Skip to main content
. 2023 Apr 19;49(5):491–504. doi: 10.1007/s00134-023-07012-z

Table 3.

Survival analyses for the association between hypo- and hypercapnia on mortality

Hypocapnia (< 35 mmHg) Hypercapnia (≥ 45 mmHg) Number of ETCO2 measurements/patients/mortality events
Complete case analysis HR 95% CI HR 95% CI
Overall 1.65 1.42–1.92 0.82 0.65–1.04 2505/1342/506
Confirmed TBI 1.64 1.40–1.92 0.89 0.70–1.13 2051/1094/462
Isolated TBI 1.77 1.43–2.19 0.93 0.67–1.28 1069/567/246
Patients with symptoms of cerebral herniation
Absence of PEARL± 1.44 1.21–1.73 0.82 0.62–1.09 1113/598/317
Signs of elevated ICP on CT$ 1.44 1.22–1.71 0.99 0.76–1.29 871/461/319
Only considering a single CO2 value per patient
First ETCO2 1.71 1.42–2.07 0.81 0.63–1.04 1224/1224/460
Second ETCO2 1.60 1.35–1.91 0.89 0.58–1.37 1281/1281/481
Nadir ETCO2 1.68 1.42–2.00 0.96 0.62–1.49 1342/1342/506
Peak ETCO2 1.58 1.32–1.89 0.83 0.66–1.06 1342/1342/506
After multiple imputation HR 95% CI HR 95% CI NA
Overall 1.63 1.42–1.88 0.78 0.62–0.97
Confirmed TBI 1.58 1.37–1.82 0.85 0.67–1.06
Isolated TBI 1.68 1.38–2.06 0.88 0.64–1.21
Patients with symptoms of cerebral herniation
Absence of PEARL± 1.43 1.22–1.67 0.81 0.63–1.04
Signs of elevated ICP on CT$ 1.40 1.21–1.63 0.95 0.75–1.19
Only considering a single CO2 value per patient
First ETCO2 1.68 1.41–2.01 0.77 0.60–0.98
Second ETCO2 1.59 1.34–1.89 0.84 0.55–1.30
Nadir ETCO2 1.72 1.44–2.06 0.65 0.36–1.17
Peak ETCO2 1.69 1.41–2.02 0.77 0.61–0.96
Confounder adjusted HR 95% CI HR 95% CI NA
Overall 1.28 1.12–1.46 0.90 0.74–1.10
Confirmed TBI 1.27 1.11–1.45 0.92 0.75–1.13
Isolated TBI 1.23 1.01–1.51 0.88 0.67–1.16
Patients with symptoms of cerebral herniation
Absence of PEARL± 1.28 1.10–1.49 0.93 0.73–1.17
Signs of elevated ICP on CT$ 1.19 1.03–1.37 0.94 0.77–1.14
Only considering a single CO2 value per patient
First ETCO2 1.21 1.01–1.47 0.87 0.69–1.09
Second ETCO2 1.34 1.13–1.58 0.94 0.64–1.38
Nadir ETCO2 1.31 1.11–1.56 0.86 0.51–1.46
Peak ETCO2 1.33 1.10–1.61 0.88 0.71–1.09

Cox regression analyses (complete-case analyses as well as after multiple imputation performed on all 1776 patients that had been selected from the BRAIN-PROTECT database) on hypo- and hypercapnia versus normocapnia for the overall population, as well as in subgroups of patients with confirmed and isolated TBI as well as in patients with signs of cerebral herniation. Additional sensitivity analyses consider only one measurement per patient, i.e., either the first or second measurement, as well as nadir and peak ETCO2 values per patient. ±At arrival of HEMS; $midline shift > 5 mm or compressed/absent basal cisterns. All analyses including multiple ETCO2 values per patient were adjusted for the measurement time point. Confounder adjusted analyses adjust for age (spline variable), sex, ASA score, systolic blood pressure (spline variable), heart rate (spline variable), oxygen saturation (spline variable), Injury Severity Score (spline variable), first Glasgow Coma Scale score, HEMS service involved in the treatment and distance between incident scene and trauma hospital (spline variable) after multiple imputation of missing variables. In all Cox regression models, mortality up to 1 year after the trauma was modeled