Analysis of Histopathological Lesions and Their Correlation With Upper Gastrointestinal Endoscopic Findings: A Retrospective Study at a District Hospital in Ghana
Prosper Adjei, Samuel Kyeremeh AdjeiABSTRACT
Background and Aims
Although upper gastrointestinal (GI) endoscopy is a crucial diagnostic tool for evaluating lesions of the upper GI tract, its findings are not always definitive, necessitating histopathological confirmation. This retrospective study aimed to assess the spectrum of histopathological lesions and their correlation with endoscopic impressions at Methodist Hospital, Wenchi, Ghana.
Methods
A retrospective observational study was conducted on patients who underwent upper GI endoscopy with biopsy between August 2024 and September 2025. Out of 513 patients who underwent upper GI endoscopy, 176 were biopsied, but 46 were excluded due to missing histopathology reports, leaving 130 patients for analysis. Endoscopic impressions were compared with histopathological diagnoses across different GI tract sites (esophagus, stomach, and duodenum). Diagnostic accuracy was calculated with sensitivity, specificity, and 95% confidence intervals (CIs). Agreement was assessed using Cohen's Kappa and prevalence‐adjusted bias‐adjusted Kappa (PABAK).
Results
Of the 130 patients, 97.7% of lesions were benign (predominantly chronic gastritis), and 2.3% were malignant. Endoscopy demonstrated a sensitivity of 100% (95% CI: 29.2–100) and specificity of 99.2% (95% CI: 95.8–100) for malignant lesions. Cohen's Kappa was 0.796, indicating substantial agreement, while PABAK was 0.966, reflecting near‐perfect concordance after adjusting for prevalence.
Conclusion
Upper GI endoscopy showed high diagnostic accuracy and substantial concordance with histopathology for benign lesions such as gastritis. These findings suggest that clinicians in resource‐deprived areas where histopathology is not readily available or may be associated with long turnaround times can safely commence targeted treatments for benign upper GI conditions based primarily on endoscopic findings.