Oxygen saturation / fraction of inspired oxygen as a predictor of mortality in patients with exacerbation of COPD treated at the Central Military Hospita

Abstract

"Clinical Research Work - Residents" award at the Xth SOLAMI / XXVI ACMI-ACP Congress, Cartagena, 17th "" 20th of August of 2017.

Introduction: the validity of the ratio arterial oxygen saturation and inspiratory oxygen fraction (SaO /FiO2), calculated by pulse oximetry and by arterial gases in patients with exacerbation of chronic obstructive pulmonary disease (E-COPD) at the altitude of Bogotá are not known. Patients with COPD can present alterations in gas exchange that can worsen with bronchospasm episodes. Obtaining values of SaO₂ by oximetry and FiO₂ can provide valuable information on the course of the exacerbation.

Objective: to determine the validity of the SaO₂ / FiO₂ ratio calculated by pulse oximetry and by arterial gases in relation to mechanical ventilation (VM) outcomes and mortality at seven and 30 days.

Methods: a prospective cohort study with diagnostic test analysis was performed, calculating DECAF, BAP-65, CURB-65, arterial blood gas and pulse oximetry scores on admission of patients with E-COPD, and the outcome of mortality was evaluated at seven and 30 days of admission and the MV requirement during hospitalization; the SaO₂ /FiO₂ ratio was calculated using the SaO₂ obtained in the arterial gases and independently the SaO₂ /FiO₂ ratio with the SaO₂ obtained by pulse oximetry. With the data obtained, the values of sensitivity (S), specificity (E), positive predictive value (VPP), negative predictive value (NPV), positive likelihood ratio (LR +), negative likelihood ratio (LR-) and low area the operating characteristics curve of the receiver (ACOR) were calculated.

Results: 462 E-COPD were analyzed; the MV requirement was 14.3% and 30-day mortality of 5.71%, the sensitivity of the SaO₂ /FiO₂ ratio calculated by pulse oximetry for MV outcome was 84.6% (95% CI : 75-94.2), specificity 42% (95% CI: 37-47), PPV: 19.6% (95% CI: 14.8-24.4), NPV: 94.3% (95% CI: 90.5-98), LR +: 1.46 (95% CI : 1.28-1.67), LR: 0.37 (95% CI: 0.20-1.67), ACOR: 0.779% (95% CI: 0.711-0.847) p <0.0001, sensitivity of SaO₂ /FiO₂ ratio for arterial blood gases for MV was 83% (95% CI: 73.2-92.9), specificity 57% (95% CI: 51.9-62.2), PPV: 24.8% (95% CI: 18.8-30.7), NPV: 95.2% (95% CI: 92.2-98.2), LR +: 1.94 (IC95%: 1.65-2.27), LR-: 0.30 (IC95%: 0.17-0.51), ACOR: 0.799% (IC95%: 0.737-0.861) p <0.0001, the sensitivity of the SaO₂ /FiO₂ ratio by oximetry for outcome of mortality has a sensitivity of 76.8% (95% CI: 58.8- 95), specificity of 39.2% (95% CI: 34.4-43.9), PPV: 7.1% (95% CI: 3.9-10.3), NPV: 96.5% (95% CI : 93.5-99.5), LR +: 1.26 (IC95%: 1.01-1.58), LR-: 0.59 (CI 95%: 0.29-1.20), ACOR: 0.689% (IC95%: 0.568-0.810) p <0.0001, the sensitivity of the SaO₂ /FiO₂ ratio for arterial blood gases was 80.8% (95% CI: 63.7-97.8), specificity 53.2% (95% CI: 48.3-58), PPV: 9.6% (95% CI: 5.5-13.8), NPV: 97.8% (95% CI: 95.7-99.9), LR +: 1.73 (95% CI: 1.39-2.14), LR-: 0.36 (IC95%: 0.16-0.80), ACOR: 0.732% (IC95%: 0.617-0.846) p <0.0001.

Conclusion: SaO₂ /FiO₂ values calculated by oximetry and arterial blood gases can be useful to predict MV outcomes and mortality in patients with exacerbation of COPD