Transient Neurologic Symptoms After Spinal Anesthesia

Abstract

To the Editor: The paper by Hampl et al. [1] is of interest in that it follows up on several reports of neurologic injury by hyperbaric 5% lidocaine after subarachnoid anesthesia and attempts to study the problem prospectively. The results indicate a high incidence of transient neurologic symptoms in the hyperbaric 5% lidocaine group (44/120) and only one case (1/150) in the hyperbaric 0.5% bupivacaine group. The difference in results between the two groups is noteworthy but must be viewed with great caution. The study did not control for the confounding variables of duration, position, and specific type of surgery, and therefore one cannot exclude those factors as being the cause of the difference in incidences of neurologic symptoms. Although the differences in position and duration of procedures are indicated and the implications discussed, the types of surgical procedures and their impact on the results are not. This latter information is crucial and could explain the authors' findings. The patients in the 5% lidocaine group underwent procedures lasting 16 min compared with 51 min in the 0.5% bupivacaine group. The difference in surgical duration suggests the procedures were different in the two groups and that the bupivacaine group underwent more extensive and traumatic surgery. In addition, one can also surmise that the patients' postoperative course, including pain from the surgical procedure, use of analgesics, level of activity, time to ambulation, psychological response to surgery, and length of stay in the hospital were also different. These factors alone might explain the differences in results found by the authors. Indeed, the patients with the more extensive surgery and the more intense surgical pain may not have noticed subtle sensory neurologic changes. Furthermore, more analgesia and sedation in the bupivacaine group with the more extensive surgery could have masked the symptoms and also potentially affected the recollection of symptoms since the evaluation by the quality assurance nurses occurred on postoperative day 3. One can assume the length of stay was shorter in the lidocaine group, and this raises the question whether the earlier return to routine activity at home had made the symptoms more apparent. In addition, it wouldn't be surprising that a larger percentage of the lidocaine group was interviewed by telephone, being out of the hospital earlier, and one can only wonder what effect this could have had on the results. Along with the other flaws in design, this variability in the type of data collection used would introduce a bias. This bias may have had unpredictable effects on the measurement of a parameter that was very subjective in nature. Was it possible that those contacted by telephone were more likely to complain? In addition to the factors mentioned by Dr. Carpenter in his editorial [2] and the authors in their paper, the lack of data on the specific surgical procedure performed, the analgesia used, the length of stay in the hospital, and differences in how the interview was performed would further confound the interpretation of these results. We would urge Dr. Hampl and colleagues to repeat this important study using bupivacaine randomly in half the short procedure group, blinding the anesthesiologist, and doing more frequent neurological assessments postoperatively, preferably using direct rather than telephone interviews. Unless the results of this study are verified in a prospective randomized study, we plan to continue to use 5% lidocaine when the clinical situation warrants its use. So far, published case reports, retrospective analyses, and the unrandomized study by Hampl et al. are not enough to outweigh hyperbaric 5% lidocaine's long history of safe use. Ralph Erian, MD Charles Hantler, MD Dale Solomon, MD Department of Anesthesiology The University of Texas Health Science Center at San Antonio San Antonio, TX 78284-7838