The hypothesis tested is that shifts in pH, induced when a cardioplegic solution is oxygenated, can be detrimental. We added either 100% nitrogen, 95% nitrogen and 5% carbon dioxide, 100% oxygen, or 95% oxygen and 5% carbon dioxide to the cardioplegic solution (St. Thomas' Hospital No. 2 plus glucose 11 mmol/L), and determined postischemic recovery of isolated rat hearts after 3 hours of 10 degrees C cardioplegic protected ischemia. Hearts were arrested and reinfused every 30 minutes throughout the ischemic period with cardioplegic solution. When 5% carbon dioxide was added to nitrogen, the pH of the cardioplegic solution decreased from 9.1 (100% nitrogen) to 7.0 (95% nitrogen: 5% carbon dioxide), a change associated with improved postischemic functional recovery. Aortic output improved from 52.3% +/- 2.7% to 63.9% +/- 2.8%, p less than 0.05, and cardiac output from 60.8% +/- 3.6% to 75.4% +/- 3.3%, p less than 0.01. This improvement was associated with diminished efflux of lactate during ischemia but increased postischemic release of lactate dehydrogenase. When nitrogen was replaced with oxygen, the addition of 5% carbon dioxide resulted in a similar decrease of pH, which again was associated with improved postischemic functional recovery. Aortic output improved from 66.3% +/- 2.8% (100% oxygen) to 88.9% +/- 3.7% (95% oxygen: 5% carbon dioxide), p less than 0.005, and cardiac output from 75.3% +/- 4.1% to 88.9% +/- 2.4%, p less than 0.01. The efflux of lactate during ischemia and the postischemic release of lactate dehydrogenase were similar in both groups. Furthermore, provision of additional oxygen with perfluorocarbons in an electrolyte solution identical to the St. Thomas' Hospital plus glucose solution and oxygenated with 95% oxygen: 5% carbon dioxide conferred no extra protection. In conclusion, the St. Thomas' Hospital No. 2 plus glucose cardioplegic solution should be oxygenated but with 95% oxygen: 5% carbon dioxide and not 100% oxygen because of the additive effect of a relatively "acidotic" pH.