Background: Visualization and palpation are standard methods for identifying peripheral IV sites. Vein diameter and depth significantly influence catheter dwell time and complication rates. This study aimed to determine the diagnostic performance of visualization and palpation in determining suitability for cannulation. Methods: In this prospective observational study, healthy adult volunteers were enrolled from a senior recreational facility and an office setting. Two vascular access nurses independently assessed both arms of each participant using the Vascular Access Score (VAS): 1 = easily palpable, distended, visible; 2 = easily palpable, not distended, visible; 3 = easily palpable, not visible; 4 = poorly palpable, visible; 5 = nonpalpable, not visible. Scores of 1–3 were considered suitable for traditional cannulation, and scores of 4–5 were considered unsuitable, prompting escalation. Ultrasound was used to measure vein parameters with vein diameter ⩽0.3 cm and vein depth >0.4 cm considered suboptimal. Primary outcome was diagnostic accuracy. Results: From March 10 to April 15, 2025, 100 adults were enrolled, yielding 1000 IV site assessments. Median age was 71 years; 63% were female; median BMI 28.2 kg/m 2 ; 93% identified as Caucasian. Mean vein depth was 0.35 cm (26.2% > 0.4 cm) and mean diameter 0.35 cm (47.8% ⩽ 0.3 cm). After adjudication, and after excluding hand and wrist sites as well as missing values, VAS distribution with 483 IV sites was 1 = 22.77%, 2 = 4.76%, 3 = 53.21%, 4 = 7.45%, 5 = 11.80%. The sensitivity of visualization/palpation method for determining if the vein was suitable for traditional insertion was 99.32%, specificity was 27.46%, PPV was 37.69%, NPV was 98.92%, and accuracy was 49.48%. Conclusion: Vein assessment via standard visualization/palpation effectively rules out inappropriate sites and should trigger early ultrasound escalation. However, clinically “suitable” veins are frequently suboptimal on ultrasound, supporting broader use of ultrasound to guide initial cannulation decisions.
Bahl et al. (2026) studied this question.