Studies on the impact of onsite evaluation must account for the per‐pass success rate
Robert L. Schmidt · Digestive Endoscopy · 2013
To the Editor, I wish to comment on the study by Suzuki et al. regarding the optimal number of needle passes for endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) for solid pancreatic lesions.1 The data provided by Suzuki (table 1) are the first empirical data showing the impact of needle passes on the per-case success rate. Their data with 22-G needles conform to a binomial sampling distribution with a per-pass sampling success rate of 0.53 (, P = 1.00). Suzuki et al. found that there was no significant difference in the per-case adequacy rate achieved with a fixed sampling policy using four needle passes compared to sampling with onsite evaluation. This result is not surprising because the per-pass success rate was relatively high (P = 0.53). Recent studies have shown that the impact of onsite evaluation is context dependent. Mathematical models based on a binomial sampling model suggest that onsite evaluation improves performance when the per-pass success rate is low, but provides little benefit when the per-pass success rate is high.2 The impact of the per-pass success rate has been empirically demonstrated in a variety of tissues, including solid pancreatic lesions.3, 4 Thus, the theoretical and empirical results suggest that a binomial model is a good model for EUS-FNA adequacy in solid pancreatic lesions. Given the per-pass success rate at their institution, a fixed sampling policy of four passes would be expected to be equivalent to sampling with onsite evaluation; however, this would not be the case at locations where the per-pass sampling rate is lower.2, 5 The per-pass adequacy rate is a key determinate of the per-case adequacy rate and studies on the effect of onsite evaluation must account for this factor.