Anne May1,2, Chris Humston3, Julie Rice4, Christopher J Nemastil1, Ann Salvator1, Joseph Tobias4,5. 1. Department of Pulmonary Medicine, Nationwide Children's Hospital, Columbus, OH, USA. 2. Department of Pediatrics, The Ohio State University College of Medicine, Columbus, OH, USA. 3. Department of Anesthesiology and Pain Medicine, Nationwide Children's Hospital, 700 Children's Drive, Columbus, OH, 43205, USA. christopher.humston@nationwidechildrens.org. 4. Department of Anesthesiology and Pain Medicine, Nationwide Children's Hospital, 700 Children's Drive, Columbus, OH, 43205, USA. 5. Department of Anesthesiology and Pain Medicine, The Ohio State University College of Medicine, Columbus, OH, USA.
Abstract
INTRODUCTION: The gold standard for measuring the partial pressure of carbon dioxide remains arterial blood gas (ABG) analysis. For patients with cystic fibrosis undergoing general anesthesia or polysomnography studies, continuous non-invasive carbon dioxide monitoring may be required. The current study compares end-tidal (ETCO2), transcutaneous (TCCO2), and capillary blood gas carbon dioxide (Cap-CO2) monitoring with the partial pressure of carbon dioxide (PaCO2) from an ABG in patients with cystic fibrosis. METHODS: Intraoperatively, a single CO2 value was simultaneously obtained using ABG (PaCO2), capillary (Cap-CO2), TCCO2, and ETCO2 techniques. Tests for correlation (Pearson's coefficient) and agreement (Bland-Altman analysis) were performed. Data were further stratified into two subgroups based on body mass index (BMI) and percent predicted forced expiratory volume in 1 s (FEV1%). Additionally, the absolute difference in the TCCO2, ETCO2, and Cap-CO2 values versus PaCO2 was calculated. The mean ± SD differences were compared using a paired t test while the number of times the values were ≤ 3 mmHg and ≤ 5 mmHg from the PaCO2 were compared using a Fishers' exact test. RESULTS: The study cohort included 47 patients (22 males, 47%) with a mean age of 13.4 ± 7.8 years, median (IQR) BMI of 18.7 kg/m2 (16.7, 21.4), and mean FEV1% of 87.3 ± 18.3%. Bias (SD) was 4.8 (5.7) mmHg with Cap-CO2 monitoring, 7.3 (9.7) mmHg with TCCO2 monitoring, and 9.7 (7.7) mmHg with ETCO2 monitoring. Although there was no difference between the degree of bias in the population as a whole, when divided based on FEV1% and BMI, there was greater bias with ETCO2 in patients with a lower FEV1% and a higher BMI. The Cap-CO2 vs. PaCO2 difference was 5.2 ± 5.3 mmHg (SD), with 16 (48%) ≤ 3 mmHg and 20 (61%) ≤ 5 mmHg from the ABG value. The TCCO2-PaCO2 difference was 9.1 ± 7.2 mmHg (SD), with 11 (27%) ≤ 3 mmHg and 15 (37%) ≤ 5 mmHg from the ABG value. The ETCO2-PaCO2 mean difference was 11.2 ± 7.9 mmHg (SD), with 5 (12%) ≤ 3 mmHg and 11 (26%) ≤ 5 mmHg from the ABG value. CONCLUSIONS: While Cap-CO2 most accurately reflects PaCO2 as measured on ABG, of the non-invasive continuous monitors, TCCO2 was a more accurate and reliable measure of PaCO2 than ETCO2, especially in patients with worsening pulmonary function (FEV1% ≤ 81%) and/or a higher BMI (≥ 18.7 kg/m2).
INTRODUCTION: The gold standard for measuring the partial pressure of carbon dioxide remains arterial blood gas (ABG) analysis. For patients with cystic fibrosis undergoing general anesthesia or polysomnography studies, continuous non-invasive carbon dioxide monitoring may be required. The current study compares end-tidal (ETCO2), transcutaneous (TCCO2), and capillary blood gascarbon dioxide (Cap-CO2) monitoring with the partial pressure of carbon dioxide (PaCO2) from an ABG in patients with cystic fibrosis. METHODS: Intraoperatively, a single CO2 value was simultaneously obtained using ABG (PaCO2), capillary (Cap-CO2), TCCO2, and ETCO2 techniques. Tests for correlation (Pearson's coefficient) and agreement (Bland-Altman analysis) were performed. Data were further stratified into two subgroups based on body mass index (BMI) and percent predicted forced expiratory volume in 1 s (FEV1%). Additionally, the absolute difference in the TCCO2, ETCO2, and Cap-CO2 values versus PaCO2 was calculated. The mean ± SD differences were compared using a paired t test while the number of times the values were ≤ 3 mmHg and ≤ 5 mmHg from the PaCO2 were compared using a Fishers' exact test. RESULTS: The study cohort included 47 patients (22 males, 47%) with a mean age of 13.4 ± 7.8 years, median (IQR) BMI of 18.7 kg/m2 (16.7, 21.4), and mean FEV1% of 87.3 ± 18.3%. Bias (SD) was 4.8 (5.7) mmHg with Cap-CO2 monitoring, 7.3 (9.7) mmHg with TCCO2 monitoring, and 9.7 (7.7) mmHg with ETCO2 monitoring. Although there was no difference between the degree of bias in the population as a whole, when divided based on FEV1% and BMI, there was greater bias with ETCO2 in patients with a lower FEV1% and a higher BMI. The Cap-CO2 vs. PaCO2 difference was 5.2 ± 5.3 mmHg (SD), with 16 (48%) ≤ 3 mmHg and 20 (61%) ≤ 5 mmHg from the ABG value. The TCCO2-PaCO2 difference was 9.1 ± 7.2 mmHg (SD), with 11 (27%) ≤ 3 mmHg and 15 (37%) ≤ 5 mmHg from the ABG value. The ETCO2-PaCO2 mean difference was 11.2 ± 7.9 mmHg (SD), with 5 (12%) ≤ 3 mmHg and 11 (26%) ≤ 5 mmHg from the ABG value. CONCLUSIONS: While Cap-CO2 most accurately reflects PaCO2 as measured on ABG, of the non-invasive continuous monitors, TCCO2 was a more accurate and reliable measure of PaCO2 than ETCO2, especially in patients with worsening pulmonary function (FEV1% ≤ 81%) and/or a higher BMI (≥ 18.7 kg/m2).
Authors: J Griffin; B E Terry; R K Burton; T L Ray; B P Keller; A L Landrum; J O Johnson; J D Tobias Journal: Br J Anaesth Date: 2003-10 Impact factor: 9.166