Kawamoto M, Mukaida K, Maehara Y, Yuge O
Department of Anesthesiology and Critical Care, Hiroshima University Faculty of Medicine, Kasumi 1-2-3, Minami-ku, Hiroshima 734-8551.
In Vivo. 2001 Jan-Feb;15(1):45-8.
An abnormally accelerated Ca-induced Ca release (CICR) rate is known to be correlated with malignant hyperthermia susceptibility (MHS).
To analyze significant clinical findings concerning CICR rate and develop a computer program for its prediction in human MHS.
Using data from 146 subjects who had received a muscle biopsy for the determination of CICR rate, because of their anesthesia-related MHS history, we analyzed 23 different clinical features. There were 71 subjects with an abnormally accelerated CICR rate and 75 with a normal rate. Accelerated CICR rate was used as the objective variable whilst clinical findings and ages were used as independent variables and control variables, respectively. A multiple logistic regression model was employed for the analyses and the most suitable formulae for prediction were determined for use in the development of a computer program.
The following 8 clinical findings were determined to be the most significant: the presence of muscle rigidity, the most serious PaCO2 reading (mmHg), peak body temperature (degree C), body temperature rate of increase over 15 minutes (degree C/15 minutes), most serious arterial pH reading, administration of dantrolene, improvement of acidosis with dantrolene, and time elapsed to peak body temperature after administration of anesthetics (minutes). By ranking the subject ages into 14 groups, we were able to minimize the prediction error rate with each corresponding formula. The computer program developed for prediction whilst consisted of these formulae yielded a sensitivity of 80% and a specificity of 86%.
This method of prediction may contribute to the accurate prediction of CICR rate at the bedside. For clinical convenience, we will distribute the computer program upon request.
已知钙诱导的钙释放（CICR）速率异常加快与恶性高热易感性（MHS）相关。
分析有关CICR速率的重要临床发现，并开发一个用于预测人类MHS中CICR速率的计算机程序。
利用146名因麻醉相关的MHS病史而接受肌肉活检以测定CICR速率的受试者的数据，我们分析了23种不同的临床特征。其中71名受试者的CICR速率异常加快，75名受试者的速率正常。将加快的CICR速率用作目标变量，而将临床发现和年龄分别用作自变量和控制变量。采用多元逻辑回归模型进行分析，并确定了最适合的预测公式以用于开发计算机程序。
确定以下8项临床发现最为重要：肌肉强直的存在、最高的动脉血二氧化碳分压读数（mmHg）、最高体温（摄氏度）、15分钟内体温升高速率（摄氏度/15分钟）、最低的动脉血pH读数、丹曲林的使用、丹曲林改善酸中毒情况以及麻醉给药后至最高体温的时间（分钟）。通过将受试者年龄分为14组，我们能够使用每个相应公式将预测错误率降至最低。为预测而开发的计算机程序由这些公式组成，其灵敏度为80%，特异度为86%。
这种预测方法可能有助于在床边准确预测CICR速率。为方便临床使用，我们将根据要求分发该计算机程序。
