Sir: In their contribution on self-extubation (SE) Moons et al. [1] note that patients in medical ICUs had a greater incidence of SE than those in surgical ICUs. Others [2] have also speculated on the possible differences in SE rates when comparing stable postoperative cardiac surgical patients to medical patients. We previously kept prospective data on all airway accidents in our ICU [3, 4]. Using these data we specifically addressed the issue of the type of patient in relation to the rate of unplanned extubation (UE) [5]. We categorized patients into three groups, cardiac surgical, other surgical and medical. We then analyzed the frequency of UE both in terms of patients and in terms of ventilator days. The data are shown in Table 1. In terms of patients we noted that cardiac surgical patients had lower rates of UE than medical patients. This, however, was explained by the shorter duration of ventilation, as there was no difference in the frequency of UE between the three groups in terms of events per ventilator days.

Table 1 Frequency of unplanned extubation (UE) in medical (M), cardiac surgical (CS), and other surgical (OS) intensive care units
Full size table

Our data, unlike those of Moons et al., thus suggest that UE occurs equally commonly in medical and surgical patients. This difference could be explained by the structure of the ICU. In our setup all patient are ventilated in a large medical-surgical ICU. All protocols regarding sedation, weaning, nursing supervision are established by consultant intensivists. All ventilator, airway, and weaning decisions are made by a single team of ICU consultants, junior medical staff, nurses, and respiratory technicians. This ensures a high degree of uniformity. Moons et al. suggest that different teams manage different ICUs with relatively varying protocols and practices. This difference in delivery of intensive care may partially or completely account for the differential SE rates noted by them in medical and surgical ICUs.

In these two studies spanning 6 years the UE rate was 0.3% per patient and 0.17% per ventilated day. We continue to keep detailed records prospectively of all airway accidents. From January 2002 to date we have ventilated 2,915 patients. There were a total of 21 (0.72%) endotracheal tube displacements; 12 (0.4%) were cases of SE, 6 of which needed immediate reintubation, while one patient was reintubated 14 h after the event. Seven (0.24%) had partial displacement of the endotracheal tube with the tip of the tube dislodged to the pharynx. Two of these patients were extubated at the time of the event while the other five had the same tube repositioned laryngoscopically. Two patients had a temporary dislodgement of the endotracheal tube during a percutaneous tracheostomy procedure.

It is plausible that our uniform and protocolized airway care is the main reason for our relatively low overall UE rate