Skip to main content
Download PDF

Ultrasound shear wave elastography for assessing diaphragm function in mechanically ventilated patients: a breath-by-breath analysis

Critical Care volume 24, Article number: 669 (2020) Cite this article

Abstract

Background

Diaphragm dysfunction is highly prevalent in mechanically ventilated patients. Recent work showed that changes in diaphragm shear modulus (ΔSMdi) assessed using ultrasound shear wave elastography (SWE) are strongly related to changes in Pdi (ΔPdi) in healthy subjects. The aims of this study were to investigate the relationship between ΔSMdi and ΔPdi in mechanically ventilated patients, and whether ΔSMdi is responsive to change in respiratory load when varying the ventilator settings.

Methods

A prospective, monocentric study was conducted in a 15-bed ICU. Patients were included if they met the readiness-to-wean criteria. Pdi was continuously monitored using a double-balloon feeding catheter orally introduced. The zone of apposition of the right hemidiaphragm was imaged using a linear transducer (SL10-2, Aixplorer, Supersonic Imagine, France). Ultrasound recordings were performed under various pressure support settings and during a spontaneous breathing trial (SBT). A breath-by-breath analysis was performed, allowing the direct comparison between ΔPdi and ΔSMdi. Pearson’s correlation coefficients (r) were used to investigate within-individual relationships between variables, and repeated measure correlations (R) were used for determining overall relationships between variables. Linear mixed models were used to compare breathing indices across the conditions of ventilation.

Results

Thirty patients were included and 930 respiratory cycles were analyzed. Twenty-five were considered for the analysis. A significant correlation was found between ΔPdi and ΔSMdi (R = 0.45, 95% CIs [0.35 0.54], p < 0.001). Individual correlation displays a significant correlation in 8 patients out of 25 (r = 0.55–0.86, all p < 0.05, versus r = − 0.43–0.52, all p > 0.06). Changing the condition of ventilation similarly affected ΔPdi and ΔSMdi. Patients in which ΔPdi–ΔSMdi correlation was non-significant had a faster respiratory rate as compared to that of patient with a significant ΔPdi–ΔSMdi relationship (median (Q1–Q3), 25 (18–33) vs. 21 (15–26) breaths.min−1, respectively).

Conclusions

We demonstrate that ultrasound SWE may be a promising surrogate to Pdi in mechanically ventilated patients. Respiratory rate appears to negatively impact SMdi measurement. Technological developments are needed to generalize this method in tachypneic patients.

Trial registration

NCT03832231.

Background

Acute respiratory failure is a common cause of admission in the intensive care unit (ICU) that can require invasive mechanical ventilation to relieve respiratory muscles work of breathing and ensure satisfactory gas exchange [1]. However, mechanical ventilation may produce harmful effects leading to the worsening of the patient’s prognosis independently of the primary reason for intubation [2]. More specifically, ventilator-induced diaphragm unloading results in time-dependent disuse atrophy of diaphragm myofibers [3, 4] and decreased diaphragm pressure-generating capacity [5, 6]. This is a serious issue that has been associated with prolonged duration of mechanical ventilation, difficult and prolonged weaning, and increased mortality [7,8,9,10,11]. Minimizing the ventilator unloading-induced diaphragm dysfunction may become a critical goal in the management of mechanically ventilated patients [12]. This strategy, namely the diaphragm protective ventilation [13], could not be implemented without a reliable and easy accessible monitoring of the diaphragm function. However, monitoring the diaphragm function in the ICU is not straightforward [7] as the gold standard relies on the recording of transdiaphragmatic pressure (Pdi) that is not widely available [14]. The later explains the growing interest of diaphragm ultrasound (US) as it provides direct visualization of muscle structure and functioning [15]. Diaphragm US is an interesting tool for assessing diaphragm function [16], monitoring its temporal structure changes (thickness, thickening, excursion, [3, 17]) and predicting weaning outcome [9, 16,17,18,19]. Beyond standard US imaging, ultrasound shear wave elastography (SWE) allows direct and real-time quantification of the mechanical properties of tissues [20]. Briefly, SWE relies on the measurement of the propagation velocity of shear waves remotely generated inside tissues by ultrasonic focused beams. This measurement is performed in three main steps. First an US pushing beam is focused remotely within the tissue. This results in the propagation of transient shear waves, propagating parallel to the surface of the US probe. Second, the probe switches to an ultrafast imaging mode, allowing the tracking of the propagating shear wave along the imaging plane. Finally, after estimating the shear wave speed between two points of the image, the US scanner is able to reconstruct a shear modulus (i.e. stiffness) map of the region of interest [20]. A typical B-Mode image overlaid with the elastography map is presented in Fig. 1. Local muscle shear modulus (i.e. stiffness) measured using SWE has been shown to provide reliable estimates of passive and active muscle force in locomotor muscles [21, 22]. Lately, we demonstrated that changes in diaphragm shear modulus (SMdi) reflect changes in Pdi during isovolumetric inspiratory efforts and ventilation against inspiratory loading in healthy subjects [23]. Recently, Flatres et al. performed measurements of shear modulus in the diaphragm (i.e. at the end of expiration only) and in limb muscles of critically ill patients and found a good intra- and inter-operator reliability, but the relationship with diaphragm function was not investigated [24]. Therefore, the present study aimed at investigating: (i) the agreement between changes in SMdi and changes in transdiaphragmatic pressure in mechanically ventilated patients and (ii), whether or not changes in SMdi are responsive to respiratory load when varying the ventilator settings. Secondary objective was to investigate changes in SMdi in patients undergoing a spontaneous breathing trial.

Fig. 1
figure 1

Typical ultrasound image obtained during shear wave elastography imaging of the diaphragm. Shear modulus map obtained from ultrasound shear wave elastography overlaid with standard ultrasound B-Mode during intercostal scanning of the diaphragm at the right zone of apposition

Full size image

Methods

This study followed the STROBE guidelines for observational study [25]. It was conducted in a medical 15-bed ICU from February 2019 to February 2020. It was approved by an ethical committee (ID RCB: 2018-A022311-54) and referenced on ClinicalTrials.gov (NCT03832231). Written informed consent was obtained from all patients or their relatives.

