Prevalence of tracheostomy in ICU patients. A nation-wide survey in Switzerland

Objective: To assess the frequency, timing and technique of tracheostomy an d its variation between different intensive care units (ICUs) in Switzerlan d. Design: Retrospective, descriptive prevalence study. Setting: A question naire was sent to all intensive care units formally recognized by the Swiss Society of Intensive Care Medicine. Excluded were paediatric ICUs. A total of 48 ICUs (70 %) responded. Patients: In 1995 and 1996 the participating units had admitted 90,412 patients for a total of 243,921 ICU days. Results : Seventy percent of the contacted ICUs answered the questionnaire. The pre valence of tracheostomy was 10% in the longterm ventilated patients (define d as > 24 h), or 1.3 % of all patients. Most tracheostomies were performed during the 2nd week of ventilation. The frequency of tracheostomy varied wi dely (0-60 %) and was only slightly associated with the different language regions of our country and with the policy of hospitals to accept or refuse intubated patients on their normal wards. Most units offered either conven tional surgical tracheostomy (69 %) andior percutaneous procedures (57 %). The decision to perform a tracheostomy was made mostly by the intensivist a nd the procedure was more often performed in the ICU (65 %) than in the ope rating theatre (35 %). Units where the intensivist had exclusive control us ed only percutaneous techniques. An overall complication rate of 13 % was r eported, bleeding and infections being at the top of the scale. Only 27 % o f the units performed late fellow-up protocols. Conclusions: Despite its fr equency, tracheostomy in Swiss ICUs is far from being standardized with reg ard to indication, timing and choice of technique.