Response to letter regarding article, "association between postresuscitation partial pressure of arterial carbon dioxide and neurological outcome in patients with post-cardiac arrest syndrome".

BACKGROUND Partial pressure of arterial CO2 (Paco(2)) is a regulator of cerebral blood flow after brain injury. Recent guidelines for the management of cardiac arrest recommend maintaining Paco(2) at 40 to 45 mm Hg after successful resuscitation; however, there is a paucity of data on the prevalence of Paco(2) derangements during the post-cardiac arrest period and its association with outcome. METHODS AND RESULTS We analyzed a prospectively compiled and maintained cardiac arrest registry at a single academic medical center. Inclusion criteria are as follows: age ≥18, nontrauma arrest, and comatose after return of spontaneous circulation. We analyzed arterial blood gas data during 0 to 24 hours after the return of spontaneous circulation and determined whether patients had exposure to hypocapnia and hypercapnia (defined as Paco(2) ≤30 mm Hg and Paco(2) ≥50 mm Hg, respectively, based on previous literature). The primary outcome was poor neurological function at hospital discharge, defined as Cerebral Performance Category ≥3. We used multivariable logistic regression, with multiple sensitivity analyses, adjusted for factors known to predict poor outcome, to determine whether post-return of spontaneous circulation hypocapnia and hypercapnia were independent predictors of poor neurological function. Of 193 patients, 52 (27%) had hypocapnia only, 63 (33%) had hypercapnia only, 18 (9%) had both hypocapnia and hypercapnia exposure, and 60 (31%) had no exposure; 74% of patients had poor neurological outcome. Hypocapnia and hypercapnia were independently associated with poor neurological function, odds ratio 2.43 (95% confidence interval, 1.04-5.65) and 2.20 (95% confidence interval, 1.03-4.71), respectively. CONCLUSIONS Hypocapnia and hypercapnia were common after cardiac arrest and were independently associated with poor neurological outcome. These data suggest that Paco(2) derangements could be potentially harmful for patients after resuscitation from cardiac arrest.