of ropivacaine and multiple-injection techniques as well as to strengthen the validity of the results of our study.