Longer withdrawal time is a validated quality indicator in colonoscopy and is strongly associated with improved adenoma detection. In contrast, no standardized benchmark exists for upper gastrointestinal endoscopy (EGD) despite increasing evidence, primarily from Asian cohorts, that prolonged inspection may improve detection of early neoplastic lesions. North American data on this topic are scarce, and practice patterns vary widely, with no consensus regarding optimal examination duration. We aimed to evaluate the association between total EGD examination time and the detection of clinically significant lesions in a large, prospectively documented North American population. We hypothesized that longer total examination time would be independently associated with higher detection yield across multiple lesion categories.

We performed a secondary analysis of an ongoing prospective observational study conducted at the Centre hospitalier de l’Université de Montréal (CHUM), a tertiary academic center, between November 2023 and August 2025. Diagnostic, surveillance, and screening EGDs were included. Procedures with pre-planned therapeutic interventions such as endoscopic mucosal resection, known pre-existing pathology requiring targeted examination (e.g., Barrett’s esophagus), or incomplete time documentation were excluded.

Total EGD examination time was defined as the interval from endoscope insertion to complete withdrawal. Recorded times included inspection and routine biopsy sampling. For each procedure, detection of abnormalities was classified into six prespecified categories, the first being the primary outcome: (1) significant visual abnormality, (2) significant biopsy/polyp abnormality, (3) any significant abnormality, (4) any visual abnormality, (5) any biopsy/polyp abnormality, and (6) any abnormality of any type. Clinically significant abnormalities were defined based on consensus criteria established prior to data extraction.

Associations between total examination time and detection outcomes were modeled using logistic regression with generalized estimating equations (GEE) to account for clustering by individual endoscopist. Time was analyzed as a continuous predictor.

A total of 1,169 EGDs met inclusion criteria. The mean patient age was 55.9 years, and 57.5% were female. The mean total examination time was 5.6 ± 3.7 minutes. Detection rates for all abnormality categories increased with longer examination time. For the primary outcome, each additional minute was associated with a 10.9% increase in the odds of identifying at least one significant visual abnormality (OR 1.109; p < 0.001). Similar associations were observed across secondary outcomes, with odds ratios ranging from 1.13 to 1.25 per additional minute (all p < 0.001).

Modeled detection curves demonstrated a steady increase in probability of abnormality detection as inspection time lengthened, with the most pronounced gains occurring between 4 and 12 minutes. Detection appeared to plateau at approximately 13–15 minutes, suggesting diminishing marginal benefit beyond this duration.

In this large North American cohort, longer total EGD examination time was strongly and consistently associated with higher detection of significant mucosal abnormalities. The relationship persisted across multiple lesion categories and demonstrated a reproducible plateau at approximately 13–15 minutes. These findings support the concept that total EGD examination time represents a meaningful, measurable quality metric analogous to withdrawal time in colonoscopy. Future studies should validate these results in multicenter settings, refine optimal time thresholds, and investigate the role of automated quality-monitoring tools capable of providing real-time feedback on inspection duration.