Abstract
The optimal sampling techniques for EUS-FNA remain unclear and have not been standardized. To improve diagnostic accuracy, suction techniques for EUS-FNA have been developed and are widely used among endoscopists. The aim of this study was to compare wet-suction and dry-suction EUS-FNA techniques for sampling solid lesions. We performed a comprehensive literature search of major databases (from inception to June 2020) to identify prospective studies comparing wet-suction EUS-FNA and dry-suction EUS-FNA. Specimen adequacy, sample contamination, and histologic accuracy were assessed by pooling data using a random-effects model expressed in terms of odds ratio (OR) and 95% confidence interval (CI). Six studies including a total of 418 patients (365 wet suction vs. 377 dry suction) were included in our final analysis. The study included a total of 535 lesions (332 pancreatic lesions and 203 nonpancreatic lesions). The pooled odds of sample adequacy was 3.18 (CI: 1.82-5.54, P = 0.001) comparing wet- and dry-suction cohorts. The pooled odds of blood contamination was 1.18 (CI: 0.75-1.86, P = 0.1). The pooled rate for blood contamination was 58.33% (CI: 53.65%-62.90%) in the wet-suction cohort and 54.60% (CI 49.90%- 59.24%) in the dry-suction cohort (P = 0.256). The pooled odds of histological diagnosis was 3.68 (CI 0.82-16.42, P = 0.1). Very few adverse events were observed and did not have an impact on patient outcomes using either method. EUS-FNA using the wet-suction technique offers higher specimen quality through comparable rates of blood contamination and histological accuracy compared to dry-suction EUS-FNA.