Participants

Patients older than 18 years old were eligible for inclusion if they had been intubated and ventilated for at least 24 h, and failed a first spontaneous breathing trial (SBT). They could be included if they met predefined readiness-to-wean criteria on daily screening [26] and were therefore ready for a second SBT. Readiness-to-wean criteria were the following: SaO2 > 90% or PaO2/FiO2 ≥ 150 mmHg with a fraction of inspired oxygen (FiO2) ≤ 40%, no or minimal vasopressor, and a positive end-expiratory pressure (PEEP) ≤ 8 cmH2O. Patients who were pregnant, under a legal protection measure, with a contraindication to the insertion of a gastric-esophageal probe (esophageal bleeding), or with known allergies to anesthetizing were not included.

Flow and pressure measurements

A flow sensor (Hamilton Medical, Bonaduz, Switzerland) connected to a spirometer (ADInstruments, Bella Vista, Australia) was used to continuously measure flow. Esophageal (Pes) and gastric (Pga) pressures were monitored using a double-balloon feeding catheter (NutriVentTM, Mirandola, Modena, Italy). The catheter was inserted through the mouth or nostril in the esophagus as demonstrated by the appearance of cardiac artifacts and appropriate negative swings of pressure during inspiration. Both balloons were inflated with 3 to 4 ml of air and connected to separated differential pressure transducers (model DP45-32, Validyne, Northridge, CA). The correct position of the esophageal balloon was confirmed by a dynamic occlusion test allowing the visualization of a corresponding deflation in esophageal pressure and airway pressure during inspiratory effort [27]. Flow and pressure signals were digitized (Powerlab, ADInstruments, Sydney, Australia) and recorded at a sampling frequency of 1 kHz (LabChart, ADInstruments). Pdi was obtained by the online subtraction of Pes from Pga.

Ultrasound imaging and shear wave elastography

The zone of apposition of the right hemidiaphragm was imaged using an ultrafast US scanner (Aixplorer, Supersonic Imagine, France) driving a linear transducer array (SL 10-2, Supersonic Imagine). The probe was placed on the mid-axillary line, vertical to the chest wall, at the 8th–11th intercostal space and the spot was skin marked. US gel was generously applied to optimize acoustic coupling and minimal pressure was applied to the transducer to limit tissue deformation and/or alteration of breathing mechanics. In this location, the diaphragm appears as a three-layered structure just superficial to the liver, consisting of a relatively non-echogenic muscular layer bounded by two echogenic lines corresponding to the diaphragm pleura and peritoneum (Fig. 1). The rotation and angle of the transducer were then finely adjusted to obtain maximal echo intensity from diaphragm pleura and peritoneum. Using the built-in SWE mode of the US scanner, the region of interest was placed at the desired depth to fully cover the diaphragm. The sampling rate of SWE ranged from 1.6 to 2 Hz, depending on diaphragm depth. B-mode images were simultaneously displayed on the US scanner with a frame rate of 12 Hz. B-mode frames and shear wave velocity modulus values maps were retrieved from the US scanner for off-line processing. All US measurements were taken by a single operator (QF).

Study protocol

Patients were in a semi-recumbent position throughout the study. Sedations were not modified during the protocol. The study was carried out as follows: i) recordings under different conditions of pressure support ventilation, ii) recordings during a SBT.

Conditions of mechanical ventilation

At baseline, patients were ventilated under pressure support ventilation mode. In each patient, 4 consecutive conditions of ventilation were applied in a randomized order: (i) initial ventilator settings predefined by the attending physician (PS), (ii) + 25% pressure support with baseline PEEP (PS+25%), (iii) -25% pressure support with baseline PEEP (PS-25%) and (iv) baseline pressure support and zero end-expiratory pressure (PSZEEP). Each breathing condition was maintained for 10 min with 30-s acquisitions performed at 3 and 9 min within the condition.

Spontaneous breathing trial

Every patient underwent a planned 30 min SBT during which no assistance was provided from the ventilator (pressure support and PEEP were set at 0 cmH2O). This modality of SBT, part of usual care of our ICU, reflects the work of breathing after extubation [28]. Thirty seconds US and pressure recordings were performed at the onset of the SBT and every five minutes, for a maximum of 30 min. Failure of the SBT was defined by the following criteria: respiratory rate ≥ 35 breaths/min or increase ≥ 50%, SpO2 ≤ 90% or PaO2 ≤ 50 mmHg (with FiO2 ≥ 50%), heart rate ≥ 140 beats/min, new onset of supraventricular or ventricular arrhythmia, systolic arterial pressure > 180 or < 90 mmHg, alteration of consciousness, and diaphoresis or any signs of respiratory distress [26]. In case of failure of the SBT, initial ventilator settings were resumed. Otherwise, the SBT was defined as successful and the decision of extubation was taken by the clinician in charge.

Data analysis

Analysis of shear modulus maps

Data were analyzed offline using standardized MATLAB (Mathworks, Natick, MA, USA) scripts. A rectangular region of interest (ROI) was drawn manually in the center of the diaphragm on the first frame of each recording. The ROI was replicated on subsequent frames. Diaphragm shear modulus was calculated as SMdi = ρ ⋅Vs2, where Vs is the velocity of the shear wave and ρ is the muscle density (1000 kg/m3). SMdi was reported as the median shear modulus within each ROI.

