Ventilator-induced diaphragmatic dysfunction in extremely preterm infants: a pilot ultrasound study

To investigate the development of diaphragmatic dysfunction in ventilated extremely preterm infants (EPI) using diaphragm ultrasound (DU). EPI of less than 28 weeks’ gestational age who required mechanical ventilation within six hours of birth were included in this prospective, observational study. DU was performed once a day until four days of life. End-inspiratory and end-expiratory thicknesses of the diaphragm were measured, and the diaphragm thickening fraction was calculated. A total of 20 EPI were enrolled. After intubation, there was a progressive reduction in end-inspiratory thickness of the diaphragm from baseline to day 1 (P < 0.001), but not from day 1 to day 2 (P = 0.092), day 2 to day 3 (P = 1.0), or day 3 to day 4 (P = 1.0). There was also a significant reduction in the diaphragm thickening fraction from baseline to day 1 (P < 0.001), but not from day 1 to day 2 (P = 1.0), day 2 to day 3 (P = 1.0), or day 3 to day 4 (P = 1.0). Conclusions: This study provides the first evidence of diaphragmatic dysfunction in ventilated EPI. We demonstrated a rapid progression of ventilator-induced diaphragmatic dysfunction, with a significant reduction in diaphragm thickness and thickening fraction within 24 h of ventilation. What is Known: • Over-assistance of the ventilator suppresses respiratory effort and induces diaphragm unloading, resulting in diaphragm atrophy or dysfunction. • Diaphragmatic dysfunction contributes to prolonged ventilator dependence and poor clinical outcomes. What is New: • Most extremely preterm infants develop diaphragmatic dysfunction after intubation within 24 hours. • Diaphragm thickness and contraction ability measured by ultrasound would be important indicators of worsening breathing or respiratory outcomes. What is Known: • Over-assistance of the ventilator suppresses respiratory effort and induces diaphragm unloading, resulting in diaphragm atrophy or dysfunction. • Diaphragmatic dysfunction contributes to prolonged ventilator dependence and poor clinical outcomes. What is New: • Most extremely preterm infants develop diaphragmatic dysfunction after intubation within 24 hours. • Diaphragm thickness and contraction ability measured by ultrasound would be important indicators of worsening breathing or respiratory outcomes.


Introduction
Mechanical ventilation is a life-saving intervention and essential tool for critically ill patients. However, the use of ventilation is associated with major complications such as infection, barotrauma, tracheal injuries, and ventilator-induced lung injuries [1]. Previous studies in adult patients have shown that mechanical ventilation also results in diaphragmatic dysfunction [2][3][4]. This is because Communicated by Daniele De Luca. over-assistance of the ventilator suppresses respiratory effort and induces diaphragm unloading, resulting in diaphragm atrophy or dysfunction [1][2][3][4]. The reduction of diaphragm thickness or contraction ability caused by mechanical ventilation is termed ventilator-induced diaphragmatic dysfunction (VIDD) [2] and contributes to prolonged ventilator dependence and poor clinical outcomes [1][2][3][4].
Most extremely preterm infants (EPI) require mechanical ventilation to maintain adequate oxygenation and ventilation due to lung immaturity, weak respiratory drive, excessively compliant chest wall, and surfactant deficiency [5]. In ventilated EPI, prolonged mechanical ventilation not only raises the above issues but also increases the risk of bronchopulmonary dysplasia, retinopathy of prematurity, or neurodevelopmental disorders [5]. However, the development of diaphragmatic dysfunction due to mechanical ventilation has not been investigated in EPI.
Ultrasonography is non-invasive, repeatable, and provides real-time information at the bedside. To date, diaphragm ultrasound (DU) has been used to assess diaphragmatic dysfunction in adult patients [1,6,7]. This technique is also used for pediatric patients to assess the development of VIDD [8][9][10][11]. Recently, DU has been used to predict continuous positive airway pressure (CPAP) failure [12] or successful extubation in preterm infants [13]. However, this technique has not been applied to assess the development of diaphragmatic dysfunction in ventilated EPI. We hypothesized that DU is also useful for the assessment of diaphragmatic dysfunction in ventilated EPI. This study is aimed at investigating the development of VIDD after intubation in EPI using ultra-sonographic evaluations.

