[Failure of spinal anesthesia. Evaluation of the practice at a university hospital].

There is little data concerning failures in spinal anaesthesia. A retrospective analysis of 337 spinal anaesthesias performed in a teaching hospital gave a 9.8% failure rate. A failure was defined as the need to carry out part or all of a surgical act under general anaesthesia when spinal anaesthesia had been carried out. The main causes of failure were insufficiently or excessively extended neural blockade, insufficient duration or a poor quality sensory blockade with the patient feeling surgical or tourniquet pain. The local anaesthetics used were hyperbaric 0.5% tetracaine with and without 0.1 mg metaraminol, hyperbaric 0.5% prilocaine and isobaric 0.5% bupivacaine. The failure rate was 19.4% with plain hyperbaric tetracaine, 7.6% with tetracaine with metaraminol, 5% with prilocaine and 2.9% with isobaric bupivacaine. This rate was related neither to the experience of the anaesthetist, nor to the clinical features of the patients, nor to the position nor to the needle size. Although there were more failures during orthopaedic procedures, probably because of tourniquet pain, this was not significant. These results confirmed that spinal anaesthesia was a reliable technique. Hyperbaric plain tetracaine did not always guarantee a successful anaesthesia and isobaric bupivacaine should be more commonly used.