We followed with interest the recent correspondence [1-3] regarding an incident of accidental administration of thiopental rather than antibiotics before caesarean section . There appears to be agreement that the largest causative factor in this and other similar incidents is human error. Suggestions for reducing this error have included the use of pre-filled syringes and avoidance of thiopental altogether. These solutions, however, are not without their own problems  and even a pre-filled syringe may be picked up and administered in error.
We believe that the method of antibiotic administration we currently use offers additional safety in avoiding this particular error. Our standard antibiotic prophylaxis before caesarean section is cefuroxime and metronidazole, mixed in theatre via the spiking of a bag of metronidazole 500 mg with an ampoule of cefuroxime 1.5 g. This is given pre-incision, as per National Institute for Health and Clinical Excellence guidelines . The process of preparing and administering the drugs is very different (mixing the ampoule into the bag then connecting to a giving set, rather than drawing drug from the ampoule into a syringe and injecting as a bolus). This in itself would reduce the chance of error. We then pondered what would happen if despite this, a thiopental ampoule was selected in error and mixed with the metronidazole. It can be seen from Fig. 2 that if this was to occur, a thick white precipitant forms in the bag, which we believe would alert the anaesthetist to their error before administration, thus acting as an extra safety net.
The preparation described has been in routine use at the Liverpool Women's Hospital since March 2011, so we estimate that it has been used without incident in over 3000 caesarean sections. Before this, for several years a similar system was used whereby the cefuroxime and metronidazole were supplied together, pre-attached. Although cheaper than the pre-attached set, the current preparation is approximately three times the cost of giving cefuroxime and metronidazole separately (or co-amoxiclav, which is slightly cheaper again), making cost the main drawback compared with alternatives. We believe that this is outweighed by the advantage of allowing antibiotics to be given more safely before incision.