Article Text
Abstract
Despite level 1 evidence for important benefits of volume-targeted ventilation (VTV), many vulnerable extremely preterm infants continue to be exposed to traditional pressure-controlled ventilation. Lack of suitable equipment and a lack of appreciation of the fact that ‘one size does NOT fit all’ appear to contribute to the slow uptake of VTV. This review attempts to improve clinicians’ understanding of the way VTV works and to provide essential information about evidence-based tidal volume (VT) targets. Focus on underlying lung pathophysiology, individualised ventilator settings and VTtargets are keys to successful use of VTV thereby improving important clinical outcomes.
- volume-targeted ventilation
- bronchopulmonary dysplasia
- tidal volume
- individualized patient care
Statistics from Altmetric.com
Background
Acceptance of volume-targeted ventilation (VTV) in newborn intensive care has been unexpectedly slow, despite sound a physiological rationale for its use and a growing body of evidence for its safety and efficacy.1 The most recent Cochrane meta-analysis that included 16 parallel studies with 977 infants and 4 cross-over trials concluded that VTV, compared with pressure-limited ventilation, offered a number of important benefits with no apparent adverse effects (table 1).2 An earlier meta-analysis by Peng et al had come to similar conclusions.3 While these meta-analyses share the inevitable limitations of combining different modalities of ventilation and study designs, they nevertheless provide a stronger evidence base for VTV than that available for currently preferred ventilation approaches. Yet, while acceptance of VTV is high in Scandinavian countries, Australia and New Zealand,4 it appears to be more spotty in much of the rest of the world5 and especially in USA (Gupta et al, 2017, unpublished).
Although it may be tempting to attribute the reluctance of clinicians to embrace VTV to simple inertia, there are real frustrations and barriers to adopting the paradigm shift that this approach to respiratory support of sick newborn infants represents. The goal of this review is to improve clinicians’ understanding of the way VTV works and the importance of individualising its application to specific clinical situations and individual patients. A major barrier to success appears to be a lack of appreciation of the fact that ‘one size does NOT fit all’. I will review basic principles of VTV, identify some of the barriers and pitfalls in its application and provide evidence-based guidelines for optimal use of VTV in different clinical situations.
What is VTV and how does it work?
Although the rationale for the use of VTV and its operational characteristics have been described in detail in earlier publications,1 6 it is important to briefly review the working principles of VTV and the distinction between VTV and volume-controlled ventilation before exploring the caveats and barriers to its use. Some confusion exists in describing approaches to regulating tidal volume (VT) delivery. Many authors refer to all modalities that seek to control VT as VTV; however, failing to distinguish between volume-controlled ventilation of the ‘adult’ type (VC) and volume-targeted modalities that are specifically designed for newborn ventilation may, in fact, be one of the barriers to its wider acceptance.
VTV is a term that should be reserved for pressure-controlled modalities of ventilation with automatic adjustment of peak inflation pressure (PIP) to target a user-set VT measured at the airway opening. Thus, VTV is fundamentally different from VC modes of ventilation that are widely used in adult and paediatric applications (figure 1). In VC ventilation (also known as volume-cycled ventilation), a user-set VT is introduced into the proximal (ventilator) end of the patient circuit. Pressure rises passively, in inverse proportion to lung compliance, reaching its peak just before exhalation. In larger patients with cuffed endotracheal tubes (ETTs), there is a good correlation between the set VT and the VT that reaches the patient’s lungs. In extremely low birthweight (ELBW) infants, much of the set VT is lost to compression of gas in the circuit and leak around uncuffed ETT,7 necessitating a much higher set VT in order to deliver a physiological VT to the patient. Additionally, because ETT leak fluctuates and the degree of loss of volume to compression varies with PIP, the relationship between set and delivered VT is not constant (figure 2). Use of a separate flow sensor at the airway opening to monitor exhaled VT can overcome some of these limitations, allowing the operator to manually adjust the set VT to deliver the desired VT to the infant’s lungs. However the ETT leak is usually variable necessitating frequent monitoring and adjustment, thus making this approach less attractive. An alternate, less desirable approach that is still used by some clinicians is to rely on clinical assessment of chest rise and breath sounds to select the set VT, with subsequent adjustments based on blood gas measurement. However, there is evidence that clinical assessment of adequacy of VT is poor8 and therefore this approach cannot be recommended. Despite these limitations, VC has been shown to be feasible, at least under research conditions, even in small preterm infants, when a flow sensor at the airway opening is used and appropriate adjustments to the set VT are made to achieve the desired exhaled VT.9
