Results
Between March and December 2021, 40 patients were enrolled, and DUS examinations were performed (figure 2). Within 48 hours, 24 (60%) of patients were successfully extubated, while 16 (40%) required invasive or non-invasive MV support. Three of the sixteen patients in the FE group required re-intubation (7.5%), and thirteen received non-invasive MV (32.5%). Between the SW and FE groups, there were no statistically significant differences in age, sex, PIM-3, PRISM-4, need for inotrope infusion, indication for IMV, maximum OSI, and treatments used. The FE group had a longer duration of inotrope infusion days, a longer duration of IMV, and a longer length of stay in the PICU than the SW group (table 1).
Diaphragm measurements of the SW and FE groups are also shown in table 1. While DTi and DTe did not differ significantly between the two groups, DTF was significantly greater in the SW group (55,05 ± 23,75% vs. 30,9 ± 10,38%) (p<0,001). DE was significantly greater in the SW group (14 ± 4,4 mm vs 11,05 ± 3,25 mm) (p<0,001), but there was no significant difference in IS or ES between the two groups. When DTF and DE were compared in terms of post-extubation intervention requirements, while DTF was significantly different between those who required reintubation and those who required NIV (p=0.021), there was no significant difference in terms of DE (p=0.620) (table 2).
The ROC curves for DTF and DE are shown in figure 3. After performing a ROC analysis and calculating the AUC, the AUCs (95% CI) for DTF and DE were 0.962 (0.911-1) and 0.880 (0.762-0.998), respectively. DTF and DE had optimal cutoff points of 40.5 and 12.15 mm, respectively. Of the 26 patients who were successfully extubated, 22 had a DTF greater than 40.5%. Of the 16 who failed extubation, 14 had a DTF less than 40.5%. DTF and DE were found to have a sensitivity and specificity of 91.67 %, 87.50 %, and 83.33 %, 81.25 %, respectively. The positive and negative predictive value of DTF and DE were also shown in table 3.
Discussion
Point of care ultrasound is increasingly used in pediatric critical care settings. Recently, ultrasound has been used to assess the diaphragm functions17, identify VIDD20, pneumonia and bronchiolitis23 and evaluate diaphragm motions during spontaneous breathing trials21. When DTF and DE were determined using DUS in the zone of apposition at the end of the ERT, we found a significant difference between the SW and FE groups. Although, these findings are similar to previous studies evaluating the role of DUS on the weaning policy, this is the first pediatric study evaluating DUS at the end of the ERT. Also, this study showed that DUS findings were significantly differ between NIV requirement and reintubated patients who were considered as extubation failure. Additionally, the FE group had a statistically significantly longer IMV duration, indicating that prolonged IMV duration was associated with increased diaphragmatic atrophy.
Extubation failure is reported as high as 33% in some studies6. The failure rate in our study was 40%, which was significantly higher than the rate in other studies. This difference could be explained by the fact that among all the patients included only three (7.5%) required reintubation and the tendency in our PICU practice to use NIV following extubation. With such high rate of extubation failure and NIV requirement, it is obvious that, more reliable tools are needed to anticipate weaning outcomes and identify indications for NIV treatment. RSBI has been demonstrated to be accurate in predicting extubation failure in adults24. However, this index reflects the contribution of all inspiratory muscles, rather than the diaphragm. In addition, due to different respiratory rates in different ages and the tidal volume is weight-related, there is no constant threshold for RSBI in children. CROP Index found by Thiagarajan et al9, was a good predictor for successful extubation, pediatric studies demonstrated that this index did not reliably predict extubation failure in children25. Although there are studies showing that volumetric capnography can also predict the successful extubation in children, it is not widely used in clinical practice10.
In mechanically ventilated children, inspiratory muscles especially diaphragm weakens and VIDD occurs rapidly. Within the first week of IMV, the diaphragm’s thickness decreases by more than 10%, and an increasing percentage of muscle fat degrades muscle quality17. After a few hours of ERT, the breathing pattern, which is usually normal at the start, deteriorates. As such, we sought to determine whether changes in diaphragm function at the conclusion of the ERT are an accurate predictor of successful extubation.
Yoo et al, compared the predictive values of DE and DTF in adults, found that best prediction thresholds for DE and DTF were 10-15 mm and 20-36% respectively; the sensitivity and specificity for DE and DTF were 0.86-0.79, 0.90-0.80, respectively26. In our study, DE of children was similar to adults: the optimal cut-off was 12.15 mm, DTF of children was higher than adults: the optimal cut-off was 40.5%. This is consistent with the clinical severity difference between FE groups between two studies. They had more severe disease, and six required tracheotomies, while five died during their intensive care unit stay. However, only three of our patients required reintubation, and the remaining thirteen were followed with NIV without reintubation. Lee et al, found that DTF was significantly different between the successful and failed extubation groups in children and a DTF value of <17% was associated with extubation failure17. Also, Ijland et al, found that over 90% of children were successful extubated with a median DTF of 15.2%27. In both studies above, there were only 3 patients in the failed extubation groups, and those 3 patients were also required reintubation. However, children who required reintubation had a DTF value of less than 20% in our study.
Our study had some limitations. First, our population was relatively small, despite the fact that we included a comparable number of patients as previous studies17,18. Specifically, only three patients in the FE group were reintubated, which precluded us from revealing the DUS findings regarding the difference between NIV requirement and reintubation. It has been demonstrated that the use of NIV reduces the need for reintubation in children28. It is necessary to conduct larger, multicenter studies comparing the DUS findings in terms of NIV requirement, reintubation, and successful extubation. Second, because no reference for DTF and DE in children has been established, and in the absence of initial DUS findings prior to IMV, it is difficult to determine whether the included children’s initial diaphragmatic function was abnormal or not. Third, the heterogeneity of the IMV indication limits the generalization of the study results on general PICU patients.