div epub:type=”chapter” role=”doc-chapter”>
11. Ventilator-Induced Diaphragm Dysfunction
11.1 Definition and Brief History
It was Vassilakopoulos and Petrof who first proposed the concept of ventilator-induced diaphragm dysfunction (VIDD) in 2004, which refers to the loss of diaphragmatic force-generating capacity that is specifically related to the use of mechanical ventilation . As early as 1988, it has been recognized that the phenomenon of diaphragm atrophy or failure of normal growth in ventilated neonates, which may contribute to difficulties in weaning . In 1994, the loss of diaphragm mass and a large reduction in maximal diaphragmatic-specific force production induced by 48 h of controlled mechanical ventilation (MV) were documented in rodent study  and supported by the following primate study, which also showed the decrease of diaphragmatic endurance .
Although the numerous animal studies indicated that MV promotes VIDD repeatedly, it was highlighted until 2008, Levine has first shown the evidence of rapid disuse atrophy of diaphragm fibers in human beings with prolonged MV . Great efforts have been put in to explore the mechanisms responsible for VIDD and trying to identify the biological target for drug intervention, all of which promote the birth of the strategy of diaphragm-protective mechanical ventilation.
11.2 Anatomy of Diaphragm
The diaphragm is the musculotendinous boundary between the thoracic cavity and abdominal cavity, which is thin (2–3 mm in healthy adults), dome-shaped, and composed of two parts: noncontractile central tendon that separates the right and left sides and extends to the dome of each hemidiaphragm, and the contracting muscle fibers: the costal and crural muscle groups, distinguished with each other in composition and function too. The costal muscle group is thin and forms the diaphragmatic leaflets, the contraction of fibers will lead to flattening of the diaphragm and lowers the ribs. Although the crural muscle groups are thicker, it contributes little to the displacement of the diaphragm. The continuation of the medial tendinous margins of the crura forms the median arcuate ligament anterior to the aortic hiatus. The aorta, inferior vena cava, thoracic duct, esophagus, vagus, and phrenic nerves, traverse the diaphragm through aortic, caval, and esophageal apertures separately.
The left and right sides of the diaphragm are innervated by the ipsilateral phrenic nerves, which originates from the anterior branches of the cervical spinal cord of segments 3–5 and traverses the neck and mediastinum before inserting into the diaphragm centrally. Each nerve divides into four trunks that innervate the anterolateral, posterolateral, sternal, and crural portions of the diaphragm on that side.
The outer rim of the diaphragmatic muscle is innervated laterally from the thoracic spinal cord of 7 through 12 and recently, the study found that the crural diaphragm also receives innervation from the vagus nerve . In terms of blood supply, it mainly comes from the phrenic artery below and the pericardiophrenic arteries above the diaphragm.
About 55% of the diaphragm fibers are of the slow-twitch, oxidative type (highly resistant to fatigue), 25% are of the fast-twitch, oxidative glycolytic type (relatively resistant to fatigue), and the remaining 20% of the fibers are of the fast-twitch glycolytic variety (susceptible to fatigue) .
11.3 The Physiological Function of the Diaphragm
The respiratory system consists essentially of two vital parts: the gas exchanging organ (lung) and the pump that drive the ventilation (respiratory muscles). The diaphragm plays an important role in respiratory mechanics. As the most powerful of the respiratory muscles, the diaphragm performs up to 75% of the work of breathing alone and assisted by accessory inspiratory muscles (e.g., scalene, parasternal portions of the internal and external intercostal muscles, and sternomastoids). The simultaneous contraction of inspiratory muscles elevates and expands the upper ribcage, increasing negative intrapleural pressure, and driving air into the lungs. Apart from the role of the respiratory pump, the diaphragm also serves as a mechanical barrier between the abdominal and thoracic cavities and maintains the pressure gradient between the cavities.
11.4 Ventilator-Induced Diaphragm Dysfunction
Although mechanical ventilation is life-saving in respiratory failure supporting, prolonged ventilator use associated with difficult weaning. Among the variety of impact factors, diaphragm weakness is believed to be the leading cause . Compared with assist modes that allow the respiratory muscles to remain active, controlled mechanical ventilation increases the risk of diaphragm dysfunction by abolishing the effort of inspiration. We will discuss the two most important reasons: ventilator-induced diaphragm atrophy and contractile dysfunction in the following.
