Participants characteristics
Descriptive characteristics are summarized in Supplementary eTable 1. There were no statistically significant differences in sex, age, serum creatinine (Scr), uric acid (UA), triglycerides (TG), total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), fasting plasma glucose (FPG), hemoglobin (HB), or body temperature (T) between the two groups (p > 0.05). Before starting heliox or nitrogen-oxygen ventilation, there was also no statistically significant difference in SBP, DBP, mean arterial blood pressure (MABP), CVP, Lac, or SpO2 between the groups (p > 0.05). There were five patients with coronary atherosclerotic heart disease, one patient with chronic pulmonary heart disease, five patients with essential hypertensive heart disease, and 29 patients without basic heart disease in the control group. There were three patients with coronary atherosclerotic heart disease, four patients with chronic pulmonary heart disease, five patients with essential hypertensive heart disease, and 27 patients without basic heart disease in the experimental group. There was no significant difference in the results of basic heart disease between the two groups (Fisher's exact test: p = 0.586). Also, there were three patients with chronic obstructive pulmonary disease, two patients with bronchiectasis, and 35 patients without basic pulmonary disease in the control group. There were eight patients with chronic obstructive pulmonary disease, no patients with bronchiectasis, and 31 patients without basic pulmonary disease in the experimental group. There was no significant difference in the results of basic pulmonary disease between the two groups (Fisher's exact test: p = 0.098).
Results per study group
Effect of different ventilation modes on blood pressure
Based on the results of Mauchly sphericity test (p < 0.05), the data did not conform to sphericity assumptions. Therefore, ANOVA with repeated measures was used, and the Greenhouse-Geisser correction was applied. The results of the ANOVA with repeated measures showed that after correcting for the within-subjects factor (p < 0.05), there was a significant difference in the mean SBP, DBP, and MABP at different time points. Furthermore, the interaction between the ventilation group and treatment time was significant (p < 0.05), indicating that the ventilation group had an effect on the measurements. Specifically, the overall mean of SBP, DBP, and MABP differed significantly between the two ventilation groups (p < 0.05), as shown in Fig 3, Supplementary eTable 3-5. Moreover, in the control group, ventilation time was negatively correlated with SBP (r = -0.264, p = 0.001), DBP (r = -0.188, p = 0.017), and MABP (r = -0.257, p = 0.001). With an increase in controlling factors, a negative correlation between ventilation time and SBP, DBP as well as MABP were still observed (Supplementary eTable 12). But there was no significant change in SBP, DBP and MABP during ventilation in the experimental group (p > 0.05) (Supplementary eTable 2).
Effect of different ventilation modes on cardiac function
After applying the Greenhouse-Geisser correction, the results of the repeated measures ANOVA for CVP showed that the overall means at different times were not equal, with a corrected p-value of < 0.05. Moreover, the interaction between the ventilation group and treatment time was significant, with a corrected p-value of < 0.05. This finding indicates that the ventilation group had an effect on the CVP measurements, and that there was an interaction between the ventilation group and treatment time. The overall mean of CVP was also found to be significantly different between the two ventilation groups, with a p-value of < 0.05 (Fig 4, Supplementary eTable 6). Moreover, in the control group, ventilation time was positively correlated with CVP (r = 0.232, p = 0.003). With an increase in controlling factors, a positive correlation between ventilation time and CVP was still observed (Supplementary eTable 13). In the experimental group, ventilation time was negatively correlated with CVP (r = -0.536, p < 0.001). With an increase in controlling factors, a negative correlation between ventilation time and CVP was still observed (Supplementary eTable 13).
Using the Greenhouse-Geisser method, the results of ANOVA of repeated measurements of ScvO2, after correction for the internal factor, showed a p-value of > 0.05, indicating no statistical significance in the overall means of ScvO2 at different times. However, the interaction between the ventilation group and treatment time, after correction, had a p-value of < 0.05, suggesting that there was an interaction between ventilation group and treatment time, and that the ventilation group had an effect on the measurements. Furthermore, the overall mean of ScvO2 was found to be significantly different between the two ventilation groups, with a p-value of < 0.05 (Fig 4, Supplementary eTable 7). Moreover, in the control group, ventilation time was negatively correlated with ScvO2 (r = -0.378, p < 0.001). Using multiple linear regression analysis to control confounding factors, the correlation between ventilation time and ScvO2 still existed (Supplementary eTable 14). In the experimental group, ventilation time was positively correlated with ScvO2 (r = 0.257, p = 0.001). With an increase in controlling factors, a positive correlation between ventilation time and ScvO2 was still observed (Supplementary eTable 14).
Using the Greenhouse-Geisser method, the results of ANOVA of repeated measurements of HR, after correction of the sphericity assumption (Mauchly's test, p < 0.05), indicated that the overall means of HR at different times were not the same. The interaction between ventilation group and treatment time was statistically significant (interaction time×ventilation group corrected, p < 0.05), indicating that ventilation group had an effect on the measurements. However, there was no significant difference in the overall mean of HR between the two ventilation groups (p > 0.05, Fig 4, Supplementary eTable 8). Furthermore, in both groups, Lac and SpO2 did not change significantly during ventilation (p > 0.05) (Supplementary eTable 2).
Effects of different ventilation modes on airway pressure
In both groups, VT and MV were unaffected by ventilation time (p > 0.05) (Supplementary eTable 2). After applying the Greenhouse-Geisser correction, the results of ANOVA of repeated measurements for Ppeak, Pplat and ∆P showed that the overall means of Ppeak, Pplat and ∆P at different times were significantly different (p < 0.05). The interaction between ventilation group and treatment time was also significant (p < 0.05), indicating that ventilation group had an effect on the measurements. Specifically, the overall mean of Ppeak was significantly different for the two ventilation groups (p < 0.05, Fig 5, Supplementary eTable 9), while there was no statistical significance in the overall mean of Pplat and ∆P between the two ventilation groups (p > 0.05, Fig 5, Supplementary eTable 10). In addition, in the control group, ventilation time was positively correlated with the Ppeak (r = 0.194, p = 0.014), Pplat (r = 0.184, p = 0.020) and ∆P (r = 0.184, p = 0.020) . On the contrary, in the experimental group, ventilation time was negatively correlated with Ppeak (r = -0.337, p < 0.001), Pplat (r = -0.312, p < 0.001) and ∆P (r = -0.312, p < 0.001) . In both groups, using multiple linear regression analysis to control confounding factors, a correlation between ventilation time and airway pressure was still observed (Supplementary eTable 15-17).