Objective: To measure red blood cell 2,3-diphosphoglycerate (RBC 2,3-DPG) concentrations in early critical illness; to investigate factors associated with high or low RBC 2,3-DPG levels; to calculate in vivo P50 in patients with early critical illness; and to explore the relationship between RBC 2,3-DPG and intensive care mortality.
Design: Prospective cohort study.
Setting: General medical-surgical intensive care unit (ICU) of a major Scottish teaching hospital.
Patients: One-hundred eleven critically ill patients during the first 24 hrs in the ICU with no history of chronic hematologic disorders or RBC transfusion within 24 hrs and 34 age- and sex-matched healthy reference subjects.
Measurements and main results: We measured RBC 2,3-DPG concentration, plasma biochemistry values, and arterial blood gas parameters. On average, RBC 2,3-DPG was lower among critically ill patients than controls (mean [sd], 14.1 [6.3] vs. 16.7 [3.7] mumol/g hemoglobin; p = .004) and had a wider range of values (patients, 3.2-32.5 mumol/g hemoglobin; reference group, 9.1-24.3). Regression analysis indicated a strong independent association between plasma pH and RBC 2,3-DPG (B, 32.15 [95% confidence interval, 19.07-46.22], p < .001) and a weak association with plasma chloride (B, -0.196 [95% confidence interval, -0.39 to -0.01], p = .044) but not with hemoglobin or other measured biochemical parameters. The mean calculated in vivo P50 level was normal (3.8 kPa) but varied widely among patients (range, 2.0-5.5 kPa). RBC 2,3-DPG concentration was similar for ICU survivors and nonsurvivors.
Conclusions: RBC 2,3-DPG concentrations vary widely among critically ill patients. Acidosis is associated with lower RBC 2,3-DPG concentrations, but anemia is not associated with a compensatory increase in RBC 2,3-DPG early in critical illness. Lower RBC 2,3-DPG concentrations during the first 24 hrs of intensive care are not associated with higher ICU mortality.