The effect of three bolus doses of remifentanil on the pressor response to laryngoscopy and tracheal intubation during rapid sequence induction of anaesthesia was assessed in a randomized, double-blind, placebo-controlled study in four groups of 20 patients each. After preoxygenation, anaesthesia was induced with thiopental 5-7 mg kg-1 followed immediately by saline (placebo) or remifentanil 0.5, 1.0 or 1.25 micrograms kg-1 given as a bolus over 30 s. Cricoid pressure was applied just after loss of consciousness. Succinylcholine 1 mg kg-1 was given for neuromuscular block. Laryngoscopy and tracheal intubation were performed 1 min later. Arterial pressure and heart rate were recorded at intervals until 5 min after intubation. Remifentanil 0.5 microgram kg-1 was ineffective in controlling the increase in heart rate and arterial pressure after intubation but the 1.0 and 1.25 micrograms kg-1 doses were effective in controlling the response. The use of the 1.25 micrograms kg-1 dose was however, associated with a decrease in systolic arterial pressure to less than 90 mm Hg in seven of 20 patients.

Percutaneous tracheostomy using a graded dilatation technique was described by Ciaglia in 1985 [1]. The description of other techniques using special forceps to dilate the trachea followed. The safety and low early complication rate of percutaneous tracheostomy compared with open surgical tracheostomy have been established by several prospective studies [2][3][4][5][6][7].The long-term complications, such as tracheal stenosis, remain a source of controversy. Conventional surgical tracheostomy has been associated with a long-term tracheal stenosis rate of between 16 and 64% [2, 8]. Studies using tomography [2] and fibreoptic laryngotracheoscopy [9, 10] have shown low stenosis rates with percutaneous tracheostomy.This postal survey was designed to assess the practice and attitudes to percutaneous tracheostomy in intensive care. Questions were included on the use of fibreoptic bronchoscopy and long-term follow-up. MethodsIn 1997, a postal survey was sent to the clinical directors of 231 general intensive care units in England and Wales.Specialist units, such as pure neuro-surgical or cardiac units were not surveyed. The recipients were requested to fill in a short and simple questionnaire (Appendix). The results were statistically analysed using the Chi-squared test. A p value less than 0.05 was considered statistically significant. ResultsQuestionnaires were returned by 176 out of 231 intensive care units, yielding a 76% response rate. Of the 176 directors who replied, 129 (73.3%) used percutaneous tracheostomy as the method of choice for tracheostomies on their unit (Table 1). Only 32 (18.2%) had never used it. Six (3.4%) had stopped using the technique. In all, 78.4% were using the technique of percutaneous tracheostomy on their ICU patients. FibreoscopyIn seventy-one (49.3%) units, fibreoscopy did not feature as a routine part of the technique of percutaneous tracheostomy (Table 2) and in only 45 (31.3%) was it used routinely.

SummaryThe aim of this study was to determine whether infusion sets containing antisyphon devices increased the time to initial¯ow from syringe drivers. The antisyphon devices assessed were those manufactured by B Braun, Wescott and Vygon. Each device was placed between a 50-ml syringe and a spiral extension set and primed with saline. A fourth syringe and spiral extension set acted as a control. The infusion sets were placed in four identical syringe drivers and started simultaneously. The time from pressing the start button until the initial¯ow for each infusion set (start-up time) was recorded. The test was conducted 15 times each at 2 ml.h À1 , 10 ml.h À1 and 50 ml.h À1 . At 2 ml.h À1 the start-up time was signi®cantly longer with all the antisyphon sets compared with the control (p`0.0001). At higher infusion rates the differences between the antisyphon sets and the control were less pronounced. Clinicians who use syringe driver infusions should be aware of this delay between the activation of the infusion pump and the onset of¯ow and take steps to prevent it.

Editor-We thank Dr Sadacharam for his comments on our study. 1 As a calcium channel blocker, magnesium sulphate causes a dose-dependent negative inotropic effect and peripheral vasodilation. However, in our study, no significant inter-group haemodynamic differences were observed. Some possible explanations are; in our study, ASA I and II patients were enrolled and patients with cardiovascular, hepatic, or renal dysfunction and neuromuscular diseases were excluded. Furthermore, pre-hydration was performed and the magnesium sulphate loading dose was infused slowly over 15 min. In elderly patients and patients with a co-existing medical disease, drugs with potential cardiovascular effects including magnesium sulphate should be administered carefully, as the pharmacokinetics and pharmacodynamics are different from those in young, healthy patients.Magnesium sulphate prolongs and potentiates neuromuscular block by non-depolarizing neuromuscular blocking agents. However, a previous study reported that this effect did not prolong emergence from general anaesthesia, if neuromuscular block was adequately monitored with a nerve stimulator. 6 In addition, minor side-effects of magnesium sulphate such as flushing, nausea, and headache are expected at the serum magnesium level .2 mmol litre 21 , and potentially life-threatening complications involving the neuromuscular, cardiovascular, or both systems occur when the serum Mg concentration exceeds 5 mmol litre 21 . 7 In our study, the peak magnesium concentration [1.31 (0.13), mean (SD)] measured administration was much lower than the level associated with a minor side-effect. Therefore, it is our opinion that magnesium sulphate can be used safely not only with regional anaesthesia but also with general anaesthesia.

No abstract