Our study investigated the EIT parameters and respiratory and hemodynamic parameters of C-ARDS patients treated with HFNC. When the flow rate gradually increased from baseline to 30L/min-50 L/min and prone position; a significant and consistent descreasing trend of GI index, CoV, RVD index, ROI ratio, respiartory rate, mean arterial pressure and inreasing P/F ratio.
In the study, based on the GI index and CoV, which are the EIT parameters, we found that when we followed the patients in the prone position and at a flow rate of 30 L/min in HFNC, the heterogeneity in the ARDS lung decreased and the homogeneity increased. Similarly, when evaluated according to the ROI ratio, the lung heterogeneity decreased in the prone position and at low flow rate. In the RVD index, we found that the difference between the supine flow rate of 50 L/min and the flow rate of 50 L/min in the prone position was statistically significant.
High dynamic lung tension (caused by tidal volume) and static lung tension (caused by PEEP administration) during invasive mechanical ventilation therapy are associated with lung injury (23). However, little is known about the effect of flow rates followed in spontaneous breathing and applied in HFNC treatment on AC fields (9). HFNC reduces inspiratory effort, improves lung mechanics, and reduces minute volume by affecting PEEP (24). However, the main challenge here is to find the individualised optimum flow rate in HFNC therapy, just as the optimal PEEP for each patient in invasive mechanical ventilation therapy. HFNC flow rates are heterogeneous, ranging from 15 L/min to 100 L/min (25). To evaluate the effects of different HFNC flow rates on AC, EIT-based and most commonly used GI index, CoV, RVD index, and ROI ratio parameters were used in this study (13).
The GI index is a functional EIT parameter widely used to measure the heterogeneity of lung ventilation (13). When the GI index values in our study were examined, we found that ventilation was homogeneously distributed at 30 L/min and 50 L/min flow rates in the prone position compared to baseline values before treatment. We found that the GI index at a flow rate of 30 L/min in the prone position was 0.5 ± 0.01 and almost homogeneous. This result shows that it may be beneficial to increase the flow rate in HFNC by titration in case of improvement in P/F ratios and a decrease in respiratory rate.
On the other hand, although it was determined that the GI index decreased and homogeneity increased at 30 L/min and 50 L/min flow rates in the supine position compared to the pre-treatment values, there was no statistically significant difference between this position and these flow rates. Similarly, in Li Z. et al.'s (8) study, no statistically significant difference was found in the GI index in the measurements they made at different flow rates after HFNC treatment was started in the supine position. It was found that the difference in our study was due to the fact that the patients were followed in the prone position, and therefore, the GI index level decreased, and homogeneity increased in the prone position compared to the supine position. As in ARDS patients, the rationale for following the patient in the prone position in C-ARDS patients is to ensure that the atelectatic areas in the dependent dorsal region are included in ventilation and to reduce heterogeneous ventilation (26).
At the same time, the prone position reduces the ventilation/perfusion mismatch, hypoxemia, and shunt, which is typical of ARDS (1). Because when the patient is in the prone position, the pleural pressure gradient between the dependent and non-dependent region decreases as a result of gravity; thus, a more homogeneous lung is formed (27).
Another EIT parameter that measures ventilation distribution is CoV. When the CoV values in this study were examined, the CoV value at a flow rate of 30L/min in the prone position was 50.67 ± 1.33. This value indicates that the ventilation distribution is centred in the dorsal-ventral direction of the thorax (14). CoV values were decreased at HFNC supine 30L/min, 50 L/min and prone 50L/min flow rates. This can be explained by the shift of ventilation from the dorsal AC areas to the ventral AC fields in ARDS patients, resulting in increased aeration and even overdistension in the ventral region while decreasing and causing atelectasis in the dorsal regions (28).
Similarly, Li Z. et al. (8), in their study, found that ventilation was reduced in the dorsal regions of the lung in patients in whom HFNC treatment was unsuccessful. This resulted in increased respiratory effort, shortened inspiratory time, decreased minute volume, and inadequate oxygenation in the patients' group. Similarly, in our study, respiratory rate and P/F ratios were found to be higher and P/F ratios lower in the supine position, where the CoV value was low and heterogeneous AC characteristics persisted, compared to the prone position. Lehmann S. et al. (29) found that ventilation was displaced due to gravity in patients with multiple trauma who were placed in the lateral position with EIT in their study.
In our study of the other EIT parameter, the RVDsponbreath index, a significant difference was found only between 50 L/min in the supine position and 50 L/min in the prone. This can be explained by the inspiratory time being too short during spontaneous breathing to calculate a fixed RVD index (30). Bickenbach et al. (16) found the RVDsponbreath index to be statistically significant in their study of spontaneous breathing trials in prolonged weaning (30). However, RVDsponbreath is strongly dependent on the individual variation, depth, and respiratory rate of the patient's spontaneous breathing (16). Therefore, when calculating the RVDsponbreath value, a sequence with spontaneous respirations should be selected (16). Although a 5-minute sequence was chosen to reduce the heterogeneity of spontaneous breathing, only the RVDsponbreath between the two times was statistically significant. As a result, RVD, which was initially developed to measure slow flow in mechanical ventilators, is used during spontaneous breathing, but it is incomprehensible among patients (20).
It is a cheap, simple and effective method to follow the patient in the prone position in ARDS patients and has a 15% benefit on absolute survival (31). It also significantly improves arterial oxygenation compared to the supine position (32). In the study we presented, significant improvements were found in P/F ratios, RR, and MAP in patients placed in the prone position. In previous studies, there are findings that a prone position can reduce ventilator-related lung damage and improve respiratory and hemodynamic parameters (33). On the other hand, in a large, randomised study conducted in recent years, the P/F ratio was found to be similar between survivors and non-survivors in ARDS patients (34). The effectiveness of the prone position at the individual patient level may depend on the sub-phenotypes of ARDS, as in C-ARDS (35).
Another parameter developed to monitor the physiological effects of EIT, which is the dynamic monitoring of regional lung mechanics at the bedside, is the ROI ratio (13). The value obtained by the ratio of the dependent ROI areas to the non-depending ROI areas approached the value of 1 at a flow rate of 30 L/min in the prone position in our study. This shows the homogeneity of AC ventilation. In the COVID-19 ARDS case report presented by Tomasino S. et al. (15), while the patient with Gattinoni type I ARDS had an ROI ratio of 1 in the supine position, the value decreased in the prone position, overdistension occurred, and oxygenation did not improve. More studies are needed to predict the ROI ratio, ARDS subtype, and future treatment.
Limitations:
In our study, in which the effects of HFNC application with EIT have monitored in patients followed up with the diagnosis of C-ARDS, the sample size was limited. However, the sample size could not be calculated since there has been no previous study on this subject in the literature. Since our study is a pioneering study in its field, it is necessary to carry out studies with larger sample groups in the future.