Assessment/Procedure

Patients randomized to the ST had a single device placed on the arm or forearm at the discretion of the inserter, and those randomized to a DT had the initial tourniquet placed followed by a second approximately 30 centimeters (cm) distal to the first (Figure 1). In both cases, vein diameter was measured both prior to and after placement of a tourniquet.

Standard tourniquet pictured left; double tourniquet pictured right.

The investigators were trained to perform uniform bedside assessment of the venous system including measuring and saving vessel depth and diameter. The linear array transducer was used for all insertions, either a L12-4s Mindray M9 unit (Mindray North America, San Jose, CA) or a HFL38xp Sonosite X-Porte (FUJIFILM Sonosite, Inc, Bothell, WA) depending on patient proximity to the device within the department. Investigators measured vessels in short-axis orientation and assessed diameter pre- and post- tourniquet. The US site director reviewed all scans and measurements for accuracy. There were no discrepancies in measurements of vein depth between the initial operators and the director’s review. A 4.78 cm 20-gauge, peripheral IV catheter was used for the evaluation. Investigators recorded first-stick success, number of attempts, and overall success of the procedure as well as the location of insertion (upper arm, antecubital, forearm) of all successful placements. A successful insertion was confirmed by full advancement of the catheter so that the catheter was no longer externally visible, immediate sampling of at least five cubic centimeters (cc) of blood, and flushing without resistance of five cc of saline.

Patient, IV, and vein characteristics at time of initial assessment were summarized for each study group using medians and interquartile ranges for continuous variables and compared by Kruskal-Wallis tests. We summarized frequencies and percentages for categorical variables and compared them by chi-squared tests or Fisher’s exact tests. Logistic regression was employed to assess the strength of association between type of IV inserters and first-stick success. All tests of statistical significance were two-sided with a p-value < 0.05 indicating a significant difference. We performed analyses using SAS version 9.4 (SAS Institute, Inc., Cary, NC).

Additional data collected from the electronic health record included age, gender, body mass index (BMI), vitals signs, and relevant past medical history.

The primary outcome was the first-stick success rate of obtaining venous access between the two groups. This was assessed by the number of attempts to establish access. We performed statistical comparisons using chi-squared and Fisher’s exact tests. All tests of statistical significance were two-sided with a p-value <0.05 indicating significant difference.