We appreciate the interest expressed in our recent publication. Whereas it was not our intention to incite such passionate and emotional responses from our respected peers, we certainly welcomed and carefully considered the many comments regarding the interpretation of our data and implications of our findings. In our article, we simply described a phenomenon, i.e. , that the deliberate elicitation of a paresthesia by advancing a needle to a nerve root during interscalene block is not always associated with the ability to elicit a motor response to electrical stimulation. Conversely, we have also observed what Carter and Sandberg reported, 1that is, that a motor response to needle advancement is not always associated with a paresthesia. Our observations are reproducible and similar observations have been made at other institutions during axillary block. 2We have videotaped the phenomenon. We do not completely understand nor pretend to be able to definitively explain the easily observed pronounced dissociation between sensory and motor nerve stimulation that occurs in many patients.
Upon review of these letters, there clearly exists some misunderstanding with regard to our intentions. It was not our intention to endorse or condemn any single technique or needle type. At our institution, we routinely use paresthesia techniques, peripheral nerve stimulation techniques, long-beveled needles, and short-beveled needles. With proper technique, we are convinced that all of these are very useful and extremely safe. Use of peripheral nerve stimulation to locate nerves for local anesthetic blockade was a major advance in regional anesthetic practice.
It appears that the questions raised by the authors of all three letters relate either to our methodology or our conclusions.
Regarding our methodology, our reported results cannot simply be dismissed as being secondary to a faulty nerve stimulator that: 1) was brand new, 2) had new batteries, the voltage output of which was checked by digital voltmeter immediately prior to each use, 3) was functionally checked by the ability to transcutaneously stimulate the facial nerve of the first author at 2 mA immediately prior to each use, 4) was equipped with a digital meter indicating the delivered current amplitude in mA. Further, this phenomenon has been observed with more than one manufactured brand of nerve stimulator.
Regarding our conclusions, we believe that our data interpretation was sound and responsible and that we kept an open mind when interpreting new data rather than clinging to preconceived notions that lack scientific support. To this end, we were puzzled by the inconsistencies, double standards, and biased conclusions included in the letter by Choquet et al. 3In their letter, these authors cite the need for a “calibrated oscilloscope” but drew significant conclusions based upon several referenced studies that lacked this same rigorous standard. They state that “confirmation of the needle tip being in contact with the nerve is obtained by eliciting a paresthesia with each pulsation,” based upon reference to a review article published in 1985 that neither specifically studied this nor provided supporting experimental data for this statement. 4It appears to be anecdotal. They state “when responses are observed for currents this low (motor response at 0.1–0.2 mA for 200 μs), injection should be avoided…” but later contradict themselves by advocating injection following motor response “at low current (0.2–0.5 mA for 100 μs),” again, without data. The fact is that there are presently no compelling data that link injection at low current to nerve injury. Injection at low minimal current has never been prospectively compared to injection at higher minimal currents. Indeed, all three catastrophic complications reported by Benumof 5that involved the use of a peripheral nerve stimulator, reported injection with current levels in the 0.81 mA to 1.0 mA range. Choquet et al. include editorial commentary on the presumed advantages of short-beveled needles compared to long-beveled needles without statistically significant data from prospective controlled studies to support their views. The fact is, as Neal et al. recently published, “there are no randomized clinical trials to support or refute the ability of various needle types and bevel configurations to impale human nerves.”6Choquet et al. omit relevant data that short-beveled needles caused more severe nerve lesions in animal studies and that the lesions took longer to heal. 7The “tendency of short-beveled needles to press or push the nerve away” is a myth. With minimal pressure, under direct vision, a short-beveled needle will travel into the substance of an isolated nerve 7,8or through an anesthetized animal nerve (William F. Urmey, M.D., unpublished observations; 22-gauge, short-bevel needle was easily passed through brachial plexus fascia and nerves in an anesthetized dog, Harvard School of Public Health, Boston, MA, October 1987). In fact, three of the four catastrophic cases reported by Benumof 5documented the use of a short-beveled needle that clearly entered the neural tissue itself. Finally, Choquet et al. claim “paresthesia(s) are usually difficult to elicit with a short beveled insulated needle,” but contradict this a few lines later by stating “frequent unintentional paresthesia” occur during block placement by withdrawal and stimulation of short-beveled needles.
On the other hand, we thank Casati et al. 9and Carter and Sandberg 1for their constructive critiques and agree that an “ability to elicit a paresthesia in every patient does not constitute irrefutable evidence of direct nerve contact.” There is no obvious way to ethically prove this. Perhaps two-dimensional ultrasonographic examination of nerves during paresthesia can eventually help to answer this question. (Joseph Kay, M.D., Assistant Professor of Anesthesiology, University of Toronto, Sunnybrook Hospital, Toronto, Canada, personal communication, July 2002). Nevertheless, it is extremely unlikely and represents very wishful thinking to believe that a needle deliberately advanced in the direction of superficial large diameter nerve roots causes paresthesia without ever making nerve contact. Our contention that nerve contact was the event that caused paresthesia is supported by the depth at which paresthesia was reported. By contrast, we agree that with use of a peripheral nerve stimulator, it is clear that a motor response may be elicited without contact with the needle's tip, at a distance from the nerve. Nevertheless, in practice, we have frequently encountered further evidence that the needle tip may be in the nerve when motor response is elicited with peripheral nerve stimulator. Often, motor responses occur at very low amperage, disappear with further advance of the needle, only to reappear when the needle is slightly retracted. This constitutes credible evidence that the shielded needle tip may travel through the nerve during our searching. It is illogical to believe that the needle's shaft rather than its point caused the mechanical paresthesias observed in our study; as the needles were all advanced axially, initial contact would be with the needle's point. This does not mean that the needle tip was intrafascicular or intraneuronal. This we believe is exceedingly rare based upon the extremely low incidence of permanent neuropraxias that occur in association with interscalene block.
In summary, there is an old saying, “In life, you will be forgiven your lies, but heaven help you if you attempt to tell the truth.” In our opinion, it is denial on our part as anesthesiologists to perpetuate and cling to folklore that we can deliberately aim and advance needles at superficial, large nerves, gently sneaking up on them while never making nerve contact. Although it is tempting to believe in such a never proven concept of “immaculate conduction,” there is compelling scientific evidence and unfortunate complications of intraneural damage that argue the contrary. That is, there is every indication and reason to believe that nerve contact occurs during peripheral nerve or plexus blockade. Maybe we are asking the wrong questions. Perhaps, when considering peripheral nerve location by needle exploration, the burden of proof should not lie with having to demonstrate that paresthesia occurs in response to needle tip to nerve contact. Instead, we might begin to ask ourselves how we can prove that no nerve contact occurs when we elicit sensory or motor responses during peripheral nerve blocks. Despite this theoretical routine nerve contact, with responsible technique, peripheral nerve blocks are extremely safe. 10Use of either accepted paresthesia or electrical nerve stimulation techniques are very safe, if performed carefully on patients who are not overly sedated or anesthetized. Our publication as well as that of Choyce et al. 2represent an initiative in determining the relationship between paresthesia and stimulator techniques used for nerve location. Additional, more sophisticated studies are needed. Forthcoming data should be analyzed openly rather than defensively.