Breath-by-breath analysis

Changes in Pdi (ΔPdi), Pes (ΔPes), Pga (ΔPga) and SMdi (ΔSMdi) were computed for each respiratory cycle. The cycles were delimited by the deflations of the esophageal pressure signal and not with the flow signal since it could mask the onset of the inspiratory effort, especially when the patient has to overcome intrinsic PEEP [29, 30]. During inspiration, ΔPes was computed as the difference between the start of the decrease in Pes and the negative peak value of Pes. ΔPga was computed as the difference between the start of the increase in Pga and the positive peak value of Pga during inspiration. ΔPdi was computed as the difference between the start of the increase in Pdi and the positive peak value of Pdi during inspiration. ΔSMdi was calculated as the difference between the value of SMdi at the start of inspiration the positive peak value of SMdi during inspiration. Transdiaphragmatic pressure time product (PTPdi) per breath was also computed [31]. For every ventilatory condition, the 3 cycles with the least variation in ΔPdi were considered as representative of a given ventilatory condition and selected for further analysis. Maximal transdiaphragmatic pressure (Pdimax) was measured before the SBT using a one-way valve allowing exhalation only [32]. Pdimax was calculated as the difference between Pdi at functional residual capacity and maximal Pdi during the Mueller maneuver. A video showing US imaging along with the temporal evolution of flow, internal pressures, and SMdi is available in Additional file 1: S1.

Statistics

Results are presented as median (Q1-Q3) for descriptive statistics. We calculated our sample size based on the expected correlation between ΔPdi and ΔSMdi of 0.7 [23]. A required sample size of 17 patients was obtained. Sample size was then increased to compensate for any patients that would have to be withdrawn from the study for any reason. Differences in measured variables across the conditions of ventilation were assessed using linear mixed models. Linear mixed models were chosen over traditional two-way repeated ANOVAs given their ability to handle unbalanced data [33]. More specifically, six patients who failed the SBT had only one measurement during the SBT so that they would have been excluded from analysis using two-way repeated ANOVAs. Linear mixed models were also used to compare measured variables between patients based on the outcome of the SBT (success or failure) and to test the interaction effect between ventilatory condition × SBT outcome. Data from the ninth minute of each PS conditions and during the first and last measurements of the SBT were used for linear mixed models. Tukey’s post hoc tests were performed when a significant main or interaction effect was found.

Pearson’s correlation coefficients were used to investigate within-individual relationships between variables. Repeated measure correlation (R, 95% CIs) was used for determining overall relationships between variables [34]. Paired t-tests were used to compare differences in breathing pattern, diaphragm function, and SMdi between the SBT start and end-points for the failure and success groups. Unpaired t-tests were used to compare differences in breathing pattern, diaphragm function, and SMdi between patients who failed and succeeded the SBT. Unpaired t-tests were also used to compare differences in patients’ characteristics (age and BMI), breathing pattern, diaphragm function, and ΔSMdi between patients who presented a significant (p < 0.05) ΔPdi–ΔSMdi correlation and their counterparts. Analyses were performed in the computing environment R [35]. Significance was set at p < 0.05 for all tests.

Results

Population

Between February 2019 and February 2020, 1087 patients were admitted within the ICU. 404 patients received invasive mechanical ventilation, 99 fulfilled the inclusion criteria leading to the enrollment of 30 patients. Twenty-five patients were considered for subsequent analysis (see the flow chart of the study in Fig. 2). The main characteristics of patients are displayed in Table 1. The main reason for intubation was acute respiratory failure, and patients were ventilated since 4 (3–7) days at the time of inclusion. At baseline, patients were receiving a pressure support level of 10 (10–13) cmH2O and a PEEP level of 5 (5–5) cmH2O. A total of 3878 breathing cycles were recorded and 930 were considered for further analysis (i.e. corresponding to triplicate for each condition of ventilation). All patients tolerated each condition of the protocol.

Fig. 2
figure 2

Flowchart of the study

Full size image
Table 1 Characteristics of patients at inclusion
Full size table

Changes in diaphragm shear modulus and diaphragm function across breathing conditions

Changes in breathing pattern and ΔPes, ΔPga, ΔPdi, PTPdi, and ΔSMdi at each condition of the protocol are shown in Table 2, and time course of ΔPdi, PTPdi, and ΔSMdi during the protocol is displayed in Fig. 3. There was a significant relationship between the level of ventilatory assistance and the breathing pattern, namely the respiratory rate increased and the tidal volume decreased while the ventilatory assistance decreased. Similarly, both PTPdi and ΔPdi significantly increased while the level of assistance decreased. Under PS ventilatory conditions, ΔPdi ranged between 0.1 and 38.1 cmH2O and between 0.6 and 50.7 cmH2O during the SBT. Similarly, ΔSMdi presented a stepwise increase corresponding to each decrease of the ventilatory assistance level.

Table 2 Physiological variables and diaphragm shear modulus under the different ventilatory conditions
Full size table
Fig. 3
figure 3

Diaphragm shear modulus and transdiaphragmatic pressure across different breathing conditions. PTPdi, pressure–time product of transdiaphragmatic pressure (Panel a); ΔPdi, inspiratory change in transdiaphragmatic pressure (Panel b); ΔSMdi, inspiratory change in diaphragm shear modulus assessed using ultrasound shear wave elastography (Panel c). The error bars correspond to 25th and 75th percentile. PS, pressure support ventilation with baseline inspiratory support and positive end-expiratory pressure; PS+25%, PS with inspiratory pressure support increased by 25%; PS-25%, PS with inspiratory pressure support decreased by 25%; PSZEEP, PS with baseline inspiratory support and positive end-expiratory pressure set at 0; SBT, spontaneous breathing trial; SBT Start, start of the SBT; SBT End, end of the SBT. a Significantly different from PS+25%; b Significantly different from PS; c significantly different from PS−25%; d Significantly different from PSZEEP (all p < 0.05)