Methods
This pilot prospective observational study was conducted at a single level III neonatal intensive care unit (NICU) at Ibaraki Children's Hospital from July 2020 to October 2022. We studied EPI of less than 28 weeks' gestational age who required mechanical ventilation within six hours of birth. Only infants who were admitted during working hours (Monday through Friday from 8 AM to 6 PM) were included in this study. Infants with chromosomal abnormalities, major malformations, congenital lung diseases, or congenital heart defects were excluded. Additionally, patients extubated within 48 h of birth were also excluded. This study was approved by the institutional review board (2020IRB-28), and written informed consent was obtained from the parents of each patient.
During the study period, a single consensus manual for EPI management [14] was used. CPAP was the first choice for providing respiratory support. Respiratory support was adjusted to maintain a pH of 7.25-7.35, oxygen saturation of 88-94%, and PaCO 2 of 40-60 mmHg. Infants were intubated for persistent apneic episodes, excessively labored breathing, or failure to meet the abovementioned targets and ventilated using a Babylog VN600 (Dräger, Lübeck, Germany). We routinely applied the synchronized intermittent mandatory ventilation mode to eliminate conflict between mechanical ventilation and spontaneous breathing by synchronizing mechanical motions with infant effort via an inspiration triggering system. The maximum inspiratory pressure and positive end-expiratory pressure were adjusted to maintain a target tidal volume of 4 to 6 mL/kg. Physicians responsible for medical decisions were unaware of the ultrasound data. Detailed respiratory management policies are described in Supplementary Text 1.
DU was conducted randomly by three physicians (Y.H., K.H., and R.M.) with sufficient experience, using a highfrequency (20 MHz) linear probe (Aplio i800, Canon Medical Systems, Japan) with a resolution limit of 0.01 mm. DU was performed with the patients in the supine position after ensuring quiet regular breathing. Diaphragm thickness was evaluated as previously described [1,[6][7][8]11]. To obtain a horizontal view of the diaphragm, the linear probe was placed perpendicular to the midaxillary line at the ninth to tenth intercostal spaces. The diaphragm was visualized superficial to the liver and consisted of three layers bounded by the echogenic membranes of the diaphragmatic pleura and peritoneum. End-inspiratory and end-expiratory thicknesses of the diaphragm (Tdi and Tde, respectively) were measured between the pleural and peritoneal layers (Fig. 1A, B). The diaphragm thickening fraction (DTF) was calculated as follows: DTF = (Tdi-Tde)/ Tde × 100% [1, 6-8, 11, 13]. First, DU, which was used as the baseline measurement, was performed after administering initial respiratory care procedure (such as intubation and surfactant administration), and subsequent recordings were acquired within a 24 ± 4 h time frame. The average value of three measurements was noted for all parameters [8,11,13]. Because VIDD develops within a few days after ventilation in adult and pediatric patients [1,2,7,8,15], we performed DU once a day until four days of life.
Results were expressed as medians (interquartile range (IQR)). Multivariate normality for the linear mixed model, including categorical time as a fixed effect and the subject as a random effect, was applied to assess the impact of time on Tdi and DTF (as a percentage of baseline measurement) after intubation. The marginal means for each time were then compared using small-sample bias-corrected sandwich variance [16], where multiplicity for these comparisons was adjusted using Bonferroni correction. The Henze-Zirkler test for multivariate normality [17] was applied to evaluate the assumptions of the model. All statistical analyses were performed using MVN package [18] of R version 4.2.2 (R Core Team, Vienna, Austria) and SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Significance levels were set at 0.05.

