Chapter 85: Controversies: Noninvasive Ventilation at the End-of-Life—Useful or Not?

Noninvasive ventilation (NIV) has been well described as effective in different patient populations, for example, hypercapnic respiratory failure due to exacerbations of chronic obstructive pulmonary disease (COPD),¹ hypoxic respiratory failure in immunocompromised hosts,² or cardiogenic pulmonary edema in the absence of acute coronary ischemia,³ to cite the most common indications.

Conceptually, the use of noninvasive ventilation can be divided into the following three categories⁴:

1. NIV as a part of “full-code” treatment (life support without preset limits)

2. NIV in patients with do-not-intubate orders (life-support with preset limits)

3. NIV as a comfort measure in patients at the end-of-life (NIV ensuring comfort while dying)

Each category has specific goals of care, response to failure, and main points to communicate with the patient and/or family. Categories 2 and 3 can be defined as palliative NIV.⁵

The goals of NIV in patients in category 1 are to alleviate symptoms of respiratory distress, improve oxygenation and/or ventilation, avoid intubation, and reduce the risk of mortality. Endotracheal intubation is performed if necessary.

Patients in category 2 are those who decline endotracheal intubation or patients in whom clinicians feel that intubation would not meet the goals of care. In this group, the use of NIV achieves the same goals as it does in category 1, except that endotracheal intubation is not an option in cases where NIV is ineffective.

The only purpose of NIV in category 3 is symptom palliation and patient comfort.

Respiratory distress is one of the most common symptoms seen in patients approaching the end-of-life. It leads to restrictions in quality of life and increases anxiety and fear.⁶

Terminal dyspnea is a manifestation of an irreversible process, such as carcinomatous lymphangitis in malignant diseases or advanced degenerative neuromuscular disease (amyotrophic lateral sclerosis).

The vast majority of patients with terminal cancer experience symptoms of respiratory distress at some point during the last 6 weeks of life, and they commonly report significantly increased dyspnea during the last two weeks.⁷ In patients with a noncancer terminal diagnosis, such as COPD or chronic heart failure (CHF), the severity of respiratory distress can be greater; however, the severity remains relatively stable until death.

The mechanism underlying the relief of respiratory distress through noninvasive ventilation remains the same in all clinical situations.⁸ NIV reduces the work of breathing by increasing transpulmonary pressure and reducing inspiratory muscle workloads. Gas exchange is improved by increasing alveolar ventilation, functional residual capacity, opening collapsed alveoli, reducing shunts, and improving the ventilation/perfusion (V/Q) ratio. Altogether, these mechanisms result in a lower respiratory rate, reduced CO₂ retention, and an overall improvement of symptoms of respiratory distress.

Recently, studies focused on the use of palliative noninvasive ventilation have provided descriptive data, new evidence, and qualitative appraisals of palliative NIV.

A prospective cohort study by Azoulay et al.⁹ reported outcomes of patients undergoing NIV in the context of “do-not-intubate” (DNI) orders (ie, category 2). A DNI order is present in about 20% of all patients receiving NIV in intensive care units (ICUs). A substantial hospital survival of 56% was observed, which was most obvious in the COPD patient subgroup. Importantly, for those who survived up to 90 days, health-related quality of life did not significantly change compared to baseline. Moreover, anxiety, depression, and posttraumatic stress disorder (PTSD)-related symptoms in patients and their families were similar to those seen when NIV was used in category 1 (full code) patients. These recent data, together with previously published reports,^10,11 thus support the use of NIV in this clinical context.

A study by Nava et al.¹² that focused on patients with end-stage cancer (ie, category 3 patients) assessed the acceptability and effectiveness of NIV versus oxygen therapy in decreasing dyspnea and its effect on the use of opioids. The study showed that NIV is faster and more effective compared to oxygen in reducing dyspnea and offers the potential to reduce the dose of opioids. Hospital mortality was similar in both groups. NIV was well accepted and well tolerated, again, with the best response in cases of hypercapnic respiratory insufficiency.

Finally, NIV has been evaluated in patients suffering from terminal phase motor neuron disease who are at the end-of-life. A qualitative study by Baxter et al.¹³ revealed important variations in patient wishes regarding the use of NIV toward the end-of-life and also a degree of uncertainty concerning NIV management among healthcare teams. Nevertheless, end-of-life use of NIV was generally perceived as beneficial, allowing a more peaceful end-of-life, free of choking or struggling to breathe during the final moments.

Even if current research findings support the use of noninvasive ventilation in palliative situations (categories 2 and 3 of the conceptual framework), important questions/objections have been raised, concerning possible discomfort and unnecessary prolongation of the terminal phase of life.^14,15 Some aspects of the current controversies are summarized in Table 85–1.

Prior to NIV Initiation

Palliative noninvasive ventilation, incorporated into a strategy of continuous patient care, should be discussed early, along with the patient’s other preferences regarding advanced or terminal stages of their disease. Specific goals of using NIV in the clinical setting should be carefully explained and thoroughly understood by the patient and their family with particular emphasis on concrete measures that can and, if approved, will be taken to quickly achieve patient comfort in cases when NIV fails to achieve the desired results.

Monitoring During NIV

It has been reported that NIV targeted at control of respiratory distress (ie, category 3) can also be successfully used outside the ICU, for example, wards, hospices, and even at home. Monitoring should be oriented toward continually reassessing whether NIV is succeeding or failing to meet therapeutic goals (in category 3, eg, the only monitored aspect would be relief of symptoms).

Discontinuing NIV at the End-of-Life

NIV used to aid patients through the end-of-life should be discontinued when patients feel that the NIV is not making them more comfortable or when patients are no longer able to communicate. This is probably the moment when the benefits of NIV have ceased, since patients have lost control over the decision to continue NIV support.

There are important questions regarding the use of noninvasive ventilation during the end-of-life period that have not yet been addressed. There are no data assessing the quality of NIV–assisted end-of-life care. We also do not have a full appreciation or analysis of life support resource utilization or the effectiveness of palliative NIV on survival.

Noninvasive ventilation can and should be considered as a feasible palliative therapeutic option as well as a comfort measure during the dying process. However, there are only limited evidence about what type of patients and what diseases would mostly benefit from palliative NIV. To date, few studies have been published that address this important issue. Presently, decisions about palliative NIV have to be evaluated carefully and on a case-by-case basis, and oriented on the potential benefits for a given patient.