The Comparison Of The Accuracy Of Transcutaneous And End-tidal CO₂ Monitoring In COPD Patients

Background:Chronic Obstructive Pulmonary Disease（COPD）is a common chronic respiratory disease that seriously endangers human health and its morbidity and mortality rate is high which resulting a heavy social and economic burden.The main feature of COPD is persistent respiratory symptoms and airflow limitation.The lungs of end-stage COPD patients are always in over-inflation state for a long time,which usually lead to respiratory muscle fatigue and ventilation dysfunction and carbon dioxide（CO₂）retention in severe patients which could induce poor prognosis.Non-invasive positive pressure ventilation（NPPV）can effectively relieve the respiratory muscle fatigue and improve CO₂ retention in COPD patients which is a commonly used assistant ventilation method in hospitals and families.Arterial partial pressure of carbon dioxide（PaCO₂）is an important indicator for evaluating the severity of CO₂ retention and the therapeutic effect of NPPV in COPD patients.At present,PaCO₂ is usually directly measured by arterial blood gas analysis which is invasive and cannot be monitored continuously.Repeated arteriopuncture may also increase the pain of patients and the infection risk.The non-invasive CO₂ monitoring methods include end-tidal carbon dioxide partial pressure（PetCO₂）and transcutaneous carbon dioxide partial pressure（PtcCO₂）.They are both non-invasive and can be monitored continuously,and already have been the research focus in recent years,but their accuracy has always in dispute.The PtcCO₂ monitor heats the local skin through the electrode and arterializes the local capillaries,accelerates the diffusion of CO₂,and then measures CO₂ through the electrochemical sensor on the skin surface.PtcCO₂ is often used in the monitoring in infants or during sleep and mechanical ventilation which can accurately predict PaCO₂.However,it is susceptible to hemodynamic status,the position and temperature of the sensor,coupled with the high cost of the instruments and consumables.It has not been widely used in clinical practice in China at present.The PetCO₂ monitor measures CO₂ concentration in expiratory air through a sensor,which is a non-invasive,simple and fast monitoring method.It is divided into the mainstream and sidestream method according to the position of the sensor in the airflow.The mainstream method measures CO₂ by placing a sensor in the breathing circuit,and the sidestream method only analyzes a small amount of gas by drawing it from the breath circuit into the sensor.PetCO₂ is commonly used in anesthesia,sleep monitoring and other fields,and has been confirmed to have a good correlation and consistency with PaCO₂,which can predict PaCO₂ levels accurately.However,the accuracy between PetCO₂ and PaCO₂ is affected by chronic airway obstruction and mismatched ventilation-perfusion ratio in COPD patients which resulting in significant differences in partial pressure of CO₂in each alveoli,and unequal emptying time of each alveoli.Previous studies have shown that the PetCO₂ measured by conventional methods differs from PaCO₂ in COPD patients,so it is difficult to predict PaCO₂ accurately.Previous researches of our research group show that monitoring PetCO₂ by prolonging exhalation method[PetCO₂（P）]can reduce the difference between PaCO₂ and PaCO₂ and to improve prediction accuracy,because COPD patients have the pathophysiological characteristics of existing different types of mismatched ventilation-perfusion ratio.In recent years,studies comparing of the accuracy of these two non-invasive CO₂partial pressure monitoring methods always focus on neuromuscular diseases,infantile diseases and other fields,and few researches have explored in COPD patients.Therefore,in order to find an accurate and convenient non-invasive CO₂ monitoring method that is suitable for COPD