11.5 Ventilator-Induced Diaphragm Atrophy and Responsible Mechanisms
Numerous animal and human experiments consistently demonstrate that prolonged MV promotes a reduction in diaphragm mass, a surrogate of diaphragm atrophy. With high anatomy and fiber type composition similar to human beings, the rat is selected as the most frequently used animal model to study VIDD. Studies showed that 10–15% reduction in the cross-sectional area of rat diaphragm fiber types (specifically, type I, type IIa, and type IIx/b) happened, even as less as 12 h with fully controlled ventilation mode [9, 10], and 30% reduction following 18–24 h of prolonged ventilatory support [11, 12].
In human studies, Levine demonstrated that 18–69 h of fully controlled MV results in the magnitude of diaphragmatic atrophy (about 50% reduction in fiber cross-sectional area) of both type I (slow) and type II (fast) muscle fibers in the costal diaphragm. This finding was confirmed by the following studies, not only limited in controlled ventilation  but also in partial support mode . On average, a rate of 6% loss of diaphragm thickness per ventilated day was detected .
Keeping the balance of protein degradation and synthesis is the basis of maintaining skeletal muscle fiber size. During controlled MV, increased proteolysis and depressed protein synthesis developed rapidly in the diaphragm fibers, all of which contribute to the net loss of protein and diaphragm fiber atrophy [15, 16].
11.6 Decrease of Protein Diaphragmatic Synthesis
In rat studies, controlled MV promotes a rapid decrease of diaphragmatic protein synthesis, both mixed and myosin heavy chain [16, 17]. On the contrary, partial support ventilation does not promote a significant decrease in diaphragm protein synthesis compared with spontaneous breathing . We have known that keeping the skeletal muscle contractile activity is essential to maintain protein synthesis. During the controlled MV, the initial decrease of diaphragm protein synthesis was likely caused by decreased protein translation, which enrolled the Akt/mammalian target of rapamycin (mTOR) signaling pathway .
11.7 Increase of Diaphragmatic Proteolysis
It is reasonable to speculate that increased proteolysis, other than decreased protein synthesis, is responsible for the rapid diaphragm atrophy which occurs as early as 12 h after controlled MV. The increased activity of four major proteolytic systems was well demonstrated, i.e., macroautophagy (autophagy), calpains, caspase-3, and the ubiquitin-proteasome system . In partial support MV, conflicting results in limited studies may be originated from the differences in the levels of ventilatory support [17, 20].
Pharmacological inhibition of calpain and caspase-3 activity in rats decreases the rate of proteolysis and offers protection against MV-induced diaphragmatic atrophy [24, 25], and there is a regulatory cross-talk between calpain and caspase-3 activity . It is promising for medication therapy in diaphragm protection during MV.
11.8 Ventilator-Induced Diaphragm Contractile Dysfunction and Responsible Mechanisms
Not only the atrophy but also the contractile dysfunction is prevalent in diaphragm with prolonged MV, which only occurs at the level of the peripheral muscle, independent with the phrenic nerve conduction and neuromuscular transmission . This reduction of maximal diaphragmatic-specific force production is rapid and time-dependent, about 15–20% with 12 h, and 50% with 48 h of controlled MV [3, 27]. This could be alleviated by keeping spontaneous breathing during controlled MV , ventilated with assisted modes [29, 30], or short periods of bilateral phrenic nerve stimulation (10 min/h) .
Aging, neuromuscular blockers, and/or glucocorticoids are independent risk factors of diaphragm dysfunction. Particularly in severe ARDS patients, nondepolarizing neuromuscular-blocking agents are commonly used to minimize patient agitation, improve ventilator-patient asynchrony, and reduce oxygen consumption. Both skeletal muscle and diaphragm weakness are anticipated and confirmed [32, 33]. Corticosteroids are useful in treating lung underlying disease or inflammation; however, steroid-induced myopathy of both limb muscles and respiratory were well documented [34, 35]. Conflicting results still exist in terms of VIDD, and it is unclear whether the species or mode of MV selected were responsible for this difference [36, 37].
