The significance of blood gas results following cardiac arrest

Those who have been resuscitated from cardiac arrest may benefit from mechanical-ventilation strategies aimed at maintaining pCO₂ above the normal (reference) range, 35-45 mmHg (4.7-6.0 kPa). This is the headline finding of recently published research from the FINNRESCUI study group, a collaboration of Finnish intensive care physicians whose aim is to evaluate treatment modalities for patients who are admitted to intensive care following resuscitation from cardiac arrest. 

Optimum care strategies, including target goals for blood gas results are not well defined for this group of critically ill patients, and both mortality and morbidity remain high; the majority either die in hospital or survive with severe neurological disability because of anoxic brain injury. The FINNRESCUI study is a prospective observational study of all 548 patients admitted to 21 intensive care units across Finland following resuscitation from out-of-hospital cardiac arrest during a one-year period, 2010-2011. 

For this most recently published part of their study, investigators aimed to assess whether arterial blood gas results (specifically pO₂ and pCO₂), obtained during the first 24 hours after admission to intensive care, correlate with neurological outcome, assessed independently by neurologists one year after cardiac arrest. 

This assessment is based on the established Glasgow-Pittsburgh Cerebral Performance Categories (CDC) 1 to 5. CDC 1 and 2 represent good outcome, i.e. ability to work and lead normal life independently (although patients may have some neurological deficit). CDC 3 to 5 represent categories of increasingly poor outcome, ranging from severe neurological disability, precluding an independent life (CDC 3); to coma (vegetative state) (CDC 4) and finally, death (CDC 5).

Of the 548 resuscitated patients in the FINNRESCUI study cohort, 139 were excluded from this analysis because of incomplete or absent arterial blood gas data, and other reasons. Thus a total of 409 patients were included. As part of their normal care/monitoring these 409 patients had blood sampled for blood gases on average eight times during the first 24 hours after admission. For each patient, the mean pCO₂ and pO₂ during the first 24 hours were calculated. 

A third of the 409 patients had mean pCO₂ in the low range, 20-34 mmHg (2.7-4.5 kPa); a third had mean pCO₂ in the intermediate range, 34-38 mmHg (4.5-5.1 kPa); and the final third had mean pCO₂ in the high range, 38-76 mmHg (5.1-10.1 kPa). The equivalent tertile ranges for mean pO₂ were: low, 57-99 mmHg (7.6-13.1 kPa); intermediate, 99-128 mmHg (13.1-17.0 kPa); and high, 128-237 mmHg (17.0-31.5 kPa). 

The proportion of time during the 24-hour period that each patient’s pCO₂ was within each of four predetermined ranges (<30 mmHg (<4.0 kPa); 30-37.5 mmHg (4.0-5.0 kPa); 37.5-45 mmHg (5.0-6.0 kPa); and >45 mmHg (>6.0 kPa)) was also calculated. A similar calculation was made for predetermined pO₂ ranges.

At neurological assessment 1 year post cardiac arrest, 168 (41 %) study patients were judged to have had a good outcome (CDC 1 or 2), and the remaining 241 (59 %) were judged to have had a bad outcome (CDC 3-5). Mean 24-hour pCO₂ value was found to be an independent predictor of outcome; it was found that the longer a patient’s pCO₂ remained in the >45 mmHg pCO₂ range, the greater was the likelihood of a good outcome. No such relationship was found for pO₂.

In summary, investigators found that hypercapnia, pCO₂ >45 mmHg (6.0 kPa) was associated with a good neurological outcome at 12 months for patients resuscitated from cardiac arrest. They found no correlation between pO₂ and outcome. In discussion of their study the authors outline the putative mechanisms by which hypercapnia might benefit patient recovery from cardiac arrest and hypocapnia might threaten recovery.