Various modalities of VTV were developed specifically to address the limitations of VC ventilation when applied to ELBW infants. Volume guarantee (VG) on the Babylog 8000+and VN 500 ventilators (Draeger Medical GmbH, Lubeck, Germany) is the most thoroughly studied of the VTV modalities and therefore there is a stronger evidence base for specific recommendations for its use. However, the basic control algorithm has increasingly been adopted by other manufacturers, making for greater generalisability of the available data. In VG, the user chooses a target VT and a pressure limit up to which the ventilator operating pressure (the working PIP) can be adjusted. The ventilator then compares the exhaled VT of the previous inflation and adjusts the PIP up or down to try to achieve the set VT. Exhaled VT is used to minimise artefact resulting from air leakage around the uncuffed tube, because leak is always greater during inflation.10 The pressure increase from one inflation to the next is limited to avoid oscillation in the system that could lead to excessive VT. Consequently, with large changes in compliance or patient inspiratory effort, several inflations are needed to reach the target VT. An important safety feature designed to avoid delivering an excessively large VT opens the expiratory valve and terminates inflation if the VT of the inflation exceeds 130% of the target (corrected for leak). The algorithm is geared towards slower adjustment for low VT and faster response to excessive, potentially dangerous VT. Autoregulation of inflation pressure leads to automatic weaning from mechanical ventilation and is one of the first examples of closed-loop ventilation technologies.1 Because weaning occurs in real time, rather than intermittently in response to blood gases, VTV achieves faster weaning from mechanical ventilation.2 Different manufacturers have implemented VTV modalities in different ways and it is important to be aware of some of the limitations related to specific devices. For example, pressure regulated volume control (PRVC) in the Servo-i ventilators (Maquet, Solna, Sweden) uses a control algorithm similar to VG, but the VT measurement used for PIP adjustment is performed at the ventilator end of the circuit. VT measurement at that site has been shown to grossly overestimate the true VT that enters the lungs.7 This issue has been corrected in the new Servo-n and Servo-u series, which should make PRVC behave much more like VG.
Barriers to successful use of VTV
The most obvious barrier is lack of suitable ventilators, especially in the USA, where many newborn intensive care units are equipped with ventilators primarily designed for adults (so-called universal ventilators), that have VC modes, but do not offer effective VTV modes suitable for ELBW infants. Even with the best available equipment, successful introduction of a new approach to ventilation can be challenging and should be undertaken deliberately with extensive education and planning.
Unfortunately, often the first attempt at using a new modality occurs when a very sick baby is failing on current support prompting a trial of the new approach, sometimes before everyone is ready. When initial attempts with a new approach disappoint, it is natural to sound the retreat and go back to what has always ‘worked’, that is, pressure-controlled ventilation. Change does not come easily; it is human nature to hold on to the approach that is familiar and comfortable. For this reason, it is important to have a local champion or ‘super-user’ who becomes comfortable with the modality and will be the resource person to troubleshoot and educate when needed. Additionally, it is best to gain experience with straightforward simple patients. As experience and confidence grow, more difficult patients can then be managed effectively.
Despite previous publications on the topic,11 12 lack of knowledge regarding the appropriate VT targets for a given clinical scenario and a good understanding of the patient-ventilator interactions in VTV modes are the other major barriers to success (Gupta et al, 2017, unpublished) and will be reviewed in the following paragraphs.