Full size image

Correlations between changes in diaphragm shear modulus and diaphragm function

Figure 4 shows individual and global correlations between ΔSMdi and ΔPdi. Repeated measure correlation showed a significant overall correlation between ΔSMdi and ΔPdi (R = 0.45, 95% CIs [0.35 0.54], p < 0.001). Regarding within-subject correlation analysis, ΔSMdi and ΔPdi exhibited a significant correlation in 8 patients and no significant correlation in 17 patients (see Supplemental Information S2 for individual correlation coefficient and p values). Figure 5 shows the temporal evolution of the airway flow, Pes, Pga, Pdi and SMdi in a patient with a strong ΔPdi–ΔSMdi correlation (r = 0.81, p = 0.002) and in a patient with no ΔPdi–ΔSMdi correlation (r = 0.14, p = 0.643) during the protocol. Patients in which ΔPdi–ΔSMdi correlation was non-significant were older, had a faster respiratory rate, lower ΔPdi, lower PTPdi and lower ΔPga as compared to that of patients with significant ΔPdi–ΔSMdi correlation (Table 3). No difference in BMI was found between patients with and without a significant ΔPdi–ΔSMdi correlation (Table 3).

Fig. 4
figure 4

Relationship between changes in diaphragm shear modulus and changes in transdiaphragmatic pressure. Averaged data (panel a, data are shown as median (Q1–Q3)) and all data points with individual and overall linear regression lines (panel b). Panel c displays the individual linear regressions in patients with a significant ΔPdi-ΔSMdi correlation (p < 0.05). Panel d displays the individual linear regressions in patients with no significant ΔPdi-ΔSMdi correlation (p > 0.05). ΔPdi, inspiratory change in transdiaphragmatic pressure; ΔSMdi, inspiratory change in diaphragm shear modulus assessed using ultrasound shear wave elastography; PS, pressure support ventilation with baseline inspiratory support and positive end-expiratory pressure; PS+25%, PS with inspiratory pressure support increased by 25%; PS−25%, PS with inspiratory pressure support decreased by 25%; PSZEEP, PS with baseline inspiratory support and positive end-expiratory pressure set at 0; SBT Start, start of the spontaneous breathing trial. In panel a., only cycles gathered at the end of each condition and at the start of SBT were used. In panel b, cycles gathered at all time-points were used

Full size image
Fig. 5
figure 5

Physiological variables and diaphragm shear modulus over time in two patients. Temporal evolution of airway flow, esophageal (Pes), gastric (Pga) and transdiaphragmatic (Pdi) pressures, and diaphragm shear modulus (SMdi) in a patient with a significant ΔPdi–ΔSMdi relationship (r = 0.81, p = 0.002, panel a) and in a patient with a non-significant ΔPdi–ΔSMdi relationship (r = 0.14, p = 0.643, panel b). Respiratory rates were of 12 and 33 breaths.min−1 for panel a and b, respectively

Full size image
Table 3 Characteristics of patients, changes in breathing pattern, diaphragm function, and diaphragm shear modulus in patients with and without a significant correlation between changes in transdiaphragmatic pressure and changes in diaphragm shear modulus
Full size table

Comparison of diaphragm shear modulus in patients who failed and succeeded the spontaneous breathing trial

Thirteen patients (52%) failed the SBT. Reasons of failure were acute respiratory distress (5/13), neurologic impairment (4/13) and weaning induced pulmonary edema (4/13). ∆Pdi and PTPdi tended to be higher at the start of the SBT in patients who failed the SBT as compared to their counterpart. No difference was found regarding ∆SMdi between the two groups of patients. Table 4 presents the breathing pattern and clinical characteristics of patients at the start and the end of the SBT, based on the outcome of the SBT.

Table 4 Physiological variables and diaphragm shear modulus according to the outcome of the spontaneous breathing trial
Full size table

Discussion

The present work provides new insights regarding the use of ultrasound shear wave elastography for the assessment of diaphragm function in patients under mechanical ventilation using a breath-by-breath analysis. First, we found that changes in diaphragm shear modulus and changes in transdiaphragmatic pressure were significantly correlated. However, when considering the relationship between diaphragm shear modulus and transdiaphragmatic pressure within individuals, the correlation was significant in only a third of patients and it was absent in the remaining patients. Second, we found no significant difference in diaphragm function and change in diaphragm shear modulus in response to changes in ventilatory conditions. Changes in transdiaphragmatic pressure and in diaphragm shear modulus significantly increased during the SBT. Third, we found no significant difference regarding SBT-induced changes in diaphragm shear modulus between patients who failed or succeeded the spontaneous breathing trial.

Diaphragm shear modulus as a surrogate of transdiaphragmatic pressure in mechanically ventilated patients

