To explore the predictive value of carbon dioxide offset for monitoring microcirculation after fluid resuscitation in patients with septic shock.

A total of 63 patients with septic shock were treated with fluid resuscitation. The levels of heart rate, mean arterial pressure (MAP), central venous pressure (CVP), cardiac index (CI), arterial blood pH, base excess, arterial oxyhemoglobin saturation (SaO₂), hemoglobin, arterial partial pressure of oxygen (PaO₂), arterial partial pressure of carbon dioxide (PaCO₂), transcutaneous oxygen pressure (TcpO₂), transcutaneous carbon dioxide pressure (TcpCO₂), oxygen offset, carbon dioxide offset and arterial blood lactic acid were detected before and after fluid resuscitation. Meanwhile, all patients were divided into the survival group (45 cases) and death group (18 cases) according to 28 d mortality, and the levels of PaO₂, PaCO₂, TcpO₂, TcpCO₂, oxygen offset, carbon dioxide offset and arterial blood lactic acid were compared at 6 h after fluid resuscitation between the two groups. The receiver operating characteristic (ROC) curve was used to predict carbon dioxide offset for microcirculation changes after fluid resuscitation in patients with septic shock.

After 6 hours of fluid resuscitation, the levels of MAP [(76 ± 13) mmHg vs. (68 ± 12) mmHg], CVP [(12 ± 4) cmH₂O vs. (10 ± 3) cmH₂O], CI [(4.1 ± 1.4) L/min vs. (3.6 ± 1.4) L/min], base excess [(-5 ± 3) mmol/L vs. (-7 ± 4) mmol/L], PaO₂ [(121 ± 16) mmHg vs. (111 ± 15) mmHg] and TcpO₂ [(77 ± 11) mmHg vs. (62 ± 12) mmHg] were significantly higher than those before fluid resuscitation (t = 3.701, 2.552, 2.071, 4.459, 3.647, 7.084; all P < 0.05), while PaCO₂ [(37 ± 6) mmHg vs. (39 ± 6) mmHg], TcpCO₂ [(42 ± 12) mmHg vs. (49 ± 13) mmHg], oxygen offset [(0.36 ± 0.16)% vs. (0.44 ± 0.13)%], carbon dioxide offset [(0.14 ± 0.12)% vs. (0.26 ± 0.19)%] and arterial blood lactic acid [(2.4 ± 0.7) mmol/L vs. (3.6 ± 0.6) mmol/L] decreased significantly (t = 2.171, 3.103, 3.080, 4.238, 10.030; all P < 0.05). In the survival group, the levels of PaCO₂ [(35 ± 6) mmHg vs. (40 ± 5) mmHg], TcpCO₂ [(41 ± 10) mmHg vs. (53 ± 11) mmHg], carbon dioxide offset [(0.15 ± 0.19)% vs. (0.34 ± 0.15)%] and arterial blood lactic acid [(2.2 ± 0.7) mmol/L vs. (4.6 ± 0.8) mmol/L] were lower (t = 2.572, 4.289, 3.790, 11.740; all P < 0.05), while the levels of PaO₂ [(111 ± 19) mmHg vs. (95 ± 20) mmHg] and TcpO₂ [(76 ± 9) mmHg vs. (60 ± 9) mmHg] were higher than those in the death group after 6 hours of fluid resuscitation (t = 2.893, 5.935; both P<0.05). The ROC curve showed that the area under the curve of carbon dioxide offset was 0.883, and the 95% confidence interval was 0.852 - 0.982.

The carbon dioxide offset can dynamically predict microcirculation perfusion after fluid resuscitation in patients with septic shock and can be used as a reliable parameter for prognosis.