1
|
Lee Y, Samarasinghe Y, Javidan A, Tahir U, Samarasinghe N, Shargall Y, Finley C, Hanna W, Agzarian J. The fragility of significant results from randomized controlled trials in esophageal surgeries. Esophagus 2023; 20:195-204. [PMID: 36689016 DOI: 10.1007/s10388-023-00985-2] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/08/2022] [Accepted: 01/05/2023] [Indexed: 01/24/2023]
Abstract
While randomized controlled trials (RCTs) are regarded as one of the highest forms of clinical research, the robustness of their P values can be difficult to ascertain. Defined as the minimum number of patients in a study arm that would need to be changed from a non-event to event for the findings to lose significance, the Fragility Index is a method for evaluating results from these trials. This study aims to calculate the Fragility Index for trials evaluating perioperative esophagectomy-related interventions to determine the strength of RCTs in this field. MEDLINE and EMBASE were searched for RCTs related to esophagectomy that reported a significant dichotomous outcome. Two reviewers independently screened articles and performed the data extractions with risk of bias assessment. The Fragility Index was calculated using a two-tailed Fisher's exact test. Bivariate correlation was conducted to evaluate associations between the Fragility Index and study characteristics. 41 RCTs were included, and the median sample size was 80 patients [Interquartile range (IQR) 60-161]. Of the included outcomes, 29 (71%) were primary, and 12 (29%) were secondary. The median Fragility Index was 1 (IQR 1-3), meaning that by changing one patient from a non-event to event, the results would become non-significant. Fragility Index was correlated with P value, number of events, and journal impact factor. The RCTs related to esophagectomy did not prove to be robust, as the significance of their results could be changed by altering the outcome status of a handful of patients in one study arm.
Collapse
Affiliation(s)
- Yung Lee
- Division of General Surgery, McMaster University, Hamilton, ON, Canada
| | - Yasith Samarasinghe
- Division of Thoracic Surgery, Department of Surgery, McMaster University, 50 Charlton Avenue East T-2105, Hamilton, ON, L8N 4A6, Canada
| | - Arshia Javidan
- Division of Vascular Surgery, University of Toronto, Toronto, ON, Canada
| | - Umair Tahir
- Division of Thoracic Surgery, Department of Surgery, McMaster University, 50 Charlton Avenue East T-2105, Hamilton, ON, L8N 4A6, Canada
| | | | - Yaron Shargall
- Division of Thoracic Surgery, Department of Surgery, McMaster University, 50 Charlton Avenue East T-2105, Hamilton, ON, L8N 4A6, Canada
| | - Christian Finley
- Division of Thoracic Surgery, Department of Surgery, McMaster University, 50 Charlton Avenue East T-2105, Hamilton, ON, L8N 4A6, Canada
| | - Wael Hanna
- Division of Thoracic Surgery, Department of Surgery, McMaster University, 50 Charlton Avenue East T-2105, Hamilton, ON, L8N 4A6, Canada
| | - John Agzarian
- Division of Thoracic Surgery, Department of Surgery, McMaster University, 50 Charlton Avenue East T-2105, Hamilton, ON, L8N 4A6, Canada.
| |
Collapse
|
2
|
Yang G, Du W, Zhang X, Huang T, Zhang J, Liu J, Ling Y. The white substance may be a potential endoscopic marker for flat esophageal mucosal neoplastic lesions: A new endoscopic observation. Medicine (Baltimore) 2018; 97:e11885. [PMID: 30170379 PMCID: PMC6392511 DOI: 10.1097/md.0000000000011885] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/26/2022] Open
Abstract
To determine whether the white substance can act as an endoscopic marker for flat esophageal mucosal neoplastic lesions.Esophageal mucosal neoplastic lesions are mainly identified using white light endoscopy, because it is cost-effective; however, this method is limited for detecting early esophageal cancer and precancerous lesions, because these are typically flat mucosal neoplastic lesions. In our experience, a white substance surrounds or covers some flat esophageal mucosal lesions that are eventually diagnosed as neoplastic lesions by biopsy pathology.After retrospective analysis of pathological and clinical data of 20,390 patients, we identified 352 patients with flat esophageal mucosal lesions on endoscopic images. Images were re-evaluated by 2 experienced endoscopists and the prevalence of the white substance recorded. Patients were divided into non-neoplastic and neoplastic groups, based on pathology.The white substance was present in 3.5% (5/144) of non-neoplastic and 14.9% (31/208) of neoplastic cases (P < .05). The diagnostic sensitivity and specificity of the white substance for neoplastic lesions diagnosis were 14.9% and 96.5%, respectively. The presence of white substance was more common in males and in those aged 50 to 79 years. It was more commonly observed in the middle third of the esophagus, and its presence did not correlate with sex, age, or lesion location (P > .05).The white substance, which is easily detected by white light endoscopy, may be an endoscopic marker facilitating detection of flat esophageal mucosal neoplastic lesions, irrespective of sex, age, and lesion location.
Collapse
|