Elevated Venous to Arterial Carbon Dioxide Gap and Anion Gap Are Associated with Poor Outcome in Cardiogenic Shock Requiring Extracorporeal Membrane Oxygenation Support

Optimal management of cardiogenic shock requiring extracorporeal membrane oxygenation (ECMO) is still an evolving area in which assessment and optimization of the microcirculation may be critically important. We hypothesized that the venous arterial carbon dioxide gap (P(v-a)CO2 gap); the ratio of this gap to arterio-venous oxygen content (P(v-a)CO2/C(a-v)O2 ratio) and the anion gap would be early indicators of microcirculatory status and useful parameters for outcome prediction during ECMO support. We retrospectively reviewed 31 cardiogenic shock patients requiring veno-arterial ECMO, calculating P(v-a)CO2 gap and P(v-a)CO2/C(a-v)O2 ratios in the first 36 hours and the final 24 hours of ECMO support. Sixteen patients (52%) survived and 15 (48%) died. After 24 hours of ECMO support, the P(v-a)CO2 gap (4.9 ± 1.5 vs. 6.8 ± 1.9 mm Hg; p = 0.004) and anion gap (5.2 ± 1.8 vs. 8.7 ± 2.7 mmol/L; p < 0.001) were significantly higher in non-survivors. In the final 24 hours of ECMO support, the P(v-a)CO2 gap (3.5 ± 1.6 vs. 10.5 ± 3.2 mm Hg; p < 0.001), P(v-a)CO2/C(a-v)O2 ratio (1.1 ± 0.5 vs. 2.7 ± 1.0; p < 0.001), anion gap (5.1 ± 3.0 vs. 9.3 ± 5.9 mmol/L; p = 0.02), and lactate (median 1.0 [interquartile range {IQR}: 0.7-1.5] vs. 2.8 [IQR: 1.7-7.7] mmol/L; p = <0.001) were all significantly lower in survivors. Increasing P(v-a)CO2 gap and increasing anion gap were significantly associated with increased risk of mortality. Optimum cut-points for prediction of mortality were 6 mm Hg for P(v-a)CO2 gap in combination with an anion gap above 6 mmol/L in the first 24 hours of ECMO in patients with cardiogenic shock requiring ECMO.