patients,we will carry out the research to compare the accuracy between PetCO₂（P）and PtcCO₂ in COPD patients with different diseases states and different breathing states.Objectives:1.In stable COPD patients and acute exacerbation of COPD（AECOPD）patients,compare the accuracy of the mainstream PetCO₂（P）and PtcCO₂ during spontaneous breath.2.In AECOPD patients combined with type Ⅱ respiratory failure,compare the accuracy of the sidestream PetCO₂（P）and PtcCO₂ during the application of NPPV.Subjects and Methods:In the first part of the experiment,we recruited 48 stable COPD patients and 35AECOPD patients.Measure PetCO₂（P）,PtcCO₂ and PaCO₂ simultaneously under spontaneous breath.Compare the accuracy and analyze the sources of differences in prediction of PaCO₂ by PetCO₂（P）and PtcCO₂ through correlation and consistency test.In the second part of the experiment,we recruited 11 AECOPD patients combined with type Ⅱ respiratory failure.Measure PetCO₂（P）,PtcCO₂ and PaCO₂ simultaneously before applying NPPV and after applying1 hour of NPPV.Compare the accuracy and analyze the sources of differences in prediction of PaCO₂ by PetCO₂（P）and PtcCO₂through correlation and consistency test.Results:1.Part one:We recruited 83 COPD patients totally,including 48 stable COPD patients and 35 AECOPD patients.In stable group,the median PaCO₂ was 42.70（IQR,38.00,49.53）mm Hg,PetCO₂（P）was 40.50（IQR,34.83,44.78）mm Hg,PtcCO₂ was39.85（IQR,36.30,46.40）mm Hg.In acute exacerbation group,the average PaCO₂ was53.07（IQR,43.83,60.43）mm Hg,PetCO₂（P）was 52.00±9.98 mm Hg,PtcCO₂ was51.68±10.93 mm Hg.（1）During spontaneous breath in 48 stable COPD patients,compare the correlation coefficients between PetCO₂（P）and PtcCO₂:the correlation and consistency between PetCO₂（P）,PtcCO₂and PaCO₂ didn’t show obvious differences（r,0.871 vs 0.854;ICC,0.926 vs 0.934）.（2）During spontaneous breath in 35 AECOPD patients,compare the correlation coefficients between PetCO₂（P）and PtcCO₂:the correlation and consistency between PetCO₂（P）,PtcCO₂and PaCO₂ didn’t show obvious differences（r,0.909 vs 0.950;ICC,0.966 vs 0.969）.（3）When measuring mainstream PetCO₂（P）in COPD patients with spontaneous breath,the accuracy is better in AECOPD patients than in stable COPD patients by prolonging the expiration to the end of 5s（r,0.909 vs 0.950;ICC,0.966 vs 0.969）.2.Part two:We recruited 11 AECOPD patients combined with type Ⅱ respiratory failure.Before applying NPPV,the average PaCO₂ was 60.40±7.55mm Hg,PetCO₂（P）was 56.39±5.40mm Hg,PtcCO₂ was58.33±7.83mm Hg.After applying 1 hour of NPPV,the average PaCO₂ was 57.45±6.49mm Hg,PetCO₂（P）was 51.65±5.79mm Hg,PtcCO₂ was55.76±6.64mm Hg（1）Compare the accuracy of PetCO₂（P）and PtcCO₂ before applying NPPV:the accuracy of PtcCO₂is better than PetCO₂（P）whether before the application of NPPV（r,0.950 vs 0.703;ICC,0.974 vs 0.799）or after the application of NPPV（r,0.895 vs 0.660;ICC,0.944 vs 0.792）（2）During applying NPPV,the accuracy of the change of PetCO₂（P）[ΔPetCO₂（P）]and PtcCO₂（ΔPtcCO₂）in predicting the change of PaCO₂:the correlation betweenΔPetCO₂（P）,ΔPtcCO₂andΔPaCO₂ is good（r,0.679 vs 0.783）,but the Lo A is wide[ΔPetCO₂（P）,（-5.48,9.06）mm Hg;ΔPtcCO₂,（-6.50,5.74）mm Hg],soΔPetCO₂（P）、ΔPtcCO₂can only predict the variation trend rather than variation range ofΔPaCO₂.Conclusion:1.In stable COPD patients and AECOPD patients,both PetCO₂（P）and PtcCO₂ can predict PaCO₂ accurately and the accuracy of the two methods is not significantly different during spontaneous breath.2.In COPD patients,when using mainstream PetCO₂（P）,prolonging the expiration to the end of 5s is more accurate in patients with acute exacerbation than in stable patients.3.In AECOPD patients combined with type Ⅱ respiratory failure,during using NPPV,PtcCO₂ can predict PaCO₂ accurately and is more accurate than PetCO₂（P）.But,the change of PtcCO₂ and PetCO₂（P）can only predict the variation trend rather than variation range ofΔPaCO₂.