We thank Lardner and Spencer for their response to our article.
Our conclusion regarding investment in ultrasound and near-infrared imaging for routine IV placement was in reference to our setting and other similar settings, specifically. In our setting, IV placement is a core nursing skill. Nurses in our emergency department place the IVs. As such, they are viewed as experts in starting pediatric IVs. Extrapolating from our enrolment data, we estimate that 70 IVs are started by nurses each week in the pediatric section of our department. Thus, our nurses are experienced at IV placement. Our data support this; across all arms (which showed no difference in first-attempt success rate), the range was 65.9%–74.7% success on first attempt.1
Our emergency department (adult and pediatric) has a pool of roughly 200 nurses who work regularly. Out of this pool, 83 nurses were keen to participate and underwent training. We estimated that only those nurses who felt that the research was important and the technologies might offer benefit, and who were comfortable with the technological procedures would consent to participate. We were satisfied with this approach and assessed that our training package was sensible and similar to other studies.2 Because we agree that training and skill maintenance with ultrasound is complex, we discussed this quandary in the paper in some detail.
Our trial was pragmatic in the sense coined by Schwartz and Lellouch.3 The study design was sound. It was a well-designed randomized controlled trial, adherent to the CONSORT standards of reporting.4
We stand by the results of our study. We are not saying that ultrasound is of no value when performed by experienced clinicians. However, the question does remain: How do we provide the best first-time success to all children who require the placement of an IV line in an emergency setting? Given the training we provided, the results were no better than the usual method.