Article Text

other Versions

Download PDFPDF
Randomised controlled trial
The use of 50% nitrous oxide improves successful intravenous access in obese and growth-limited children in an outpatient setting
  1. David Rosen
  1. Department of Anesthesiology, Perioperative and Pain Medicine, Harvard Medical School, Children's Hospital Boston, Boston, Massachusetts, USA
  1. Correspondence to David Rosen
    Department of Anesthesiology, Perioperative and Pain Medicine, Harvard Medical School, Children's Hospital Boston, 300 Longwood Ave, MA 02115, USA; hdrosen{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Commentary on: OpenUrlCrossRefPubMedWeb of Science


Intravenous therapy is a common intervention in paediatric medicine. Its uses include the delivery of fluids, nutrition, medication and venous blood sampling. Patient anxiety and difficulty with intravenous placement are common. Patients who require frequent peripheral intravenous placement can have progressive difficulty with intravenous access and can also experience increased anxiety related to the procedure. Obesity or growth retardation can make intravenous insertion more difficult.

Topical anaesthetic applied to the skin over an intravenous insertion site has been shown to decrease pain and anxiety related to the procedure.1 Oral midazolam and inhaled nitrous oxide are well studied common pharmacologic therapies noted to relieve anxiety and facilitate cooperation with procedures in children,2 3 but their role in improving successful intravenous access in children has not previously been evaluated.


A total of 90 children were studied prospectively. Of these, 60 were obese and 30 were growth limited. The inclusion criteria included previous difficulty with intravenous access, American Society of Anaesthesiologists Physical Status Classification of 1 and the ability to understand and contribute to the different treatments and ratings. The participants were randomly assigned to three study groups using an opaque envelope technique: oral midazolam (0.3 mg/kg), 10% nitrous oxide and 50% nitrous oxide. All groups were set up to maintain blinding to the treatment assignment (oral placebo or oxygen-only inhalation). All groups had topical anaesthetic cream applied to four potential intravenous sites.

The nurses performing the venous cannulation procedures were blinded to the treatment groups as were the patients and families. Primary end points were total procedure time (intravenous access and recovery from sedation) and the number of attempts required to secure two intravenous sites. Secondary end points were patient and parent satisfaction and procedure cancellations.


Total procedure time was longer in the midazolam group than in the other two groups (p<0.001). The number of attempts was not significantly different between groups (p=0.09). Success was greater with 50% nitrous oxide than for patients treated with 10% nitrous oxide or midazolam (67%, 40% and 37%, respectively). Subgroup analysis revealed a difference in the number of attempts between study groups for the growth limited population (p=0.02) (though the published paper does not indicate which group had fewer attempts). It also demonstrated a more pronounced effect on procedure prolongation for the obese population.(p<0.05) Patient satisfaction was inversely related to both the number of attempts (correlation coefficient r=0.6) and the pain of the procedure (r=0.7). Parents' and nurses' evaluation scores were significantly higher for patients who received 50% nitrous oxide. Adverse effects were observed in two patients: dizziness after receiving midazolam and nausea after receiving 50% nitrous oxide. Four patients in the midazolam group reached a level of sedation that was considered to require a level of monitoring that was not generally available in outpatient clinics. There were no cardio-respiratory adverse events.


Overall, this is a well-designed study. It was prospective, double blinded, randomised and powered to a scientifically-determined end point. One of the few drawbacks was the inclusion of two distinct study populations in one small trial. The two groups – obese and growth-limited patients were found to differ in their demographics, pathophysiology, pharmacokinetics and technical aspects of venous cannulation. The reliance on subgroup analysis to assign a treatment effect could have been obviated if a single population were studied in adequate numbers. There were no patients with normative body mass index included in this study.

The authors' comments about the long recovery time in obese patients after midazolam may have related more to the total dose than a pharmacokinetic effect explained by drug clearance. Obese patients would have received 15 mg while the growth-limited patients of the same age might have received half that total dose. Oral midazolam in higher doses can have a longer duration of action than is required for brief procedures. It can also have unpredictable absorption and pharmacodynamic effects such as agitation and the exacerbation of obstructive sleep apnoea. Care must be taken when providing oral procedural sedation to children at risk for sleep disordered breathing.

The efficacy of 10% nitrous oxide for treating anxiety or pain is not supported in the literature, and the trial could have more efficiently studied two groups only: one receiving oral midazolam and the other receiving 50% nitrous oxide. It is likely that, for the 10% nitrous oxide group, the stress of having a tight-fitting mask applied was greater than any potential beneficial drug effect.

There was little depiction of results data in the publication beyond p values and correlation coefficients. A description was included but it was difficult to track the primary and secondary outcomes as many of them were not presented in detail. The authors did not provide information on actual differences in procedure times, number of attempts or satisfaction measures. This information would be helpful in determining whether the statistical differences between the treatments are clinically meaningful. This study provides some evidence to support the use of 50% nitrous oxide in outpatient settings to provide short-term analgesia and anxiolysis for paediatric patients undergoing intravenous insertion. With proper training, credentialing and patient monitoring, this appears to be a safe and effective way to perform mild to moderately painful procedures such as intravenous insertion in children.


View Abstract


  • Competing interests None.