Partial arterial carbon dioxide and oxygen pressure in patients with cardiogenic shock

In patients with acute cardiovascular diseases, hypocapnia, hypoxia and hyperoxia are known to be associated with increased mortality. This monocentric prospective registry study included 238 consecutive patients with cardiogenic shock (CS). The study aimed to assess the prognostic impact of partial arterial carbon dioxide (PaCO₂) and oxygen pressure (PaO₂) on 30-day all-cause mortality. Statistical analyses included t-tests, Spearman´s correlation, Kaplan-Meier and Cox regression analyses. No difference was found between quartiles of PaCO₂ (log-rank p = 0.416) and PaO₂ (log-rank p = 0.946) in the entire cohort. In the subgroup of patients with ventilation on admission, patients with PaCO₂ ≤ 33 mmHg showed the highest 30-day all-cause mortality compared to the other quartiles (82.6% vs. 46.9% vs. 54.0% vs. 59.6% log-rank p = 0.026). No differences were found between levels of PaO₂, when stratified by quartiles (log-rank p = 0.895). After differentiation between patients with PaCO₂ ≤ 33 mmHg and PaCO₂ > 33 mmHg the association with 30-day all-cause mortality remained significant (82.6% vs. 54.5% log-rank p = 0.006) in ventilated patients, whereas still no difference could be seen in the entire cohort (log-rank p = 0.264). Even after multivariable adjustment PaCO₂ ≤ 33 mmHg remained an independent risk factor for 30-day all-cause mortality (HR 1.936; 95% CI 1.131-3.316; p = 0.016) in ventilated CS-patients. In conclusion, no association was found between different levels of PaCO₂ and PaO₂ with 30-day all-cause mortality in patients with CS. However, in the subgroup of CS-patients requiring ventilation, PaCO₂ ≤ 33 mmHg was associated with an increased 30-day all-cause mortality.