How do you monitor a ventilated patient?

• Access through your institution

European Respiratory Review 2018 27: 170101; DOI: 10.1183/16000617.0101-2017

Abstract

Noninvasive ventilation [NIV] is currently one of the most commonly used support methods in hypoxaemic and hypercapnic acute respiratory failure [ARF]. With advancing technology and increasing experience, not only are indications for NIV getting broader, but more severe patients are treated with NIV. Depending on disease type and clinical status, NIV can be applied both in the general ward and in high-dependency/intensive care unit settings with different environmental opportunities. However, it is important to remember that patients with ARF are always very fragile with possible high mortality risk. The delay in recognition of unresponsiveness to NIV, progression of respiratory failure or new-onset complications may result in devastating and fatal outcomes. Therefore, it is crucial to understand that timely action taken according to monitoring variables is one of the key elements for NIV success. The purpose of this review is to outline basic and advanced monitoring techniques for NIV during an ARF episode.

Introduction

Noninvasive ventilation [NIV] is a method of treatment that can be applied in a broad spectrum of indications [1]. NIV for acute respiratory failure [ARF] should be performed in a clinical environment with adequate nurse-to-patient ratios and monitoring. The choice of facility level should be selected according to disease severity and the co-existence of other organ failure. Irrespective of the clinical context, monitoring of patients treated with NIV should be adjusted to the patient's general status and the severity of respiratory insufficiency. These two factors should determine the location of the performance of NIV, since only the proper setting can provide suitable features for adequate monitoring. Monitoring of NIV during ARF is of the utmost importance and can be defined as the real-time evaluation of physiological functions to guide treatment strategies [table 1]. Lives are not saved by monitoring itself; it is the actions taken based upon the monitoring data.

TABLE 1

Monitoring of noninvasive ventilation during acute respiratory failure

An inadequate level of care will have consequences in the likelihood of NIV failure. In fact, close monitoring to detect a favourable response or signs of deterioration is essential in ascertaining NIV success or preventing unnecessary delays in intubation [2].

To better understand the goals of care, a Society of Critical Care Medicine task force developed a categorical approach for NIV support in critical and palliative care [3]. The use of NIV for patients with acute respiratory failure was classified into three groups, as follows. 1] NIV as life support with no preset limitations on life-sustaining treatments; 2] NIV as life support in patients who have chosen to forego endotracheal intubation; and 3] NIV as a palliative measure in patients who have chosen to forego all life support. Attentive monitoring of NIV should be performed to provide optimal quality of NIV and other therapeutic procedures in a patient with no preset limitations. In contrast, it should be kept in mind that excessive implementation of monitoring tools may worsen the effect of NIV used for palliative purposes.

In spite of convincing and unquestionable proof about the efficacy of NIV in acute exacerbations of chronic obstructive pulmonary disease [AECOPD], an observational survey of clinical practice performed in the UK by Roberts et al. [4] demonstrated surprising results. The audit comprised a large group of 9716 patients from 232 units. The results raised significant concerns about the standard of medical management of AECOPD patients. One-third of all eligible patients did not receive NIV, while 11% of those with metabolic acidosis did. Only 5% of all acidotic patients received invasive mechanical ventilation [4]. However, the most striking data showed that mortality was significantly higher in patients treated with NIV than in patients who did not receive NIV with the same level of respiratory acidosis. Unfortunately, data on how many of NIV patients were managed in high-dependency units [HDUs] or intensive care unit [ICUs] were lacking. Even taking into account that the study was not a randomised controlled trial [RCT] and there could be other factors apart from pH which could determine patients' prognosis, it demonstrated clearly that NIV achieves its benefits only in cases of appropriate administration, adequate monitoring and predicted escalation pathway in cases of deterioration [4]. Otherwise, NIV may be harmful, because it can postpone or waive the decision about effective treatment, e.g. invasive ventilation [5].

It is important to stress that there is a “learning curve” impact on the success of NIV [6], so adequate staff training should be implemented and regularly checked [7].

The aim of this article is to review how to monitor NIV in patients hospitalised for ARF and who have no preset limitations on life-sustaining treatments.

Levels of care and monitoring in patients receiving NIV

In most countries, there are three levels of facilities: the ICU, a respiratory intermediate care unit with particular expertise in NIV [respiratory HDU [RHDU]] and the respiratory general ward [table 2] [8]. The efficacy of NIV has been proven in each of these three locations [9].

TABLE 2

Levels of monitoring of patients receiving noninvasive ventilation [NIV]

The use of NIV in the ICU is supported by the strongest scientific evidence. Most of the controlled studies that demonstrated that treatment with NIV decreases mortality were performed in ICUs [10–13]. Nonetheless, the ICU should be recommended as a location for NIV treatment only for severely ill patients, with a high risk of NIV failure: severe respiratory acidosis [pH 40% of patients [74]. The number of asynchronies is correlated with the magnitude of leak and higher pressure support [74]. It was shown that high number of asynchronies is associated with decreased patient comfort, and possibly tolerance to NIV. Asynchronies are not shown to have any influence on gas exchange or any other clinical parameters.

