In Reply:-Dr. Marsch and Dr. Staender correctly criticize our study for including only 10 volunteers. [1]In defense, however, it took several years to convince 10 volunteers to undergo both a spinal anesthetic and magnetic resonance imaging. Furthermore, we do not have funding to pay for additional imaging procedures. Consequently, it is extremely unlikely that we will be able to expand our database.

Dr. Marsch and Dr. Staender are also correct that one volunteer had an extremely limited spread of sensory anesthesia. Indeed, it is quite logical to conclude that the anesthetic was a technical failure and that the data should be discarded. However, we do not believe this was a technical failure because, as the authors suggested, we performed two spinal anesthetics on this volunteer and he had a similar spread of sensory analgesia on both occasions. Furthermore, the volunteer was thin, anatomic landmarks were normal, and both procedures were technically uncomplicated. Cerebrospinal fluid (CSF) was aspirated before and after injection of the local anesthetic during each procedure. One anesthetic was performed with 50 mg lidocaine and the other with 50 mg lidocaine plus 0.2 mg epinephrine. The maximum sensory block level was L3 on one occasion and L4 on the other.

As you know, our group has performed numerous studies in which each volunteer received two, or more, spinal anesthetics. During the conduct of these studies, we became impressed by the relative consistency in peak sensory block level achieved in individual patients. Indeed, the consistency in volunteers with extremely low or high sensory block levels was the primary incentive for performing axial imaging. We suspected that anatomic variability may correlate with variability in spread of spinal anesthesia.

In retrospect, we are fortunate to have enrolled a volunteer with such an extreme CSF volume (approximately 4 standard deviations more than the mean). The finding that peak sensory block height in this volunteer was several standard deviations less than the mean of the group supports our conclusion regarding the relation between CSF volume and the distribution of spinal anesthesia. Although we believe that data from this patient should be included in the statistical analysis and that our conclusions are valid, we caution that there is no magic to the 0.05 threshold for statistical significance. Despite achieving the threshold for statistical significance, the correlation we observed could still be caused by chance. Similarly, we do not believe it is prudent to completely dismiss an interesting correlation just because the P value is 0.07. We fully agree that additional data are necessary to conclusively establish the relation between lumbosacral CSF volume and the extent and duration of spinal anesthesia. We hope that additional studies will be performed to either confirm or refute our conclusions.

Randall L. Carpenter, M.D.

Director; Clinical Research; Astra Pharmaceuticals, L.P.; Westborough, Massachusetts 01581

Quinn H. Hogan, M.D.

Department of Anesthesiology; Medical College of Wisconsin; Milwaukee, Wisconsin 53226

Spencer S. Liu, M.D.

Staff Anesthesiologist; Virginia Mason Medical Center; Seattle, Washington 98111

(Accepted for publication November 13, 1998.)

Carpenter RL, Hogan QH, Liu SS, Crane B, Moore J: Lumbosacral cerebrospinal fluid volume is the primary determinant of sensory block extent and duration during spinal anesthesia. Anesthesiology 1998; 89:24-9