Gopalasingam 2014.

Study characteristics
Methods	Single‐centre prospective randomised controlled 2‐arm cross‐over participant‐blinded study Denmark Duration: November 2012 to July 2013
Participants	40 participants randomised (experimental = 20, comparator = 20) 40 participants and 80 arteries analysed mean age (years, range): 71 (46 to 91) gender (male/female): 28/12 severity of condition (experimental/comparator): scheduled surgeries (coronary artery bypass graft = 25, mitral valve replacement = 1, aortic valve replacement = 11, combination procedures = 3) comorbidities (experimental/comparator not reported): hypercholesterolaemia = 34, diabetes = 10, smoking = 16, hypertension = 34 body weight (kg): 71 (46 to 91) height (cm): 170 (157 to 190) artery of interest: radial diameter (mm), mean ± SD: not reported. Catheter diameter 20 G catheterisation purpose (experimental/comparator): all for diagnosis (pressure monitoring or blood test, not detailed) Inclusion criteria aged 20 to 90 years undergoing elective cardiac surgery with routine preoperative catheterisation of the radial artery Exclusion criteria lack of patient consent ultrasound verification of arterial plaques no‐flow in either radial or ulnar artery or atrial fibrillation
Interventions	Experimental: ultrasound‐guided RA puncture (real‐time, artery short axis, and DNTP) as first intervention Comparator: RA puncture with palpation and landmarks as first intervention Level of experience of person carrying out the procedure: anaesthesiology residents (operators) had performed at least 20 of each procedure previous to this study Concomitant medications: after injection of 0.5 to 1.0 mL of lidocaine (10 mg/mL) Excluded medications: not reported
Outcomes	Primary (specified) attempts [Time Frame: 1 hour]: number of attempts (skin punctures) per catheterisation withdrawals [Time Frame: 1 hour]: number of withdrawals of the guide needle per catheterisation time consume [Time Frame: minutes]: time spent on catheterisation procedure catheters [Time Frame: 1 hour]: number of utilised catheters Primary (collected) first‐attempt success rate Secondary (specified) pain [Time Frame: momentan]. Pain induced by the conventional method inclusive of preoperational lidocaine injection will be the same or more intense than using DNTT with local anesthesia measured on a VAS score ease of method for the operator [Time Frame: momentan]. Use of ultrasound will increase the operator's subjective feeling of having accomplished a successful procedure on a Likert scale Secondary (collected) number of skin perforations number of attempts targeting the vessel (withdrawal of the guide cannula while the catheter remained inserted) needle manipulation time (beginning with catheter perforation of the skin) total time (group beginning when transducer was first placed on the patient’s skin (DNTP group), and when the operator started to palpate the patient’s skin (standard group)) fraction of total time > 180 seconds defining unexpected difficult catheterisation number of catheters used frequency of aborted attempts or cross‐overs (pooled) measurement of pain on a visual analogue scale (VAS) Time points reported: up to the end of the procedure (not specified)
Notes	Funding: quote: "this work was funded by the Edgar Schnohr and wife Gilberte Schnohr’s fund and Helga and Peter Korning’s foundation" Conflicts of interest: quote: "none of the authors has financial interest related to this study to disclose" Protocol (NCT01690416) available Study authors provided data for the first phase, before the intervention cross‐over
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "computer randomisation (www.randomization.com) was conducted"
Allocation concealment (selection bias)	Unclear risk	No details provided Quote: "the randomisation order was revealed to the observer just prior to catheterisation"
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participant blinded, but personnel not blinded Quote: "the study was randomised, controlled, patient‐blinded"
Blinding of outcome assessment (detection bias) All outcomes	High risk	Blinding only for participants Quote: "the study was randomised, controlled, patient‐blinded"
Incomplete outcome data (attrition bias) All outcomes	Low risk	There were no losses. Besides the cross‐over design, another cross‐over (not planned) occurred in 4 participants in the comparator group (to ultrasound aid) and in none in the experimental group
Selective reporting (reporting bias)	Low risk	All prespecified outcomes were reported
Other bias	High risk	There is no plausible reason to change the outcome from primary on the protocol to secondary on the final report