Appilication Of Partial Pressure Of End-Tidal Carbon Dioxide In Patients With Acute Exacerbation Of Chronic Obstructive Pulmonary Disease

Objective To explore the correlation and concordance between partial pressure of end-tidal carbon dioxide( PETCO2) and partial pressure of Arterial Carbon Dioxide(Pa CO2) in patients with acute exacerbation of chronic obstructive pulmonary disease(AECOPD) in various degrees, and analyse the potential pathological and physiological mechanism and influence factors of them. Further evaluating the appilication value of PETCO2, which is noninvasive.Methods It was a retrospective self controlled study included the patients who were in AECOPD accepting inpatient treatments in department of Respiratory of Zhong Shan hospital Xiamen university from January 2015 to December 2015. According to the FEV1(%pred)from the results of pulmonary function test, completing the GOLD classification,and all the patients were classified into four groups,including group I,group II,group III,group IV. Carbon dioxide trend chart was dynamically monitored about five to ten minutes by sidestream capnometry, then booking in the mean of PETCO2. At the same time, the radial arterial blood in left forearm was sampled to begin arterial blood gas analysis and detection.Finally, we got Pa CO2(which all were not corrected by temperration). To observe the difference of age, height, weight, smoking index, the time of COPD, pulmonary ventilation function index: FEV1(% pred) and FEV1/FVC(%), small airway function index: MMEF75/25(%pred), airway resistance index: s Raw(%pred), volume function index: RV/TLC(%), pulmonary diffusing capacity : TLCO SB. The spearman's correlation analysis and paired t-test were used to determine the relationship between PETCO2 and Pa CO2.Next, the concordance and bias between PETCO2 and Pa CO2 were analysed by using the method of Bland-Altman.Finally,appropriate statistical methods were used to compare the difference in each group between Pa-ETCO2(Pa CO2-PETCO2)and every index.Multiple linear regression analysis were used for these statistics.Results 1.A total of 103 patients including 97 males and 6 females were enrolled in our study during this period; the mean age was 48.49±11.64 years; the mean Pa CO2 was 44.80±11.40 mm Hg, the mean PETCO2 was 33.98±9.56 mm Hg. Compared with every group, height, weight, smoking index, the time of COPD did not show significant difference(P>0.05). Age and pulmonary function indexs including like FEV1(%pred), FEV1/FVC(%), MMEF75/25(%pred), s Raw(% pred), RV/TLC(%) and TLCO SB showed significant difference(P<0.05).2. There was significant difference between Pa CO2 and PETCO2 in four groups through paired t-test(P<0.05), we could consider that PETCO2 was different with Pa CO2. Besides, most of Pa CO2 was higher than PETCO2. Pa-ETCO2 significantly increased with the aggravation of AECOPD. PETCO2 was significantly positive correlated with Pa CO2 in four groups, and the four correlation coefficients were moderate(r of 4 groups were nearly 0.5). In addition, the correlation between PETCO2 and Pa CO2 became worse along with the improvement of GOLD classification(r=0. 489,0. 459,0.449,0.439, all P <0. 05).3. Generally, the Bland-Altman figure showed that Pa-ETCO2( the mean bias) was 10.82±9.84 mm Hg,(95% limits of consistency,-8.5 to 30.1 mm Hg), and 4.85%(5/103)of the points were beyond the 95% confidence interval. The points in the limitation were grouped, however, the restults' difference of 95% confidence interval through two measurement techniques was not accepted in actualiy clinical job, which meaned that the concordance between PETCO2 and Pa CO2 was not high. Moreover, there were poor consistency between Pet CO2 and Pa CO2 in different patients with COPD.4. The time of COPD( r=0.203),s Raw(%pred)(r= 0.369) and RV/TLC(%)(r= 0.436)were significantly positive correlated with Pa-ETCO2(P<0.05); weight(r=-0.224),FEV1(%pred)(r=-0.463),FEV1/FVC(%)(r=-0.353),MMEF75/25(%pred)(r=-0.496)and TLCO SB(r=-0.227)were significantly negative correlated with Pa-ETCO2( P<0.05). Multiple linear regression analysis showed that MMEF75/25(%pred)(adjusted r2 =0.207, ?=-0.583, P=0.019) was the major dependent contributing factors for elevated Pa-ETCO2.Conclusion There were moderate correlation and poor consistency between PETCO2 which was measured by sidestream capnometry and Pa CO2 in different patients with AECOPD who were varying widely in severity. PETCO2 cannot be applied directly to judge the varity of Pa CO2 in patients with AECOPD, whose appilication value was limited. Pa-ETCO2 could be used to reflect the severity of patients with AECOPD. The function of small tract was a important influence for the accuracy of PETCO2.