A Live Human Model Comparison Evaluating ThoraSite^® Accuracy for Needle Thoracostomy

Needle thoracostomy (NT) is a time-sensitive procedure infrequently performed by EMS clinicians with variable success rates. Our primary objective was to evaluate the accuracy of NT site selection by paramedics using ThoraSite^® compared to traditional anatomic landmarks (ALs). Secondarily, we assessed paramedic-rated confidence and ease of ThoraSite^® use.

We conducted a randomized, two-arm crossover study including fire-based paramedics. Emergency physician investigators determined a NT placement zone for live human models in three size groups, confirming with ultrasound and demarcating the zone with "invisible" ultraviolet ink. Following training, paramedics performed NT site selection on the models using ThoraSite^® and ALs by placing a sticker at the selected insertion site. Accuracy of placement was confirmed with ultraviolet flashlight. If placement was outside the demarcated zone (DZ), we identified underlying structures with ultrasound. We evaluated the effect of approach on placement accuracy and time-to-NT placement using linear models with covariates of paramedic, approach, and model size. For the outcome of accuracy, we used a log link function. For time-to-NT, we log-transformed the values for the parametric analysis allowing interpretation of the coefficients as percent differences. We compared paramedic confidence in performing the NT procedure and perceived ease of procedure using a 5-point Likert scale.

There were 112 paramedics that performed 223 ThoraSite^® and 223 landmark attempts with 383 correct placements within the DZ: 198 attempts using ThoraSite^® compared to 185 with ALs, odds ratio (OR) 1.91 (95%CI 1.01-3.62), p = 0.04. Placement accuracy by model size followed similar trends. Incorrect placement over critical structures occurred in 1 ThoraSite^® and 3 AL attempts. The mean time for NT site selection was 14.3s (SD = 7.11) using ThoraSite^® and 18.7s (SD = 7.40) using ALs (p < 0.01). Overall procedural confidence improved with training. However, there was no statistically significant difference in the change in confidence with ThoraSite^® as compared to ALs (OR = 1.55 95%CI = 0.89-2.72). Paramedics rated ease of NT placement significantly higher using ThoraSite^® (median = 5, IQR = 4-5) compared to ALs (median = 4, IQR = 4-5; p < 0.01).

ThoraSite^® was associated with increased odds of NT site selection in the DZ, reduced time-to-NT site selection, and increased self-rated ease reported by paramedics.