Continuous spinal anaesthesia and non-invasive ventilation for total knee replacement in a patient on home ventilation.

to 10 mg of i.v. midazolam for the above procedures, administered by non-anaesthetists. Yet, diazepam seems to be the drug of choice for treating seizures on the ward, when loraze-pam is unavailable. Patients are at increased risk of a lowered consciousness/prolonged post-ictal phase after receiving bolus doses of i.v. diazepam 10 mg. Midazolam may be a better choice in these situations because of its shorter half-life when compared with diazepam, especially if advanced airway support can be avoided in the absence of other indications. Editor—We thank Dr Patle for bringing our attention to the recent publication of results from the RAMPART study. 1 We were unable to include these results in our review, 2 as this trial was published after the review article was accepted for publication. The RAMPART study randomized 1023 patients with status epilepticus (SE) treated before arrival in emergency departments by emergency medical services to treatment with i.v. lorazepam and i.m. midazolam. Of these, 893 were included in the intention-to-treat analysis. The primary outcome measure, cessation of treatment without need for rescue therapy, was achieved in 329 of 448 subjects (73.4%) in the i.m. midazolam group and in 282 of 445 (63.4%) in the i.v. lorazepam group (absolute difference , 10%; 95% confidence interval, 4.0–16.1; P,0.001 for both non-inferiority and superiority). The two treatment groups had similar incidence of recurrence of seizures and need for tracheal intubation. The time to administration of medication was shorter with i.m. midazolam, and the time from administration to cessation of seizures was shorter with i.v. lorazepam. Adverse event rates were similar in the two groups. We agree that the results of the RAMPART study should have a significant impact on the early management of seizures. As Dr Patle points out, lorazepam is not readily available in many UK hospitals, and i.v. diazepam is traditionally used as an alternative. The results of this study, while supporting the use of i.m. midazolam as a possible alternative, do not provide any information about the safety of this form of treatment compared with i.v. diazepam in the in-patient setting. However, a previous meta-analysis has suggested that midazolam by any route is superior to diaze-pam in terminating SE in children and young adults, with similar rates of respiratory suppression. 3 We agree that i.m. midazolam should be considered as the first-line therapy for convulsive SE, especially in patients where i.v. access is difficult or delayed. 1 …