Endoscopy 2021; 53(01): 102
DOI: 10.1055/a-1243-0569
Letter to the editor

Reply to Romero-Castro et al.

Carlos Robles-Medranda
Endoscopy Division, Instituto Ecuatoriano de Enfermedades Digestivas, Guayaquil, Ecuador
,
Roberto Oleas
Endoscopy Division, Instituto Ecuatoriano de Enfermedades Digestivas, Guayaquil, Ecuador
,
Manuel Valero
Endoscopy Division, Instituto Ecuatoriano de Enfermedades Digestivas, Guayaquil, Ecuador
,
Miguel Puga-Tejada
Endoscopy Division, Instituto Ecuatoriano de Enfermedades Digestivas, Guayaquil, Ecuador
,
Jorge Baquerizo-Burgos
Endoscopy Division, Instituto Ecuatoriano de Enfermedades Digestivas, Guayaquil, Ecuador
,
Jesenia Ospina
Endoscopy Division, Instituto Ecuatoriano de Enfermedades Digestivas, Guayaquil, Ecuador
,
Hannah Pitanga-Lukashok
Endoscopy Division, Instituto Ecuatoriano de Enfermedades Digestivas, Guayaquil, Ecuador
› Author Affiliations

We read with great interest the letter by Romero-Castro et al. Our recent randomized trial demonstrated that endoscopic ultrasound (EUS)-guided combined coiling and cyanoacrylate injection (CYA) is superior to coiling alone for the management of gastric varices [1]. We estimated a one-sided sample size of 30 participants for each group based on the obliteration rate of 82 % for coiling alone and 53 % for CYA injection that was previously described in a two-arm study [2].

Bhat et al. retrospectively analyzed a one-arm cohort treated with EUS-guided combined therapy, reporting a 93 % obliteration rate [3]. If we consider the sample size proposed by Romero-Castro et al., our study should include at least 34 participants in each group for a 75 % statistical power. However, we did not consider this appropriate as the obliteration rate for EUS-guided combined therapy would be driven from a single-arm study, and combining two independent proportions from two different studies with different methodologies may be controversial, for example with the inclusion of patients with noncirrhotic variceal bleeding or patients treated with two different types, sizes, and diameters of coils (mean 5.8 coils) [3].

Regarding the number of coils, there are several factors, such as the gastric variceal diameter, and the type, length, and diameter of the coil itself, that can influence the number of coils required to achieve obliteration. Our study findings should be interpreted based on the data rather than the “possibilities,” as patients within our EUS-guided coiling cohort were treated with a significantly higher median number of coils (P = 0.006). Additionally, we reported a nonsignificant difference in obliteration rates between the therapies (100 % vs. 90 %; P = 0.12), contradicting the assertion of “undertreatment.” Nevertheless, we agree with Romero-Castro et al. that sample size estimation in a boarded context is challenging, with larger multicenter trials awaiting on the horizon before the most effective technique can be defined.



Publication History

Publication Date:
17 December 2020 (online)

© 2021. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany