535" Correspondence Divided doses of methohexitone improves ECT outcome To the Editor: We read with interest the recent report on substitution of mivacurium for succinylcholine for ECT treatment. ~ Since the effect of mivacurium outlasts the effect of methohexitone (MH) the authors suggested reversal of the neuromuscular block after treatment to avoid awareness. While we are unaware of a correlation between the length of seizures and the antidepressant effect of ECT, the duration of grand real seizures is thought to be important.2 Occasionally, when using lower doses of MH, we have also experienced awareness upon recovery. Further increase of the MH dose to prevent recall shortens the duration of seizures, especially in older patients. To solve this problem we decided to split the total amount of MH to be injected (0.65--0.8 mg.kg -t) into two doses. The first (75% of the assigned dose of MH) is given together with succinylcholine (0.8-1.0 mg. kg -~) before electrical stimulation and the remainder immediately after the convulsions. Labetalol and/or glycopyrrolate are used as indi- cated. We have observed that with the lower first dose of MH the duration of seizures has increased by 16.7% (52 • 4 sec vs 31 _+ 3 see). For the last five months the proposed technique was successfully used 146 times on 17 patients of different ages without any complications. Simon Gurmarnik MD Department of Anesthesiology Richard Young MD Eugenya Alesker MD Center for Behavior Medicine Southwood Community Hospital I I I Dedham St. Norfolk, MA 02056 REFERENCES I Janis K, Hess J, Fabian JA, Gillis M. Substitution of mivacurium for succinylcholine for ECT in elderly pa- tients. Can J Anaesth 1995; 42: 612-3. 20nosonn JO. Seizure characteristics and therapeutic effi- cacy in electroconvulsive therapy: an analysis of the anti- depressive efficiency of grand mal and lidocainc modified seizures. J Nerv Ment Dis 1962: 135: 239-51. REPLY We share their concern about short acting anesthetics for ECT which might create conditions facilitating awareness. However, our use of active reversal of the neuromuscular block was intended to avoid subtle airway problems when a non-depolarizing relaxant was used for short procedures. This was r concern to us since our patients were pre- dominantly elderly with systemic illness which would compro- mise their ability to compensate for residual weakness. We have also used incremental doses of methohexital to assure lack of awareness. Recent work suggests that perhaps etomidate could replace methohexital to aUow for a longer seizure duration along with slower awakening. Kenneth M. Janis MD Associate Professor of Anesthesiology University of New Mexico REFERENCE I Avramov MN, Husain MM, While PF. The comparative effects of methohexital, propofol and etomidate for elec- troconvulsive therapy. Anesth Analg 1995; 81: 596-602. Midazolam for caudal analgesia in children To the Editor: In an editoriaP which accompanied our article, 2 Dr. Goresky, expressed several concerns and doubt about the clinical tttility of caudal midazolam ftbr postoperative analgesia in children. He also raises several questions that warrant answers. In the Kingdom of Saudi Arabia, the rules and regulations that gov- ern the approval of a new drug or a new route of application of an existing drug arc very stringent. The regulatory body (Ministry of Health) is different from that in North America (USA and Canada). Additionally, in our institution (King Khalid University Hospital and King Saud University) file approval of the CMRC (College of Medicine Research Council) and the Pharmacy and Therapeutic Committee are mandatory for any research proposal. Our study was not con- ducted under sponsorship of the manufacturer. Nevertheless, even if this study were supported by the manufacturer, the review process would have been the same. Dr. Goresky chose to ignore the information provided in our paper regarding the safety of both intrathecal and epidural midazolam in primates (as he suggested) and in humans. He quoted only one study in his editorial that reported ch,'mges in the blood-brain barrier after administration of very large doses of midazolam intracisternaUy in rabbits (0.3 ml of 0.1%; equivalent to I I I pg. kg-~). 3 We addressed the limitations of this study in our paper and pointed out that even high concen- trations of lidocaine can produce neurotoxic effects in humans.4.5 We wonder why Dr. Goresky did not refer to the remaining 18 studies in the literature (Medline search reveals 17 published papers in addition to another study that appeared in a chapter of a book)? The safety and the efficacy of neuraxi- al application of midazolam have been invariably demonstrat- ed in all of these studies, both in humans after intrathecal (3 studies) and epidural (6 studies) administration as well as in different species of animals (rats, rabbits and dogs). Due to limitations of the space availability, we referred to some (but not all) of these studies in our paper. Similarly, several studies have shown that the analgesic effect of neuraxial administration of midazolam was due to ,'mtinociceptive properties of midazolam but not due to seda- tion (see, for example6.7). CAN J ANAESTH 1996 /43:5 /pp 535-8