Changes in SMdi have been demonstrated to be strongly related to changes in mouth pressure during isovolumetric inspiratory efforts in healthy subjects (R2 = 0.94 ± 0.05) [36]. In a recent work, our group demonstrated a strong relationship between ΔSMdi and ΔPdi during both isovolumetric inspiratory effort and inspiratory threshold loading in healthy subjects [23]. Our findings in turn demonstrate in mechanically ventilated patients the linear relationship between ΔSMdi and ΔPdi (Fig. 4a). However, in the current work, ΔSMdi significantly correlated to ΔPdi in only one third of patients (Fig. 4b–d). There are several potential explanations for these findings. First, the range of ΔPdi measured, all conditions of ventilation considered, was largely smaller than those induced by isovolumetric inspiratory effort or inspiratory threshold loading (both performed between 0 and 60% of maximal inspiratory pressure) in our previous study. Consequently, the range over which ΔPdi and ΔSMdi values were measured (0–120 cmH20) was much wider than the one in the present study (0–50 cmH2O with 62% of ΔPdi < 10 cmH2O). Therefore, the moderate relationship between ΔSMdi and ΔPdi observed in our study may be partly explained by the relatively low, albeit more physiological, range over which diaphragm effort was analyzed. Our results support this hypothesis, as patients with a significant ΔPdi–ΔSMdi correlation displayed higher ΔPdi (Table 3). Second, the sampling rate of SWE (~ 2 Hz) is a critical factor for the monitoring of cyclic activity of a muscle such as the diaphragm. Indeed, the computation of ΔSMdi relies on the assumption that both the maximal and minimal value of SMdi during a breathing cycle are successfully recorded. Consequently, tachypnea emerges as a major issue when ΔSMdi needs to be determined. As the patient increases its respiratory rate, it becomes likely that minimal and/or maximal SMdi values are flawed because the low sampling rate of SWE does not ensure that the measurement is performed at the end of inspiration and expiration. More precisely, maximal and minimal SMdi values are expected to occur at the end of the inspiration and expiration, respectively. An increase in respiratory rate leads to a reduction of the inspiratory time. As the number of SMdi values acquired during a breathing cycle is limited (i.e. two SMdi values per second), the shortening of the inspiratory time limits the likelihood of measuring both the minimal and maximal SMdi values during a given respiratory cycle. Consequently, ∆SMdi may be underestimated when SMdi is not recorded at the very end of inspiration and/or expiration, thus weakening the relationship between ∆SMdi and ∆Pdi. Our results also support this idea. When ΔPdi–ΔSMdi correlation was not significant, patients presented a significantly higher respiratory rate compared to their counterparts (Table 3). Also, ΔSMdi were significantly lower in patients with no ΔPdi–ΔSMdi correlation compared to patients with a significant ΔPdi–ΔSMdi correlation. This finding corroborates the idea that ΔSMdi is underestimated when tachypnea occurs. To illustrate this phenomenon, videos showing US imaging along with the temporal evolution of flow, internal pressures, and SMdi in one patient with a low respiratory rate and one patient with a high respiratory rate are available in Additional file 1: S1 and Additional file 3: S3, respectively. These findings emphasize the issue that changes in diaphragm stiffness cannot be captured when tachypnea occurs, a situation that is frequent in critically ill patients. Therefore, substantial technological developments aiming at increasing the frame rate of SWE when used in the diaphragm are required to make its use generalizable in all ICU patients. Currently, SWE relies on the measurement of propagating shear wave velocity at multiple laterally spaced points. Recent work showed that by reducing the number of lateral points over which the shear wave velocity is calculated, accurate estimates of the mechanical properties of a viscoelastic material can be obtained [37]. This promising technique could significantly reduce the computational time needed to obtain a shear modulus map and theoretically increase the sampling rate of SWE by four. Such frame rate could improve the accuracy of SMdi measurement in the case of tachypnea. Combining SWE with previously identified indices such as diaphragm excursion, thickening fraction [19, 38, 39], tissue Doppler imaging [40], and strain [41] might also improve the performance of diaphragm US for gauging diaphragm function. Between-day, intra- and inter-operator reliability of diaphragm SWE elastography was not assessed in the current work. This shall be investigated when technical limitations, in particular regarding the frame rate, will be resolved. In summary, the absence of significant ΔPdi–ΔSMdi correlation in two-third of the patients included may be explained by two main factors: i) the narrower range of ΔPdi values in mechanically ventilated patients (0–50 cmH2O), as compared to our previous work in healthy subjects (0–120 cmH2O, [22]) and ii) the higher inspiratory rate observed in individuals for no significant ΔPdi–ΔSMdi correlation was found.

The sensitivity of diaphragm shear modulus for detecting modification in respiratory load

As expected, removing PS was associated with a significant increase in diaphragm function as assessed using ΔPdi and PTPdi as repeatedly observed [30, 42, 43]. However, when the inspiratory support level or PEEP was modified, we found no significant change in PTPdi and ΔPdi as compared to baseline PS settings (Fig. 3). This might be explained by the chosen ventilatory condition. More specifically, increasing or decreasing PS by 25% led to relatively small absolute changes in PS. Similarly, removing PEEP but maintaining initial PS settings did not lead to an increase in PTPdi or ΔPdi, pointing out that the chosen ventilatory conditions appear to be too close to each other to detect changes in diaphragm function. Interestingly and despite limitations of SWE mentioned above, we found that ΔSMdi also increased during the SBT and that differences with other breathing conditions were identical to those observed in PTPdi and ΔPdi. These findings highlight that an increase in diaphragm function may be detected by diaphragm SWE that is a promising track in the field of noninvasive diaphragm function in the ICU.

Comparison of patients who failed or succeeded the spontaneous breathing trial

A secondary aim of the present study was to investigate differences in diaphragm function and diaphragm shear modulus according to the outcome of a SBT. As previously reported, we observed that patients who failed the SBT had larger (i.e. almost four times higher) ΔPdi and ΔPTPdi at the start of the SBT as compared to patients who succeeded (Table 4, [44]). Regarding ΔSMdi, no difference was observed between patients who succeeded or failed the SBT, whether at the start or the end of the SBT. Possibly, the relatively small sample size, further divided in two groups exacerbated the limitations of SWE mentioned above, limiting its use to discriminate patients according to the SBT outcome in the present work. Also, the outcome of the SBT depends on a large range of clinical parameters (i.e. desaturation, increased arterial pressure, etc.), which would primarily differentiate patients succeeding or failing the SBT before any difference could be observed on ΔSMdi.

Strength and limitations

The current work is based on a breath-by-breath analysis. All US acquisitions were synchronized with the physiological parameters. This method allowed a direct comparison of diaphragm indices for a given breathing cycle. This strategy ensures an unbiased comparison of the various diaphragm function indices, as ΔPdi and ΔSMdi are compared for the exact same diaphragm contraction. All data were analyzed offline, using standardized scripts, by an operator blinded to the ventilatory condition. Conversely, this study has several limitations. As mentioned above, the relatively low sampling rate of SWE hinders its applicability in tachypneic patients. Our team is currently working on the development of specific US sequences that would allow a significant increase in SWE sampling rate. Such improvement is needed to accurately measure ΔSMdi in case of tachypnea. In this work, PS and PEEP were purposely changed to increase or decrease respiratory load. However, no differences in ΔPdi, PTPdi or ΔSMdi were found across the conditions of PS ventilation. Inspiratory load was only significantly increased during the SBT. This may be explained by the limited range of inspiratory effort, which did not induce significant changes in diaphragm function.

