Reconsidering the Clinical Efficacy of Wide-Awake Local Anesthesia Without Tourniquet: A Trial Sequential Analysis

We read with great interest the article by Levit et al ¹ comparing trigger finger release using wide-awake local anesthesia without tourniquet (WALANT) and local anesthesia with tourniquet (LAWT). The findings resulted in a statistically nonsignificant reduction in pain during injection (mean difference [MD] −1.69; 95% CI −4.14 to 0.76; P = .18). This suggests that WALANT may not have clinical benefits over LAWT.

A trial sequential analysis (TSA) should be conducted to get the required information size (RIS) or the number of patients necessary to ascertain conclusive results and assess whether additional trials could potentially alter the reported nonsignificant findings. ² Based on the TSA graph, if the z-curve crosses the trial sequential monitoring boundary or RIS, sufficient evidence for the anticipated effect may have been obtained, and no further trials are needed.

To address this gap, we used raw data from the original meta-analysis to perform a TSA with TSA version 0.9.5.10 Beta (The Copenhagen Trial Unit, Center for Clinical Intervention Research, The Capital Region, Copenhagen University Hospital) software, focusing on pain on injection. We defined a type 1 error rate of 5% and aimed to achieve 80% statistical power. TSA showed that the Z-curve did not cross the monitoring boundary or RIS. The calculated RIS for this outcome was 659, while the included trials only comprised 151 patients. These findings suggest a lack of power to confidently exclude the possibility of a type 2 error. Further large, high-quality, randomized trials are needed to evaluate the efficacy of WALANT.