The diaphragm is the primary muscle of inspiration. It is a thin dome-shaped muscle that inserts into the lower ribs, xiphoid process, and lumbar vertebrae. It runs parallel to the rib cage in the zone of apposition before curving along the inferior pleural surface (Mead 1979). Shortening of the diaphragm muscle fibres during inspiration results in a piston-like action, drawing the lungs downward and forcing the lower chest wall outward (de Troyer and Loring 2011; Gauthier et al. 1994). Downward excursion of the diaphragm varies from about 1 cm during normal tidal breathing to over 1-2 cm during deep inspiration. Diaphragm function is a crucial determinant of respiratory capacity in respiratory failure; injury to the diaphragm in the intensive care unit (ICU) has a substantial impact on patient outcome (Dres et al. 2017; Goligher et al. 2018). Recent advances in ultrasound imaging enable clinicians to more feasibly assess diaphragm function and potentially protect the diaphragm during mechanical ventilation.
Diaphragm ultrasound techniques
Two techniques to image the diaphragm using ultrasound have been described. First, a low-frequency phased array transducer can be used to quantify downward excursion of the diaphragm in M-mode. The probe is positioned subcostally at the mid-clavicular line and angled upward, so that the ultrasound beam projects perpendicular to the dome of the diaphragm—the probe may be angled slightly medially if needed. Figure 1 shows an example of the images obtained with this approach. Normal values of diaphragm excursion during various inspiratory manoeuvres have been reported and the technique has excellent reproducibility (Boussuges et al. 2009). This evaluation is only valid in the absence of ventilatory support, as thoracic insufflation by applied positive pressure will also lead to downward diaphragm excursion.
The second approach measures diaphragm cross-sectional thickness in the zone of apposition. A high-frequency linear array transducer can be employed to measure the cross-sectional thickness of the diaphragm (B-mode) and the percentage change in thickness from expiration to inspiration, referred to as the thickening fraction (TF) (M-mode) (Figure 2) (Sarwal et al. 2013; Matamis et al. 2013). This technique has been described extensively and is well validated (Cohn et al. 1997). The probe is positioned in the 8th or 9th intercostal space between the mid- and anterior axillary lines in the zone of apposition. The diaphragm is visualised as the space intervening between the pleural and peritoneal lines. Both thickness and thickening fraction measurements show excellent reproducibility in ventilated patients (Goligher, Laghi et al. 2015).
Application: monitoring diaphragm function
Both measurements can help to diagnose diaphragm dysfunction. Reduced excursion during various inspiratory manoeuvres is an established diagnostic criterion for dysfunction (Lerolle et al. 2009). Serial measurements of diaphragm thickness can reveal acute decreases in diaphragm thickness, indicative of rapid muscle atrophy. As the thickening fraction is related to the degree of diaphragmatic shortening during inspiration, it correlates with the changes in lung volume (Wait and Johnson 1997; Wait et al. 1989), inspiratory pressure development (Ueki et al. 1995; Dubé et al. 2017) and work of breathing (Vivier et al. 2012; Umbrello et al. 2015). As a result, a reduced maximal thickening fraction is a marker for diaphragm weakness. The validity of this technique to diagnose diaphragm weakness is crucially dependent on obtaining maximal volitional inspiratory effort. Using ultrasound to diagnose diaphragm weakness may help to guide the management of patients with difficult weaning from mechanical ventilation (Vorona et al. 2018; Heunks et al. 2015).
The thickening fraction measured during a spontaneous breathing trial (SBT) predicts the weaning outcome (DiNino et al. 2014; Zambon et al. 2016). In contrast to commonly used parameters like the rapid shallow breathing index (RSBI), the thickening fraction specifically reflects diaphragmatic action. This way, diaphragm weakness is not masked by increased accessory muscle activity during the SBT, which may explain the excellent predictive results of this parameter (DiNino et al. 2014).