Practical guidelines for VTV
Initiation of VTV
Synchronised modes that support every patient breath (assist/control (A/C) or pressure support ventilation) are preferred when using VTV.13 The recommended VT targets are based on A/C; synchronised intermittent mandatory ventilation requires a slightly larger VT for the same alveolar minute ventilation,14 because fewer breaths are supported and volume-targeted. The choice of appropriate VT is the key to success and depends on the infant’s size, postnatal age and underlying disease process. One size truly does not fit all babies. Table 2 lists appropriate VT/kg and initial PIP limit for various conditions. The larger VT/kg requirement in the smallest infants is due to the proportionally larger impact of instrumental dead space of the flow sensor, which ranges from 0.7 mL to about 1.1 mL, depending on the device and sensor used.15 Infants with bronchopulmonary dysplasia (BPD) or meconium aspiration need larger VT/kg because of increased alveolar dead space due to air-trapping and heterogeneous lung inflation, which results in wasted ventilation.16 Older former preterm infants with evolving or established BPD need larger VT/kg as well, due to a combination of increased anatomical and alveolar dead space.17 Many clinicians believe that infants with pulmonary hypoplasia due to congenital diaphragmatic hernia (CDH) should be ventilated with correspondingly small VT/kg. However, observational data suggest otherwise18 and reflect the fact that, regardless of the degree of lung hypoplasia, the CO2 production/kg in infants with CDH is similar to babies without CDH and therefore they need similar alveolar minute ventilation. Although a faster rate can increase alveolar minute ventilation to a degree, rapid shallow ventilation is inefficient because it results in progressively higher dead space:VT ratio. When the VT approaches dead space volume (approximately 3 mL/kg with ETT and flow sensor) a faster rate can no longer maintain adequate alveolar minute ventilation.
When electively switching from pressure-controlled mode to VTV in a patient with satisfactory gas exchange, the simplest approach is to select a target that matches the average VT measured prior to the change-over. When initiating VTV immediately after intubation, the target VT should be selected based on the infant’s size, age and lung pathology (see table 2). Inflation pressure limit should initially be set 3–5 cm H2O above the level estimated to be sufficient to achieve a normal VT. If the target VT cannot be reached with this setting, increase the pressure limit until the desired VT is reached. If more PIP is needed than anticipated, it is important to make sure the ETT is not kinked, malpositioned in the mainstem bronchus or obstructed on the carina. Significant volutrauma and/or air leak could result from failure to recognise endobronchial intubation. Pressure limit is subsequently adjusted to be about 25% above the current working pressure and adjusted periodically as lung compliance improves and working pressure comes down. If the ventilator is unable to reach the target VT with the set inflation pressure limit, an alarm will sound. This serves as an early warning system that should prompt an evaluation of the reason for this change, for example, atelectasis, pneumothorax, ETT malposition or abdominal distention with upward pressure on the diaphragm.19
Subsequent adjustment/weaning
The suggested initial VT settings are typical values, useful as a starting point. However, as is true for all physiological variables, there is considerable variation between similar patients. Patient activity, CO2 production and presence of a base deficit for which the baby is attempting to compensate will affect the VT requirement for an individual patient. It is therefore essential that the infant’s response to the initial settings is promptly evaluated at the bedside and adjustments are made, if needed, even before a blood gas measurement is obtained. One of the important benefits of avoiding routine use of sedative and paralytic medications is the ability to observe the infant’s response to initial ventilator settings unobscured by drugs. Careful observation of the patient’s respiratory effort and rate, as well as assessment of the flow and pressure curves on the ventilator display will provide clues about the appropriateness of the selected VT for this particular patient. Transcutaneous CO2 monitoring, when available, can provide additional indication of appropriateness of alveolar minute ventilation. With actively breathing infants, the displayed values will fluctuate, therefore observations should be made over a number of ventilator cycles.
When ventilator support is adequate, the infant should be breathing comfortably without distress or tachypnoea. Persistence of significant tachypnoea and retractions indicate inadequate support and a need for a larger VT, especially when coupled with relatively low working PIP and measured VT that often exceeds the target VT. To appreciate this concept it is essential to recognise that the measured VT is the result of transpulmonary pressure, the combined positive inflation pressure from the ventilator and the negative inspiratory pressure of the infant. In the extreme case of inadequate VT setting, the PIP may be reduced to or be near the level of positive end-expiratory pressure (PEEP). This occurs because, as long as the patient is able to spontaneously generate a VT that exceeds the VT target, the algorithm will continue to lower the PIP. Eventually, the infant becomes exhausted and can no longer sustain the effort. The ventilator takes over at the set backup rate and, in the absence of the infant’s contribution to transpulmonary pressure, the working pressure will return to the level needed to reach the target VT. As the partial pressure of carbon dioxide (PCO2) rises and the pH drops, the infant will again respond to his/her respiratory drive and attempt to restore normal pH. This cycle leads to fluctuation of PCO2 and intermittent drop in mean airway pressure that may contribute to intracranial haemorrhage and atelectasis. On the other hand, if the infant is generating VT above the target value and is comfortable without retractions, tachypnoea or increasing oxygen requirement, despite the low working PIP, it indicates improved respiratory status and that extubation is in order.