The basic functional element of diaphragm muscle fibers is the sarcomere, and the fundamental unit to generate force production is the interaction between the sarcomeric proteins (actin and myosin). The relationship between cytosolic free calcium, cross-bridge attachment/cross-bridge cycling rate, and sarcomere length determined the mechanical properties of diaphragm muscle fibers . Although the precise mechanisms responsible for diaphragm contractile dysfunction caused by MV remains unclear, oxidative stress, by decreasing calcium sensitivity , or promote calpain activation , combined with the loss of myosin heavy chain protein , contribute to the damage of sarcomere structure. There is a linear relationship between the prevalence of the ultrastructure damage and the time on the ventilator. Studies shown that the average magnitude of sarcomeric injury varied from 10% of the total fiber area for patients ventilated 24 h, to 20% for ventilated 50 h, and 30–80% for ventilated 100–249 h .
11.9 Other Mechanisms Responsible for VIDD
Although the disuse atrophy caused by the suppression of inspiratory effort is the main culprit, other mechanisms may contribute to VIDD either. Concentric or eccentric load-induced injury, which refers to the acute diaphragm injury, inflammation, and weakness, occurs when the diaphragm contract against excessive load during muscular shorting or lengthening [41–45] and heightened in the existence of sepsis and systemic inflammation.
During the ventilation of patients with acute exacerbation of the chronic obstructive pulmonary disease, insufficient assist expose diaphragm to the high load and induce a rapid rise in diaphragm thickness, which believed to be a unique form of injury (i.e., inflammation and fiber swelling), instead of the reflection of hypertrophy. In patients with decreased end-expiratory lung volume (EELV) and atelectasis, the diaphragm contract during expiration trying to maintain lung volume, also known as “expiratory braking” . In the case of certain types of patient-ventilator asynchronies, such as ineffective efforts, reverse triggering, and premature cycling, the diaphragm contracts even as lung volume decreases leading to eccentric myotrauma. Additionally, the fiber length was reduced with the application of positive end-expiratory pressure (PEEP). Decreased PEEP level abruptly such as withdrawn from ventilator may lead to the diaphragm overstretched and impaired its performance [47, 48].
11.10 VIDD: Evidence for Impact on Patient Outcomes
Among the numerous etiologies, the depressed strength and endurance of diaphragm play a crucial role in difficult weaning. When critically ill patients were ready to extubate, 63% of them had diaphragm dysfunction, almost twice as frequently as limb muscle weakness . Compared with the patients that are successfully weaned, the inspiratory muscle endurance, the maximal inspiratory pressure generation, and diaphragmatic contractile function decreased in patients that experience prolonged weaning [49–51]. Diaphragm weakness at the time of extubation increased the risk of ICU readmission and the risk of mortality within the year after ICU discharge [52, 53]. Diaphragm dysfunction should be suspected in the setting of difficult weaning or repeated episodes of respiratory failure. On the other hand, inspiratory muscle training helped the patients who failed repeated weaning attempts to remove from the ventilator successfully [54–56].
11.11 Diaphragm-Protective Mechanical Ventilation
The harmful effects of mechanical ventilation on the lungs are well recognized, and ventilator-induced lung injury (VILI) directly lead to the development of lung-protective ventilation strategies. After decades of research on ventilator-induced diaphragm dysfunction, and deep understanding of the various potential mechanisms, Heunks proposed the concept of diaphragm-protective mechanical ventilation in 2018, highlight the priority of this new approach to ventilatory management .
Maintaining the appropriate level of inspiratory effort is paramount in the performance of diaphragm-protective mechanical ventilation, by avoiding over- and under-assist combined with titrating the dose of sedatives carefully. This should be monitored and targeted as early as possible since the VIDD occurs within mere hours of initiating ventilation. Partial support modes (e.g., pressure support, neutrally adjusted ventilatory assist, proportional assist), other than controlled modes, may help to alleviate disuse atrophy; however, insufficient assist should be avoided.