Introduction: Postcolonoscopy colorectal cancer (PCCRC) due to low adenoma detection rate (ADR) prompted search for methods (e.g., cap, endocuff) to increase ADR. Despite human limitations, water exchange (WE) increases ADR vs. gas insufflation (GIE 2021;93:1411). Artificial intelligence (AI) overcomes human limitations to increase ADR (Expert Rev Gastroenterol Hepatol 2019;13:1153). Both WE and AI were amongst the top 10 advances in endoscopy (2017-2021) (GIE 2018;88:1; 2019;90:35; 2020;92:241; 2021;94:441). AI-assisted ADR is hampered by false positives (e.g., bubbles, debris, mucus) (Diagnostics [Basel] 2021;11:1113). WE insertion salvage cleaning improves cleanliness (GIE 2021;93:1411) and reduces false positives (Expert Rev Gastroenterol Hepatol 2019;13:1153). Thus, the strengths of AI and WE complement the weaknesses of each other (GIE 2022;95:1198). However, WE also increases left colon mucus production (Endoscopy 2020;52:1118), a potential source of false positives (GIE 2020;92:900) that can diminish the benefit of WE. We test the hypothesis that saline produces a dose-related inhibition of left colon mucus production. Methods: Patients were randomly assigned to carbon dioxide (CO2) insufflation, WE with water, 25% saline, or 50% saline. The primary outcome was the mean left colon mucus scale (LCMS) score obtained by two blinded observers: score 0: no visible mucus; to score 4: opaque mucus in thicker clumps covering views of the lumen (Figure). Results: Among 296 patients, baseline demographics, indications and procedural data were similar. The LCMS score for WE with water was higher than those for WE with saline and CO2 insufflation (mean scores were 1.4 [WE water] vs. 0.7 [WE 25% saline] vs. 0.5 [WE 50% saline] vs. 0.2 [CO2]; P < 0.0001), with a kappa value of 0.636 for the interobserver agreement (P < 0.0001). More patients who underwent WE with water required additional cleansing for mucus than those who underwent WE with 50% saline (6.7% vs. 0%, P = 0.032). Water filling was the only predictor of moderate mucus production (odds ratio, 33.3; 95% confidence interval, 7.2-153.2). Conclusion: Since mucus is defined as false positive in AI studies, the reduction of WE-induced mucus production by saline opens new avenue of research into the complementary relationship between AI and WE. WE outperforms cap and endocuff (JGH 2021;36:3268; DDAS 2021;66:1175). WE with 50% saline plus AI will be the optimal approach (enhanced ADR) to prevent PCCRC.Figure 1.: Left Colon Mucus Scale score: (score 0) No visible mucus; (score 1) Minimal amounts of clear mucus in thin streaks or strands; (score 2) Mild opaque mucus in thin strands; (score 3) Moderate opaque mucus in thicker clumps covering one side of the surface; (score 4) More opaque mucus in thicker clumps covering more views of the lumen.