It is important to recognise that pH, not PCO2 is the primary driver of respiratory control; the physiological response to metabolic acidosis is hyperventilation. Thus PCO2 values must be interpreted in the context of pH. Moderate degree of metabolic acidosis is common in ELBW infants with immature renal tubular function and high protein intake in the first days after birth. These infants need a relatively lower PCO2 target to maintain a reasonably normal pH, avoid excessive work of breathing and loss of lung volume recruitment when the PIP falls intermittently to, or near, the level of PEEP. Failure to appreciate these interactions and inappropriate weaning of VT based on PCO2 alone is a common reason why VTV ‘does not work’. Available evidence indicates that perivascular pH, not PCO2 primarily controls cerebral vascular tone,20 suggesting that mild stable hypocapnia with a neutral or mildly acidotic pH is not likely to have adverse effects. However, PCO2 much below 35 torr (approximately 4.5 kPa) should be avoided, especially if coupled with an alkalotic pH, as should rapid fluctuations in PCO2, which have been shown to be the strongest predictor of severe intraventricular haemorrhage.21
Another challenge for users of VTV is leakage of air around uncuffed ETT, which is present to some degree in most intubated ELBW infants.22 23 The problem tends to increase with time as the immature tissues of the trachea and larynx stretch due to positive pressure ventilation (acquired tracheomegaly)24 and as the infant grows. Even though the exhaled VT is less subject to leak-related underestimation of VT, when the leak exceeds 35%–40%, VT measurement is no longer accurate, potentially resulting in inadvertent hypocapnia because a proportion of the exhaled gas escapes around the ETT during expiration and is thus not measured by the flow sensor as exhaled VT.25 When the microprocessor detects a below-target VT it increases PIP, resulting in an excessive VT. For this reason, many VTV devices cannot be safely used when the ETT leak approaches this limit. The clinician must choose one of two options: replace the ETT electively to eliminate the leak, or abandon VTV in favour of pressure-controlled ventilation, which is not affected by ETT leak. Because of this issue, there has been some interest in reassessing the use of cuffed ETT in newborn care.26 However, some of the newest infant ventilators now have the ability to calculate an estimated value for the true VT even in the face of a very large leak,23 thus avoiding this common problem and rendering cuffed tubes unnecessary. Nonetheless, unless extubation is imminent, the ETT should be changed to a larger size if the leak consistently exceeds 50% in order to provide a lower resistance airway for the infant during the weaning process. Understanding the capabilities your ventilator is therefore crucial to optimal care.
The process of weaning also causes some confusion. The effective closed-loop system of VTV is counterintuitive to some practitioners who are accustomed to manual adjustments of ventilator settings. This sometimes leads to inappropriate lowering of target VT in an effort to wean the patient off the ventilator. However, the physiological VT required by the patient does not decrease (over time it may actually increase); what comes down is the pressure required to achieve that VT as compliance of the respiratory system improves and the infant breathes more effectively. In fact, it is important to remember to adjust VT for weight gain in infants who remain ventilated for extended periods. Decreasing VT target below the patient’s physiological need will only increase the work of breathing27 and may delay successful extubation.
Conclusion
Faced with level 1 evidence of important benefits of VTV, it is hard to justify continuing to expose infants to pressure-controlled ventilation. The way forward is for us is to be willing to abandon our comfort zone and embrace the paradigm shift that VTV represents. The transition should be undertaken deliberately and only after much training and appraisal of the available literature. Focus on underlying lung pathophysiology, individualised ventilator settings and VT targets are keys to success. A formal ventilation protocol is an effective way to implement respiratory support, especially when transitioning to a new approach.28
Acknowledgments
The author thanks Dr Kabir Abubakar for his critical review of the manuscript.
References
Footnotes
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests MK received research support and lecture honoraria from Draeger Medical GmbH. The company had no input into the content of this paper.
Provenance and peer review Commissioned; externally peer reviewed.
Linked Articles
- Fantoms