Objective We investigated whether ultrasound guidance was advantageous over the anatomical landmark technique when performed by inexperienced paediatricians.

Design Randomised controlled trial.

Setting A paediatric intensive care unit of a teaching hospital.

Patients 80 children (aged 28 days to <14 years).

Interventions Internal jugular vein cannulation with ultrasound guidance in real time or the anatomical landmark technique.

Main outcome measures Success rate, success rate on the first attempt, success rate within three attempts, puncture time, number of attempts required for success and occurrence of complications.

Results We found a higher success rate in the ultrasound guidance than in the control group (95% vs 61%, respectively; p<0.001; relative risk (RR)=0.64, 95% CI (CI) 0.50 to 0.83). Success on the first attempt was seen in 95% and 34% of venous punctures in the US guidance and control groups, respectively (p<0.001; RR=0.35, 95% CI 0.23 to 0.54). Fewer than three attempts were required to achieve success in 95% of patients in the US guidance group but only 44% in the control group (p<0.001; RR=0.46, 95% CI 0.32 to 0.66). Haematomas, inadvertent arterial punctures, the number of attempts and the puncture time were all significantly lower in the ultrasound guidance than in the control group (p<0.015 for all).

Conclusions Critically ill children may benefit from the ultrasound guidance for internal jugular cannulation, even when the procedure is performed by operators with limited experience.

Trial registration number RBR-4t35tk.