BMC Anesthesiology (Nov 2023)

Nominal logistic regression analysis of variables determining needle visibility in ultrasound images – a full factorial cadaver study

  • Hanns-Christian Dinges,
  • Julia Hoeft,
  • Valér Michael Cornelius,
  • Thorsten Steinfeldt,
  • Thomas Wiesmann,
  • Hinnerk Wulf,
  • Ann-Kristin Schubert

DOI
https://doi.org/10.1186/s12871-023-02339-y
Journal volume & issue
Vol. 23, no. 1
pp. 1 – 8

Abstract

Read online

Abstract Background Needle visualization is essential to avoid vascular puncture and nerve injury in ultrasound-guided regional anesthesia. Several factors that statistically influence needle visibility have been described but the dimensions of their individual impact remain unclear. This study aimed to quantify the impact of various independent factors on ultrasound needle visibility. Methods A total of 1500 ultrasound videos of in-plane needle insertions were obtained in embalmed cadavers with ten different commercially available echogenic and non-echogenic needles at different insertion angles and bevel orientations in a full factorial study design. The visibility of needle tip and shaft were rated as “good” or “poor” visibility. Nominal logistic regression analyses were calculated for the visibility of the needle tip and shaft. Results SonoPlex Stim Sprotte, SonoTAP Facet (needle tip and shaft) and Spinostar PencilPoint (needle tip)), insertion angle and bevel orientation were associated with good ultrasound visibility, reaching statistical significance (p < 0.05). The range of the effect on the log-odds scale for needle tip visibility was largest for the insertion angle with 6.33, followed by the tissue condition (3.76), bevel orientation (1.45) and the needle types (1.25). Regarding the needle shaft visibility, the largest effect range was observed with the insertion angle (7.36), followed by the tissue conditions with 3.96, needle type (1.86) and bevel orientation (0.95). Conclusion In-plane needle visibility in ultrasound images depends mainly on the insertion angle, as expected. This is closely followed by the tissue condition, which is a factor related to the patient, thus cannot be altered to improve needle visibility. In the dimensions of the log-odds scale, the choice of a specific needle is far less important towards achieving a good visualization, whereas optimizing the bevel orientation can have a larger impact than the needle choice. Concluding from the relative dimensions of factors that determine needle visibility in this model, the importance of needles with echogenic features may be overrated.

Keywords