DISCUSSION
Point-of-care LUS has gained significant popularity in emergency settings recently. The earliest studies emphasized that LUS was a valuable tool to identify bronchiolitis and it reduced the need for CXR. Moreover, LUS was in agreement with clinical scores to predict the severity of the disease. Lung ultrasound was found to be an accurate device for both diagnosis and management in emergency departments ³ˉ⁵´ ¹⁰ˉ¹². We have reported that sonographic scores had significant agreement with clinical scores to predict the disease severity and the need for respiratory support, consistent with previous studies. However, clinical scores may vary individually as non-objective parameters. Performing LUS for bilateral anterior and posterior chest areas to obtain echographic scores may be difficult and time-consuming as infants may not stay calm during the whole period of assessment. For this reason, these scoring tools may not be rapid, practical, or objective enough to accurately quantify the severity of the disease and guide management in the emergency department.
Diaphragm ultrasound (DUS) has also been used in adult and pediatric intensive care units in order to identify weaning success from mechanical ventilator, diaphragm paralysis after cardiac/thoracic operations, or normal values for healthy individuals. In these studies, TF, EXC, and TEE were found to be good predictors for successful weaning or diaphragmatic fatigue in particular ¹⁵ˉ¹⁷. There are few such studies involving the pediatric population, unfortunately.
To our knowledge, there is one study evaluating DUS parameters in infants with bronchiolitis. Buonsenso et al. ²¹ concluded that clinical score was correlated with echographic score. The IS values differed significantly between echographic score groups and patients needing respiratory support had higher IS values, supporting our findings. They also reported that EXC was lower in the moderate echographic group compared with the mild group, but there was no statistically significant difference between the mild and normal group. In our study, IS, ES, and EXC values were higher in the moderate-severe group than the mild group for both clinical and echographic scores. Because there were no patients with normal echographic scores, we could not compare that category. Buonsenso et al. ²¹ did not conclude any correlation between respiratory rate and diaphragm parameters, although we found IS to have a significant correlation with respiratory rate at admission. This discrepancy may be related to the different numbers of patients in the two studies (61 versus 104). Although TF was found to be a valuable predictor for extubation success from mechanical ventilator in previous studies, it did not differ between bronchiolitis severity groups in either of these studies. Buonsenso et al. ²¹ also concluded that patients with lower TF values required respiratory support, in contrast to our findings, wherein we considered IS and TEE for predicting the need for respiratory support. In their study, 25 of 61 patients required respiratory support, 5 of whom were in the mild group, 15 in the moderate group, and 5 in the severe group, although all 6 patients who received HFNC oxygen therapy were in the severe group. Measurements were evaluated at admission in both studies. We could repeat the measurement of diaphragm parameters of 2 patients in 2 hours after starting HFNC oxygen therapy; IS, ES, EXC, TEI, and TEE were all decreased as the respiratory support reduced the work of breathing, but this difference was not observed for TF. The number of patients for this observation is too low for a firm conclusion, but this may have been related to the pathophysiology of bronchiolitis. El-Mogy et al. ²⁸ concluded that diaphragm thickness increased and excursion decreased at higher levels of continuous positive airway pressure (CPAP), although there were no significant differences found among diaphragmatic dimensions for levels of either nasal CPAP or high-flow therapy. To our knowledge, there is no study evaluating the effects of HFNC oxygen therapy on DUS parameters among children. In comparison with a study that evaluated normal TEE and EXC measurements in healthy children, our measurements were lower than those among the age group of 1-12 months, suggesting that bronchiolitis causes diaphragmatic dysfunction. Nevertheless, our EXC measurements were also lower than those in a study that evaluated bronchiolitis patients ²¹. We think that measurement in a semi-recumbent or supine position could change EXC values. In an adult study aiming to show changes in diaphragmatic function during breathing control in healthy participants, EXC values before the test were similar to our findings. Diaphragm excursion measurements differed in various studies, interestingly ¹⁹´²⁹.
Our study has some limitations. First, there were few patients in the severe bronchiolitis group according to both mRDAI and BUS values, so we had to compare diaphragm parameters between mild and moderate-severe groups. Second, we could only repeat the examination of DUS after starting HFNC oxygen therapy for 2 of 6 patients; if we had been able to perform the examination for all 6 patients, we could have obtained more useful information.
In conclusion, there was agreement between the clinical and sonographic severity scores, so lung ultrasound was considered valuable for the evaluation of bronchiolitis. To our knowledge, this is the first study evaluating ES measurements with DUS in bronchiolitis patients. To determine the severity of the disease, IS and ES measurements were both qualified as reliable predictors. Additionally, IS values provided convenient information about the length of stay in the hospital. Diaphragm ultrasound appears to be an objective and useful tool to help the physician make decisions about the evaluation and management of bronchiolitis in emergency settings.