Bupivacaine and intravenous regional anaesthesia

When the late Lord Nuffield proposed to endow a Chair in Anaesthesia he was told ‘Anyone can give an anaesthetic’, to which he replied ‘That’s what I’m afraid of‘ (Sir Robert MacIntosh, personal communication). More recently, intravenous regional anaesthesia (IVRA) has grown in popularity as a simple technique that that same ‘anyone’ can give, on a ‘see one -do one-teach one’ basis. Is the spate of deaths reported in the lay press the result of such philosophy? Or can the problem be solved simply by prohibiting the use of bupivacaine for the purpose, as the removal of IVRA from the Marcain Data Sheet would imply? The situation is actually more complex. In addition to the lay press reports, adverse reactions to bupivacaine IVRA before cuff deflation have been reported to the Scientific and Technical Branch of the Department of Health and Social Security because faulty equipment was implicated,’ and those following cuff deflation to the Committee of Safety of Medicine because bupivacaine was implicated. Meanwhile, numerous serious but non-fatal adverse reactions have been reported in anaesthetic journals. These occurrences have stimulated, firstly a ding-dong correspondence in the pages of this Journal over the past 3 years, and secondly some careful research in the field, two such papers (pages 147 & 150) appearing in this issue. Death from IVRA may result from error of technique or drug dosage, together with failure of resuscitation. Disasters have certainly resulted from tourniquet failure. Tourniquet equipment should be simple, and continuously supervised by a medical attendant other than the operator. However, it has recently become clear that injected solution can leak beneath a correctly inflated cuff.* In this issue Lawes and his colleagues (page 147) have shown that venous pressure at the elbow may exceed a cuff pressure of 50 mmHg above systolic. This is true particularly if the injection is made rapidly, or into a proximal vein, if the volume injected is large, or if the limb is not well exsanguinated. To the latter end an Esmarch bandage is clearly optimal, but rarely applicable in Casualty. Brachial artery compression during arm elevation is an effective and often neglected alternative. However, even the most assiduous technique cannot obliterate interosseous venous channels, nor does prolonged tourniquet time necessarily make cuff deflation harmless (though rapid reinflation can be helpful), so a safe dose of local anaesthetic is also essential. Bupivacaine 0.2% was recommended for IVRA in 1975 by Ware, who found 0.125% less reliable. Since that time he has reported its use for over 14000 cases without m i ~ h a p . ~ However, the concentration used has crept up in other quarters and serious convulsions have been reported, while early cardiac depression is also described,2 and has been a feature in some of the fatalities. Similar cases of early circulatory depression following epidural b~pivacaine,~ prompted the suggestion that bupivacaine was more cardiotoxic than lignocaine. Convulsions from local anaesthetic toxicity are associated with a rapid onset of severe hypoxia and acidosi~,~ and this, coupled with medullary depression, is a more likely cause of cardiac depression than any direct cardiotoxic effect. Moreover, in ventilated cats6 and awake and ventilated anaesthetised dogs,’ the ratio of cardiovascular to central nervous system toxic doses is at least as high for bupivacaine as for lignocaine. However, cardiotoxicity of local anaesthetics (whether direct or indirect) may be increased by hypoxia and hyperkalaemia, and in this respect bupivacaine may in theory be a worse culprit than lignocaine or prilocaine, as its higher pK. may cause it to be more concentrated in acidotic tissues. Herein lies the vital importance of rapid and efficient resuscitation. Bupivacaine 0.2% has been shown to be more effective than lignocaine 0.5% for IVRA,8 which is not surprising as bupivacaine is four times as potent. However, side effects, albeit mild, were only