Conclusions

Monitoring changes in diaphragm shear modulus as assessed using ultrasound shear wave elastography is promising as a noninvasive and specific approach to assess diaphragm function within the ICU. However, limitations of ultrasound shear wave elastography arise from its limited sampling rate when tachypnea occurs. Further technological and methodological developments are required to optimize the use of diaphragm shear wave elastography for the ICU.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

  1. Goligher EC, Ferguson ND, Brochard LJ. Clinical challenges in mechanical ventilation. Lancet. 2016;387(10030):1856–66.

    Article  Google Scholar 

  2. Demoule A, Jung B, Prodanovic H, Molinari N, Chanques G, Coirault C, Matecki S, Duguet A, Similowski T, Jaber S. Diaphragm dysfunction on admission to the intensive care unit: prevalence, risk factors, and prognostic impact-a prospective study. Am J Respir Crit Care Med. 2013;188(2):213–9.

    Article  Google Scholar 

  3. Goligher EC, Dres M, Fan E, Rubenfeld GD, Scales DC, Herridge MS, Vorona S, Sklar MC, Rittayamai N, Lanys A, et al. Mechanical ventilation-induced diaphragm atrophy strongly impacts clinical outcomes. Am J Respir Crit Care Med. 2018;197(2):204–13.

    Article  CAS  Google Scholar 

  4. Levine S, Nguyen T, Taylor N, Friscia ME, Budak MT, Rothenberg P, Zhu J, Sachdeva R, Sonnad S, Kaiser LR, et al. Rapid disuse atrophy of diaphragm fibers in mechanically ventilated humans. N Engl J Med. 2008;358(13):1327–35.

    Article  CAS  Google Scholar 

  5. Hermans G, Agten A, Testelmans D, Decramer M, Gayan-Ramirez G. Increased duration of mechanical ventilation is associated with decreased diaphragmatic force: a prospective observational study. Crit Care. 2010;14(4):R127.

    Article  Google Scholar 

  6. Jaber S, Petrof BJ, Jung B, Chanques G, Berthet JP, Rabuel C, Bouyabrine H, Courouble P, Koechlin-Ramonatxo C, Sebbane M, et al. Rapidly progressive diaphragmatic weakness and injury during mechanical ventilation in humans. Am J Respir Crit Care Med. 2011;183(3):364–71.

    Article  CAS  Google Scholar 

  7. Dres M, Demoule A. Monitoring diaphragm function in the ICU. Curr Opin Crit Care. 2020a;26(1):18–25.

    Article  Google Scholar 

  8. Dres M, Dubé BP, Mayaux J, Delemazure J, Reuter D, Brochard L, Similowski T, Demoule A. Coexistence and impact of limb muscle and diaphragm weakness at time of liberation from mechanical ventilation in medical intensive care unit patients. Am J Respir Crit Care Med. 2017;195(1):57–66.

    Article  Google Scholar 

  9. Dres M, Goligher EC, Dubé BP, Morawiec E, Dangers L, Reuter D, Mayaux J, Similowski T, Demoule A. Diaphragm function and weaning from mechanical ventilation: an ultrasound and phrenic nerve stimulation clinical study. Ann Intensive Care. 2018;8(1):53.

    Article  Google Scholar 

  10. Jung B, Moury PH, Mahul M, de Jong A, Galia F, Prades A, Albaladejo P, Chanques G, Molinari N, Jaber S. Diaphragmatic dysfunction in patients with ICU-acquired weakness and its impact on extubation failure. Intensive Care Med. 2016;42(5):853–61.

    Article  Google Scholar 

  11. Kim WY, Suh HJ, Hong SB, Koh Y, Lim CM. Diaphragm dysfunction assessed by ultrasonography: influence on weaning from mechanical ventilation. Crit Care Med. 2011;39(12):2627–30.

    Article  Google Scholar 

  12. Schepens T, Goligher EC. Lung- and diaphragm-protective ventilation in acute respiratory distress syndrome: rationale and challenges. Anesthesiology. 2019;130(4):620–33.

    Article  Google Scholar 

  13. Goligher EC, Dres M, Patel BK, Sahetya SK, Beitler JR, Telias I, Yoshida T, Vaporidi K, Grieco DL, Schepens T et al. Lung and diaphragm-protective ventilation. Am J Respir Crit Care Med. 2020.

  14. Laveneziana P, Albuquerque A, Aliverti A, Babb T, Barreiro E, Dres M, Dube BP, Fauroux B, Gea J, Guenette JA, et al. ERS statement on respiratory muscle testing at rest and during exercise. Eur Respir J. 2019;53(6):1801214.

    Article  Google Scholar 

  15. Ueki J, De Bruin PF, Pride NB. In vivo assessment of diaphragm contraction by ultrasound in normal subjects. Thorax. 1995;50(11):1157–61.

    Article  CAS  Google Scholar 

  16. Dubé BP, Dres M, Mayaux J, Demiri S, Similowski T, Demoule A. Ultrasound evaluation of diaphragm function in mechanically ventilated patients: comparison to phrenic stimulation and prognostic implications. Thorax. 2017;72(9):811–8.

    Article  Google Scholar 

  17. Tuinman PR, Jonkman AH, Dres M, Shi ZH, Goligher EC, Goffi A, Korte C, Demoule A, Heunks L. Respiratory muscle ultrasonography: methodology, basic and advanced principles and clinical applications in ICU and ED patients-a narrative review. Intensive Care Med. 2020. 46(4): 594–605.

  18. DiNino E, Gartman EJ, Sethi JM, McCool FD. Diaphragm ultrasound as a predictor of successful extubation from mechanical ventilation. Thorax. 2014;69(5):423–7.