Results
During the study period, 36 EPI were admitted to our unit, and 32 patients were intubated after birth. Seven patients were admitted outside working hours, two patients were extubated within 48 h after birth, and three patients had major abnormalities. Finally, 20 patients were included in this study (Supplementary Fig. 1). 19 patients were intubated in the delivery room. One patient was extubated within four days of life. None of the infants was sedated at the time of the first DU. Demographic and clinical characteristics of the included patients are summarized in Supplementary Table 1.
Tdi at admission was 1.27 mm (IQR: 1-1.34 mm). There was a significant reduction in Tdi over time (P < 0.001) after intubation compared to baseline measurements ( Fig. 2A).

Discussion
To the best of our knowledge, this is the first study to evaluate the development of diaphragmatic dysfunction in ventilated EPI. We found a significant progressive reduction in Tdi and DTF within the first 24 h after intubation and a gradual reduction thereafter. This pilot study allowed us to observe the development of VIDD in EPI after intubation.
Diaphragmatic dysfunction has become a subject of major concern in adult and pediatric intensive care patients [1, 2, Tdi, end-inspiratory thickness of the diaphragm 11] as it is associated with a higher rate of weaning failure and increased duration of mechanical ventilation. Diaphragm is the major respiratory drive muscle, especially for neonates [13]. However, the incidence of VIDD in EPI has not yet been investigated. We found that most ventilated EPI developed VIDD within the first 24 h after intubation. This finding is consistent with those of previous studies in adult and pediatric patients [7,8]. Although VIDD in EPI is undefined, 10% Tdi reduction from baseline has been defined as diaphragm atrophy in previous studies in adult and pediatric patients [1,9,11], and here, it was found in 75% (n = 15) of patients on day 1 and in 90% (n = 18) on day 2. This study provides the first evidence of VIDD development in EPI.
Although we could not evaluate the impact of the incidence of VIDD on respiratory or clinical outcomes due to the small sample sizes, we speculate that diaphragmatic dysfunction might be also associated with poor outcomes in ventilated EPI, as in adult and pediatric patients [1,2,7,8,11]. Considering the quick onset of VIDD, diaphragmatic protective respiratory strategies may be required to improve the respiratory outcomes in EPI. Over-assistance by the ventilator suppresses inspiratory effort, resulting in diaphragm atrophy [1,2]. Conversely, under-assistance by the ventilator increases respiratory effort, which results in an increase in Tdi, indicating loadinduced diaphragmatic injury [1]. Previous studies have confirmed that 12-25% of patients showed an increase in Tdi in ventilated adult patients [1,6]. In contrast, in the present study cohort, only one patient (5%) showed an increase in Tdi. This result might be because preterm infants are unable to compensate for insufficient ventilator assistance with respiratory effort because of their prematurity.
DU is simple and easy for clinicians to practice and monitor diaphragmatic function [8]. The high repeatability and reproducibility are comparable with those reported in previous adult, pediatric, and infant studies [7,8,13,19,20]. No other test exists to assess diaphragm thickness or contractility except DU. Diaphragm thickness and contraction ability measured by DU would be important indicators of worsening breathing or respiratory outcomes [7].
A recent article reported on the simultaneous use of DU and lung ultrasound in preterm infants [21]. Since the relationships between diaphragm function and lung aeration remain unclear, the collaborative use of DU and lung ultrasound may become a subject of major concern in terms of neonatal respiratory care.
This study has some limitations. First, we did not include any non-intubated control patients. In this study cohort, most EPI (32/36) required mechanical ventilation; in particular, all EPI less than 26 weeks were intubated. It would be inherently preferable to conclude the results after comparing our findings with control groups. However, managing EPI using non-invasive respiratory supports may be difficult due to their prematurity. Second, this was a small pilot observational study, and the results need confirmation in a larger study with adequate sample size and power. Moreover, we did not have real baseline diaphragm thickness data for this study cohort, and all baseline thicknesses were measured after NICU admission.

Conclusion
This study suggests the first evidence of diaphragmatic dysfunction in ventilated EPI. We demonstrated a rapid progression of VIDD, with a significant reduction in diaphragm thickness and thickening fraction within 24 h of ventilation. Considering the quick onset of VIDD, diaphragmatic protective respiratory strategies may be required to improve the respiratory outcomes in EPI.