Precipitation of thiopental with muscle relaxants: a potential hazard

We report an interesting phenomenon we recently encountered in our hospital. Our patient was a healthy man who came in for an emergency hernia repair due to suspected obstruction. Previous anaesthesia for an orthopaedic operation the previous year had been uneventful. The plan was to anaesthetize him with a rapid sequence induction. This was done in view of the increased risk of aspiration given that he was obstructed. It was decided to use a modified rapid sequence induction with rocuronium and thiopental sodium. The induction was uneventful; preoxygenation was followed by 375 mg of sodium thiopental and 40 mg of rocuronium through a 20 G cannula connected to a 1000 mL Hartmann’s solution. Upon intubation it was noted that the patient was not fully relaxed despite an appropriate dose of relaxant. We then noticed that our previously free running Hartmann’s infusion had stopped and that there was a 10 cm long column of a fluid containing a flaky precipitate in the IV line. Efforts to flush the line with 10 mL of saline through the injection port were unsuccessful. The patient was re-cannulated on the opposite hand, and a dose of suxamethonium was given to aid fast muscle relaxation with cricoid pressure still in place. The patient was then intubated without any harm. Surgery, subsequent extubation and recovery period were uneventful and the patient was discharged home later that day. The initial cannula was removed and examined more closely to show that the lumen had indeed occluded by the flaky precipitate. Following this clinical scenario we carried out some simple experiments. We drew up a syringe of commonly used muscle relaxants and pipetted five 1 mL aliquots onto a flat, sterile surface. To each of these we then added 1 mL of sodium thiopental to see their interaction in vitro. The results are shown in Figure 1. Whereas the interaction with mivacurium produced practically no reaction, this did not appear to be the case with the other relaxants we tested. Suxamethonium produced a mild precipitate, followed by atracurium. Vecuronium and rocuronium, however, produced a rather marked crystallization, producing macroscopically visible precipitates. This may have been responsible for our scenario where rocuronium given through a poorly flowing cannula may have led to crystallization of thiopental and subsequent occlusion of the cannula/vessel. To further test how dilution effected the crystallization we added small amounts of Hartmann’s solution to see if the precipitate would re-dissolve (if at all). Mixing with Hartmann’s solution brings about some degree of dilution, however, no evident dissolving of the crystals. Rocuronium, in particular, still produced a considerably flaky precipitate.