    Article  Google Scholar 

  19. Goligher EC, Laghi F, Detsky ME, Farias P, Murray A, Brace D, Brochard LJ, Bolz SS, Rubenfeld GD, Kavanagh BP, et al. Measuring diaphragm thickness with ultrasound in mechanically ventilated patients: feasibility, reproducibility and validity. Intensive Care Med. 2015;41(4):642–9.

    Article  Google Scholar 

  20. Gennisson JL, Deffieux T, Fink M, Tanter M. Ultrasound elastography: principles and techniques. Diagn Interv Imaging. 2013;94(5):487–95.

    Article  Google Scholar 

  21. Gennisson JL, Cornu C, Catheline S, Fink M, Portero P. Human muscle hardness assessment during incremental isometric contraction using transient elastography. J Biomech. 2005;38(7):1543–50.

    Article  Google Scholar 

  22. Hug F, Tucker K, Gennisson JL, Tanter M, Nordez A. Elastography for muscle biomechanics: toward the estimation of individual muscle force. Exerc Sport Sci Rev. 2015;43(3):125–33.

    Article  Google Scholar 

  23. Bachasson D, Dres M, Nierat MC, Gennisson JL, Hogrel JY, Doorduin J, Similowski T. Diaphragm shear modulus reflects transdiaphragmatic pressure during isovolumetric inspiratory efforts and ventilation against inspiratory loading. J Appl Physiol (1985) 2019. 126(3):699–707.

  24. Flatres A, Aarab Y, Nougaret S, Garnier F, Larcher R, Amalric M, Klouche K, Etienne P, Subra G, Jaber S, et al. Real-time shear wave ultrasound elastography: a new tool for the evaluation of diaphragm and limb muscle stiffness in critically ill patients. Crit Care. 2020;24(1):34.

    Article  Google Scholar 

  25. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7.

    Article  Google Scholar 

  26. Boles JM, Bion J, Connors A, Herridge M, Marsh B, Melot C, Pearl R, Silverman H, Stanchina M, Vieillard-Baron A, et al. Weaning from mechanical ventilation. Eur Respir J. 2007;29(5):1033–56.

    Article  Google Scholar 

  27. Baydur A, Behrakis PK, Zin WA, Jaeger M, Milic-Emili J. A simple method for assessing the validity of the esophageal balloon technique. Am Rev Respir Dis. 1982;126(5):788–91.

    CAS  PubMed  Google Scholar 

  28. Sklar MC, Burns K, Rittayamai N, Lanys A, Rauseo M, Chen L, Dres M, Chen GQ, Goligher EC, Adhikari NKJ et al: Effort to breathe with various spontaneous breathing trial techniques: a physiologic meta-analysis. Am J Respir Crit Care Med 2017. 195(11):1477–1485.

  29. Akoumianaki E, Maggiore SM, Valenza F, Bellani G, Jubran A, Loring SH, Pelosi P, Talmor D, Grasso S, Chiumello D, et al. The application of esophageal pressure measurement in patients with respiratory failure. Am J Respir Crit Care Med. 2014;189(5):520–31.

    Article  Google Scholar 

  30. Mauri T, Yoshida T, Bellani G, Goligher EC, Carteaux G, Rittayamai N, Mojoli F, Chiumello D, Piquilloud L, Grasso S, et al. Esophageal and transpulmonary pressure in the clinical setting: meaning, usefulness and perspectives. Intensive Care Med. 2016;42(9):1360–73.

    Article  Google Scholar 

  31. Vivier E, Mekontso Dessap A, Dimassi S, Vargas F, Lyazidi A, Thille AW, Brochard L. Diaphragm ultrasonography to estimate the work of breathing during non-invasive ventilation. Intensive Care Med. 2012;38(5):796–803.

    Article  Google Scholar 

  32. Multz AS, Aldrich TK, Prezant DJ, Karpel JP, Hendler JM. Maximal inspiratory pressure is not a reliable test of inspiratory muscle strength in mechanically ventilated patients. Am Rev Respir Dis. 1990;142(3):529–32.

    Article  CAS  Google Scholar 

  33. Krueger C, Tian L. A comparison of the general linear mixed model and repeated measures ANOVA using a dataset with multiple missing data points. Biol Res Nurs. 2004;6(2):151–7.

    Article  Google Scholar 

  34. Bakdash JZ, Marusich LR. Repeated measures correlation. Front Psychol. 2017;8:456.

    Article  Google Scholar 

  35. R Core Team: R: A Language and Environment for Statistical Computing. In. Vienna, Austria: R Foundation for Statistical Computing; 2020.

  36. Chino K, Ohya T, Katayama K, Suzuki Y. Diaphragmatic shear modulus at various submaximal inspiratory mouth pressure levels. Respir Physiol Neurobiol. 2018;252–253:52–7.

    Article  Google Scholar 

  37. Kijanka P, Ambrozinski L, Urban MW. Two point method for robust shear wave phase velocity dispersion estimation of viscoelastic materials. Ultrasound Med Biol. 2019;45(9):2540–53.

    Article  Google Scholar 

  38. Dres M, Demoule A. Monitoring diaphragm function in the ICU. Current Opinion in Critical Care. 2020b;26(1):18–25.

    Article  Google Scholar 

  39. Tuinman PR, Jonkman AH, Dres M, Shi ZH, Goligher EC, Goffi A, de Korte C, Demoule A, Heunks L Respiratory muscle ultrasonography: methodology, basic and advanced principles and clinical applications in ICU and ED patients-a narrative review. Intensive Care Med 2020.

  40. Fayssoil A, Nguyen LS, Ogna A, Stojkovic T, Meng P, Mompoint D, Carlier R, Prigent H, Clair B, Behin A, et al. Diaphragm sniff ultrasound: normal values, relationship with sniff nasal pressure and accuracy for predicting respiratory involvement in patients with neuromuscular disorders. PLoS ONE. 2019;14(4):e0214288.