Application: monitoring diaphragm activity
When the diaphragm is activated during tidal breathing, it shortens and hence thickens. This tidal thickening is visualised on ultrasound as diaphragm thickening. Hence, diaphragm thickening fraction measurements can be used as a marker of inspiratory effort and diaphragm contractile activity, even under mechanical ventilation. Diaphragm activity during mechanical ventilation is a key determinant of the risk of diaphragm injury. Both inappropriately low or excessive levels of diaphragm activity predict the risk of deleterious changes in diaphragm thickness and poor clinical outcomes (Goligher et al. 2018). Insufficient activity leads to rapid disuse atrophy (Levine et al. 2008; Goligher, Fan et al. 2015; Goligher et al. 2018); some evidence suggests that excessive respiratory muscle loading may result in inflammation, injury and dysfunction (Reid et al. 1994; Jiang et al. 2012), although it is unclear whether this happens frequently in the clinical setting.
Monitoring diaphragm activity by ultrasound might be used to guide the titration of ventilatory support. By targeting support levels producing relatively normal levels of respiratory effort (thickening fraction in the range of 15-30%), diaphragm atrophy and injury might be prevented. This concept is referred to as diaphragm-protective ventilation (Heunks and Ottenheijm 2018). A recent study found that the length of ICU stay, the duration of ventilation, and the risk of complications of respiratory failure (prolonged ventilation, reintubation, tracheostomy or death) were minimised when patients’ thickening fraction averaged 15-30%—similar to that of healthy subjects breathing at rest—during the first 3 days of ventilation (Goligher et al. 2018). This may represent the optimal level of spontaneous breathing during mechanical ventilation to protect the diaphragm from injury (Tobin et al. 2010). Ultrasound measurements can detect adequate, insufficient or excessive inspiratory effort (see Figure 2) (Umbrello et al. 2015).
Ultrasound can also be used to diagnose patient-ventilator dyssynchrony (Thille et al. 2006; Matamis et al. 2013), as it provides real-time information of contractile activity. Recognising and treating dyssynchrony may help to improve outcomes, as frequent dyssynchrony strongly predicts poor clinical outcomes (Vaporidi et al. 2016; Blanch et al. 2015).
Future of diaphragm ultrasound
Several emerging ultrasound techniques have been evaluated for their potential to further evaluate the diaphragm. These include speckle tracking, which uses naturally occurring speckle patterns to assess tissue deformation and motion. Originating from cardiac imaging, it has been used to measure diaphragm muscle strain (Ye et al. 2013; Hatam et al. 2014). It may provide a novel noninvasive technique for bedside evaluation of respiratory workload, but further research is required to validate this technique.
Changes in diaphragm tissue structure apart from atrophy might be detected by assessing changes in diaphragm echodensity, which reflects the development of muscle oedema, inflammation and necrosis (Puthucheary et al. 2015). These provide insights on muscle quality apart from muscle bulk. This technique has been described for imaging quadriceps muscle; it has not yet been described for diaphragm ultrasound. Other ultrasound techniques including 3D imaging and elastography have yet to find their way into the ICU and clinical practice.
Mechanical ventilation can injure the diaphragm, resulting in significant morbidity and mortality in critically ill patients. Optimising diaphragm activity during mechanical ventilation is important to mitigate diaphragm dysfunction. Ultrasound has the capability to detect structural changes in the diaphragm (thickness and quality), assess inspiratory effort, diagnose diaphragm weakness, predict liberation from ventilation, and possibly detect patient-ventilator dyssynchrony. Widespread implementation of diaphragm ultrasound in clinical practice has the potential to improve outcome by optimising ventilator management based on patient effort and synchrony, guiding assessment of the difficult-to-wean patient, and informing decisions about liberation from mechanical ventilation. Diaphragm ultrasound can also provide crucially important mechanistic insights for research studies: measurements of thickness, echodensity, thickening fraction, and maximal thickening may provide clinically relevant physiological biomarkers of disease and treatment effects. Future trials are required to explore how to facilitate diaphragm-protective ventilation and whether this new paradigm for ventilation improves the outcome for ventilated patients.