The gold standard for measuring patient–ventilator asynchrony is recording of the electrical activity of the diaphragm and pressure changes in the oesophagus, which requires sophisticated equipment and is invasive. Otherwise, asynchrony can be detected by observation of patient and ventilator rhythm of respiration. The most practical method should be analysis of the pressure and flow waveforms [75], demonstrated in figure 2. Modern portable ventilators allow for online monitoring of pressure and flow waveforms, which make them an appropriate tool of ventilation in a hospital environment. However, clinicians must be aware that observation of respiratory waveforms is not the ideal method of detecting of asynchrony: the patient's effort may not change the flow–time or pressure–time curves, and other factors may influence them, e.g. airway secretions or cardiac oscillations [71]. In a study by Younes et al. [76], 20% of ineffective efforts were not detected by waveform analysis.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2

Examples of patient–ventilator asynchrony [arrows] revealed in pressure and flow waveforms [data from the software ResScan version 5.6.0.9419]. a] Ineffective effort; b] double triggering.

Respiratory waveforms

Observation of pressure–time and flow–time waveforms during NIV can be useful not only for detection of patient–ventilator asynchrony, but also other additional information about the quality of the ventilation: magnitude of leaks, obstruction of airways and I/E ratio. The visualisation of these phenomena makes them easier to spot and to react adequately in terms of titration of the settings of the ventilator. In a multicentre RCT on a cohort of patients with AECOPD, Di Marco et al. [77] showed that titration of ventilator settings on the basis of analysis of respiratory waveforms in real time resulted in more rapid improvement in pH and PaCO2 and better tolerance of ventilation by patients.

Sleep studies

A patient with ARF is usually agitated and vigilant when starting NIV. However, in case of hypercapic coma or in the later phase of treatment, when the patient's condition has improved, ventilation is performed mainly during sleep, when the physiological parameters change and may interfere with the efficacy of ventilation. These disturbances, if not very significant, may be overlooked by nursing staff. Thus, the analysis of the memory of a ventilator with the use of the adequate software can be very useful. It allows for the analysis of overnight ventilation, practically breath by breath, and can reveal events such as excessive leaks, reduction in flow and even desaturations, if pulse oximetry is integrated with the ventilator. As far as we know, there are no data demonstrating clinical benefits of such an approach. However, we recommend the analysis of the ventilator software data after a night of NIV.

Although the acute setting is not an optimal timing for assessment of sleep, Roche Campo et al. [37] performed an interesting study assessing sleep disturbances with the use of 17-h polysomnography performed 2–4 days after starting NIV for hypercapnic ARF. The authors found that sleep disturbances [abnormal electroencephalographic pattern, greater circadian sleep-cycle disruption and less rapid eye movement sleep] were associated with late NIV failure. The practical significance of this finding is unclear; however, this observation undoubtedly broadens the spectrum of the possibility of monitoring patients under NIV. In spite of this interesting study, it seems that sleep studies could be of help mostly in the recovery phase of ARF, especially for detection of breathing-related sleep disorders and making decisions about chronic treatment.

Monitoring of cardiac function under NIV

Clinician applying NIV in order to reinforce the ventilatory pump and improve gas exchange must be aware of its effects on the functioning of the cardiovascular system. These effects must be assessed continuously or periodically, to reveal potential side-effects and manage them adequately, as the improvement of respiratory function can be associated with the impairment of cardiac function. Positive pressures are very effective in decreasing PaCO2, but simultaneously may considerably decrease cardiac output and oxygen delivery to tissues [78].

Positive pressures applied during NIV, in contrast to spontaneous breathing, produce positive intrathoracic pressure [ITP] throughout the respiratory cycle. The cardiovascular consequences of the increased ITP are as follows. 1] Decrease in venous return of right ventricle, which decreases right ventricle preload; 2] increase in pulmonary vascular resistance, which increases right ventricle afterload; 3] increase in central venous pressure; 4] decrease in left ventricle afterload due to decrease in systemic blood pressure; and 5] decrease in left ventricle preload, which decreases left ventricle ejection.

Clearly, some effects may be deleterious, while others are beneficial. Because the level of cardiac output depends mainly on preload, the general effect of NIV will be a decrease in cardiac output. In a cohort of stable COPD patients, ventilation with a mean pressure support of 18 cmH2O reduced cardiac output by 16%, while pressure of 24 cmH2O reduced it by 24% [79]. Application of continuous positive airway pressure ≥15 cmH2O in healthy subjects reduced cardiac output by 20–30% [80]. In patients with AECOPD treated with average pressures, expiratory positive airway pressure 3 cmH2O and IPAP 12 cmH2O, cardiac output and oxygen delivery were decreased by 13% and 8%, respectively [81].

The overall consequence of NIV mostly depends on the underlying cardiovascular conditions. In patients with hypovolaemia or restrictive cardiomyopathy, a compensatory sympathetic response may develop: tachycardia, vasoconstriction, oliguria and retention of water and sodium chloride [82], while in patients with congestive heart failure with fluid overload and hypertension, ITP may improve cardiac function.

Electrocardiography

12-lead ECG should be performed in all patients with ARF, irrespective of the history of cardiac disorders. Then, an ECG trace should be monitored continuously in all patients treated in the HDU or ICU. The monitoring should be continued until constant improvement is obtained. BTS/ICS guidelines recommend monitoring ECGs in all patients with tachycardia >120 beats·min-1, dysrhythmias or possible cardiomyopathy [33].

Blood pressure

Blood pressure is one of the vital signs and should be systematically assessed throughout the treatment with NIV. Hypotension [systolic blood pressure