395

Abstract

Introduction: Early tracheostomy has been used in many ICUs to help weaning patients from mechanical ventilation. A recent publication of a randomised controlled trial casted doubt about the effectiveness of this strategy [1]. We carried out a meta-analysis of the available published data on the insertion of an early tracheostomy to assess the impact on mortality, duration of mechanical ventilation, duration of sedation and the incidence of ventilator-associated-pneumonia when compared to prolonged endotracheal intubation. Methods: We included only randomised controlled trials in the meta- analysis – cross-over and quasi-randomised designs were not permitted. We searched the MEDLINE, EMBASE and CINAHL databases from 1946 to 2013. Our outcome measures included the duration of mechanical ventilation, 28d mortality, 28d incidence of VAP and duration of sedation. We used Mantel-Haenszel method with random effects model for the dichotomous data and inverse variance method with random effects model for the continuous variables. Results: We have identified 8 eligible trials with 1860 recruited patients. Duration of mechanical ventilation was not significantly different in the two groups and despite being in favour of early tracheostomy, and the I2 test suggests significant heterogeneity (-3.20 95%CI -7.15, 0.76 days, respectively; I2=91%). We observed similar results when assessed duration of sedation and duration of ICU stay. 28 day mortality was also not significantly different between the two groups (RR 0.84 95%CI 0.67, 1.04; I2=31%). Similarly, the incidence of VAP was not statistically different in the two groups, with large heterogenity observed between the studies (RR 0.92 95%CI 0.76, 1.11; I2=73%). Conclusions: Despite hypothesised and plausible clinical benefits, early tracheostomy does not appear to carry any benefit in a heterogeneous patient population. This finding is further compounded by the problem of significant heterogeneity of the studies. We believe possible explanations for the negative findings are: Early tracheostomy was thought to be beneficial when our sedation practices were very different. Most of the early studies lacked sedation scoring, sedation breaks and adequate analgesia. This change in clinical practice might explain the negative results in the more recent trials. The other common limiting factor in all of these studies was the clinicians ability to predict the length of mechanical ventilation needed and therefore the inability to accurately predict, which patients would benefit from such an intervention. We recommend that further studies are carried out with strict sedation protocol and a clear definition of early tracheostomy and prolonged endotracheal intubation. Reference: 1. Young D, et al Journal of the American Medical Association May 2013;309(20):2121–2129.