    Article  CAS  Google Scholar 

  41. Oppersma E, Hatam N, Doorduin J, van der Hoeven JG, Marx G, Goetzenich A, Fritsch S, Heunks LMA, Bruells CS. Functional assessment of the diaphragm by speckle tracking ultrasound during inspiratory loading. J Appl Physiol (1985) 2017, 123(5):1063–1070.

  42. Jubran A, Tobin MJ. Pathophysiologic basis of acute respiratory distress in patients who fail a trial of weaning from mechanical ventilation. Am J Respir Crit Care Med. 1997a;155(3):906–15.

    Article  CAS  Google Scholar 

  43. Tobin MJ, Laghi F, Jubran A. Ventilatory failure, ventilator support, and ventilator weaning. Compr Physiol. 2012;2(4):2871–921.

    PubMed  Google Scholar 

  44. Jubran A, Tobin MJ. Passive mechanics of lung and chest wall in patients who failed or succeeded in trials of weaning. Am J Respir Crit Care Med. 1997b;155(3):916–21.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

The authors thank all participants, their families, and the ICU staff for their kind support during the conduct of the study.

Funding

This study was supported by the Association Française Contre Les Myopathies (AFM), and the Fondation EDF that supports the RespiMyo project at the Institute of Myology, Paris. The PhD fellowship of TP is funded thanks to the funds provided to the Institute of Myology by the Fondation EDF. QF master work was funded by 2018–2019 “Année Recherche”, a research grant funded by the Ile-de-France Regional Health Agency.

Author information

Author notes

  1. Quentin Fossé, Thomas Poulard, Damien Bachasson, and Martin Dres have contributed equally to this work

Authors and Affiliations

  1. Sorbonne Université, INSERM, UMRS1158 Neurophysiologie respiratoire expérimentale et clinique, Paris, France

    Quentin Fossé, Marie-Cécile Niérat, Sara Virolle, Elise Morawiec, Thomas Similowski, Alexandre Demoule & Martin Dres

  2. AP-HP. Sorbonne Université, Hôpital Pitié-Salpêtrière, Service de Pneumologie, Médecine intensive – Réanimation (Département “R3S”), 75013, Paris, France

    Quentin Fossé, Sara Virolle, Elise Morawiec, Thomas Similowski, Alexandre Demoule & Martin Dres

  3. Institut de Myologie, Laboratoire de Physiologie et d’Evaluation Neuromusculaire, Paris, France

    Thomas Poulard, Jean-Yves Hogrel & Damien Bachasson

  4. Laboratoire d’Imagerie Biomédicale Multimodale, BioMaps, Université Paris-Saclay, CEA, CNRS UMR 9011, INSERM UMR1281, SHFJ, Orsay, France

    Thomas Poulard & Jean-Luc Gennisson

Authors
  1. Quentin Fossé
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Thomas Poulard
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Marie-Cécile Niérat
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sara Virolle
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Elise Morawiec
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Jean-Yves Hogrel
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Thomas Similowski
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Alexandre Demoule
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Jean-Luc Gennisson
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Damien Bachasson
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Martin Dres
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

DB and MD conceptualized and designed the study. MD, DB, QF, MCN set up the experiment. QF, TP, SV, EM performed the experiments. QF, TP, DB, and MD analyzed and interpreted the data. QF, TP, DB, JLG, AD, TS and MD drafted the original version of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Martin Dres.

Ethics declarations

Ethics approval and consent to participate

This study was approved by an ethical committee (ID RCB: 2018-A022311-54) and referenced on ClinicalTrials.gov (NCT03832231). Informed consent was obtained from all patients or their relatives.

Consent for publication

Not applicable.

Competing interests

JLG is a scientific consultant for Supersonic Imagine, Aix-en-Provence, France. MD received personal fees from Lungpacer Medical Inc., Vancouver, Canada. AD reports personal fees from Medtronic, grants, personal fees and non-financial support from Philips, personal fees from Baxter, personal fees from Hamilton, personal fees and non-financial support from Fisher & Paykel, grants from French Ministry of Health, personal fees from Getinge, grants, personal fees and non-financial support from Respinor, grants, personal fees and non-financial support from Lungpacer, personal fees from Lowenstein, outside the submitted work. The remaining authors do not declare any competing interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Additional file 1

. Figure S1: Movie clip of temporal changes in esophageal pressure, gastric pressure, transdiaphragmatic pressure, and diaphragm shear modulus during pressure support ventilation in one patient with a breathing frequency of 12 breaths/min in which a strong correlation (r = 0.81, p = 0.002) was found between changes in transdiaphragmatic pressure (ΔPdi) and changes in diaphragm shear modulus (ΔSMdi) can be found at the following link: https://figshare.com/s/f53dbea5b18c420a1490.

Additional file 2

. Figure S2: Individual correlation coefficients between changes in transdiaphragmatic pressure (ΔPdi) and changes in diaphragm shear modulus (SMdi) can be found at the following link: https://figshare.com/s/c0f06ddf34d5a784ccb5.

Additional file 3

. Figure S3: Movie clip of temporal changes in esophageal pressure, gastric pressure, transdiaphragmatic pressure, and diaphragm shear modulus during pressure support ventilation in one patient with a breathing frequency of 33 breaths/min in which no correlation (r = 0.14, p = 0.643) was found between changes in transdiaphragmatic pressure (ΔPdi) and changes in diaphragm shear modulus (ΔSMdi) can be found at the following link: https://figshare.com/s/fb33c7701fb50c35c98d.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Fossé, Q., Poulard, T., Niérat, MC. et al. Ultrasound shear wave elastography for assessing diaphragm function in mechanically ventilated patients: a breath-by-breath analysis. Crit Care 24, 669 (2020). https://doi.org/10.1186/s13054-020-03338-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13054-020-03338-y

Keywords

Critical Care

ISSN: 1364-8535