Controversies in resuscitation: to infuse or not to infuse (1)

Abstract

In an influential article on immediate (on site) versus delayed (in hospital) fluid resuscitation for hypotensive patients with penetrating torso injury, Bickell and his colleagues [1] from Houston suggested that the latter strategy was usually optimal and that ‘aggressive’ infusion should be delayed until the time of definitive operative intervention. This view was based on the results of observations that were prospective and randomised. Although at variance with recent conventional wisdom [2] they were not open to the principal criticisms of Jones and Brenneis [3] who, in 1991, reviewed critically the nine relevant studies available at that time. A plea was made for prospective randomised trials involving standard&d or widely accepted levels of service for victims with well documented blunt, penetrating, and thermal injuries of at least moderate severity. Many observations based on retrospective studies with variable case mix comprising patients with penetrating injury, blunt injury, and even brain trauma, had drawn conclusions of limited validity. General guidance can be drawn only from an analysis of studies that are better focused. The study of Bickell et al. warrants careful consideration. Bickell and colleagues [1] hypothesise that a major disadvantage of on-site aggressive fluid replacement is accentuated bleeding and haemodilution. But in the pre-hospital phase this is very difficult to measure. Another objection to immediate infusion relates to the time wasted in line placement, especially by paramedics [4,5]. This criticism has substance, although the delays quoted in published data have been very variable. Cwinn and others [6] reported that the time for insertion of one or two venous lines was approximately 14 min, but with an overall success rate of 94%. Donovan et al. [7] found the time for line placement was approximately 20 min, increasing onscene time by 6 or 7 min. There was no evidence of any benefit to the patients from this intervention. On the other hand, Spaite and colleagues [8] reported that the average times for positioning onscene intravenous lines were 1.3 min for successful attempts and 2.1 min for unsuccessful ones. Even during transport, only 2 min were required, and overall the success rate was 98%. Another paper from the same centre [9] concluded that, in an urban Emergency Medical System with strong medical control, short on-scene times can be achieved without adversely affecting subsequent advanced life support. Jones et al. [IO] reported that intravenous line placement could be achieved by over 90% of paramedics in approximately 2.5 min. Similarly, Gervin [ 1 I] investigated the use of large bore intravenous cannulae in trauma patients and found that 83% were successfully introduced by paramedics in a mean time of 50 s. As a result of infusions averaging 4.2 litres of crystalloid over an 18&n transport time, mean blood pressure increased from 70 to 96 mmHg. The clinical need for pre-hospital infusion, as well as its logistics, has also been contentious. Scalea [ 121 conducted a study using retrospective