Background: Gastric cancer survival rates in Africa are low as many cases are diagnosed late. Currently, there are no inexpensive, non-invasive and simple techniques that can be employed in poor resource settings for early case detection. In this study, we explored the possibility using blood in gastric juice as a screening tool to identify patients requiring referral for endoscopy.
Methods: The study was conducted at the University Teaching Hospital endoscopy unit in Lusaka, Zambia. During esophagogastroduodenoscopy, gastric juice was aspirated and the pH determined using pH paper test strips. The presence of blood was tested using urinalysis reagent strips.
Results: We enrolled 276 patients; 147(53%) were female and median age was 49 years (IQR 40-64 years). The presence of blood was associated with mucosal lesions, [OR 2.1; 95% CI 1.2-3.7, P = 0.004]. It was also associated with gastric cancer, [OR 6.7; 95% CI 2-35, P = 0.0005], even at 1:10 and 1:100 dilutions, [OR 5.4; 95% CI 2.3-13.8, P<0.0001] and [OR 9.1; 95% CI 3.5-23, P<0.0001] respectively. The sensitivity for gastric cancer detection using blood in gastric juice was 91% and the specificity was 41%. Analysis using the intensity of blood in gastric juice yielded an area under the receiver operating characteristic curve of 0.78; 95% CI 0.71-0.86 with a sensitivity of 79% and a specificity of 77%.
Conclusions: The presence of blood in gastric juice is associated with gastric mucosal lesions. It has a high sensitivity but low specificity for gastric cancer detection.