Abstract Background Mitochondrial oxidative phosphorylation (OXPHOS) is responsible for meeting the vast majority of cellular energy demands. Ever since the description of aerobic glycolysis by Warburg in 1927, it has been recognised that cancers have dysfunctional metabolism. Clinically, this Warburg effect is the basis of FDG-PET imaging, increased FDG uptake represents increased tumour glycolysis and translates to worse upper GI adenocarcinoma outcomes. The cause of OXPHOS disruption in upper GI adenocarcinoma is poorly understood, its impact on potential therapeutic strategies is yet to be explored. This is particularly pertinent in upper GI adenocarcinomas as despite recent improvements, 5-year survival remains alarmingly low. Method Oesophageal and gastric adenocarcinoma samples collected in the UK from 175 patients (89 oesophageal adenocarcinoma, 86 gastric adenocarcinoma, 131 male, 44 female, median age=65, IQR=12) who underwent radical resection with curative intent between 1995 and 2009. Quantitative immunofluorescence were used to determine expressions of OXPHOS Complex I and IV at a single cell level. Patients were divided into high vs. low expressors of Complex I and IV using K-means clustering to allow for subsequent subgroup analysis. Result Survival analysis demonstrated significantly worse outcomes for low expressors as compared to high expressors of Complex I at 5 years post-op (5-year survival probability low expressors=0.440, 95%CI=0.358-0.539, high expressors=0.629, 95%CI=0.515-0.769, P-value=0.048). Similarly comparison of Complex IV expression also demonstrates disadvantageous survival for Complex IV low expressors (5-year survival probability low expressors=0.432, 95%CI=0.346-0.538, high expressors=0.603, 95%CI=0.499-0.729, P-value=0.042). The low expressors of Complex I and IV encompassed significantly more Stage III disease as per AJCC 7th edition. The low vs. high expression Complex I and IV groups are similarly matched for baseline patient characteristics. Conclusion These results demonstrate reduced capacity to carry out OXPHOS in oesophageal and gastric adenocarcinoma is characteristic of more advanced cancer and results in worse patient outcomes. Further work is needed to determine if OXPHOS status impacts on neo-adjuvant therapy response and how to exploit OXPHOS deficiency therapeutically in oesophageal and gastric adenocarcinomas. The limitations of this preliminary work include small sample size and tissue derived from historic patients over a long period of time, during which treatment and outcomes for upper GI cancer have significantly changed.