Noninvasive Ventilation in Seriously Ill Older Adults at the End of Life—The Evidence Remains Elusive

Acute respiratory failure is a frequent cause of hospitalizations among seriously ill older adults at or near the end of life. These terminal hospitalizations are often complicated by distressful respiratory symptoms and chaotic transitions that result in high-intensity, high-risk interventions such as invasive mechanical ventilation (IMV) and noninvasive ventilation (NIV), which delivers positive pressure ventilation through a mask. In this issue of JAMA Internal Medicine, Sullivan and colleagues¹ studied trends in IMV and NIV use among older decedents hospitalized at the end of life. The authors found a substantial increase in NIV use and a slight increase in IMV use in the past 2 decades, potentially signifying a major shift in the way that clinicians provide ventilatory support at the end of life. Although use of NIV in older adults with terminal respiratory failure may seem appealing, high-quality evidence supporting its use across serious illnesses remains elusive.

In the current study, Sullivan and colleagues¹ analyzed patterns in NIV and IMV use over 17 years in almost 2.5 million older Medicare beneficiaries who were hospitalized in the last 30 days of life. Overall, the mean age of the cohort was 82 years, and 21.3% of Medicare beneficiaries were admitted for pneumonia or sepsis. Use of NIV during terminal hospitalizations increased 9-fold between 2000 (0.8%) and 2017 (7.1%), whereas IMV use remained relatively stable in the same period (15.0% in 2000 to 18.2% in 2017). The increase in NIV use was pronounced in older patients, with a 10-fold increase in those with congestive heart failure (CHF) and a 5-fold increase in those with chronic obstructive pulmonary disease (COPD). In addition, NIV use increased by 9-fold in patients with cancer and dementia at the end of life. Although the increase in NIV use in CHF and COPD was reciprocated by a decrease in IMV use, this reciprocal reduction was not observed in patients with cancer and dementia, who experienced a slight increase in IMV use during the study period. Furthermore, two-thirds of older adults who received NIV required an intensive care unit, suggesting that resource use remained high for older decedents at the end of life. However, as a signal for potential palliative advantage in an exploratory aim of the study, Sullivan and colleagues¹ found that older decedents who received NIV had higher frequency of hospice enrollment and lower rates of in-hospital death overall; the advantage did not appear to be as pronounced in those with cancer and dementia as in those with CHF and COPD.

Routine NIV use to deliver positive pressure ventilation grew in the 1990s owing in part to the proliferation of more comfortable masks and a push to avoid the potential complications of IMV.² The evidence quickly expanded to support NIV to deliver bilevel positive airway pressure in patients with acute hypercapnic respiratory failure with COPD, acute cardiogenic pulmonary edema, acute hypoxemia in those who were immunocompromised, and in select patients after surgical procedures to facilitate liberation from IMV.³ Guidelines delineated when NIV should be avoided, including when a patient is medically unstable, is acutely agitated, is unable to cooperate with the NIV mask, cannot protect their airway, cannot clear their secretions, or has facial trauma or anatomical abnormalities that preclude proper mask fit.⁴ The data from the current analysis by Sullivan and colleagues¹ seem to illustrate that clinicians were practicing beyond these guidelines and potentially broadly using NIV to provide palliative ventilatory support at the end of life across populations of seriously ill older adults.

Where the evidence for NIV use remains elusive is in older patients with de novo acute hypoxemic respiratory failure without a previous diagnosis of chronic respiratory disease (ie, pneumonia or acute respiratory distress syndrome), as in many decedents in the current analysis.¹ Evidence does not clearly support NIV use in these situations, resulting in the European Respiratory Society and the American Thoracic Society not making a recommendation for its use in these instances in a 2017 joint statement.³ Although NIV use in younger patients with de novo acute hypoxemic respiratory failure under close monitoring and with experienced clinicians could be supported,⁵ its use for this indication in older adults may confer greater risks.⁶

As a means of providing palliative ventilatory support, NIV reduces the work of breathing and dyspnea and is often initiated to avoid the need for IMV or to offer more time for families. However, the European Respiratory Society and the American Thoracic Society provided only a conditional recommendation for palliative use of NIV based on a moderate certainty of evidence in studies with small samples. For instance, in a study of older patients with a do-not-intubate status, NIV use compared with standard medical therapy in acute hypercapnic respiratory failure reduced respiratory distress, decreased the need for IMV, and improved survival.⁷ In a randomized feasibility study of hospitalized patients with solid tumors in the last 6 months of life, NIV use compared with oxygen delivered by a venturi or reservoir mask demonstrated reduced dyspnea and morphine needs and showed a signal for in-hospital and 6-month survival in a subgroup of participants with acute hypercapnia.⁸ Participants found NIV acceptable when it was thoroughly explained and when voluntary withdrawal from NIV was ensured.⁸ However, the mean age of participants was 10 years younger than participants in the study by Sullivan and colleagues¹, thus emphasizing the need for more research in older adults with terminal respiratory failure. Furthermore, that feasibility study⁸ was conducted in preparation for a larger trial and did not compare NIV to high-flow nasal cannula oxygenation, which is well tolerated by patients, is at least noninferior if not superior to NIV in many post hoc analyses, and has a superior safety profile.^9,10

The rapid proliferation of NIV use in older adults with advanced dementia in the current study by Sullivan and colleagues¹ is concerning, especially given the absence of evidence to support its use as palliative ventilatory support in this specific population. Noninvasive ventilation is life support that can be difficult to discontinue in older patients with advanced dementia. Its use must be tempered because of the potential for causing aspiration, delivering injuriously large tidal volumes over extended periods, and prolonging or increasing distress when a patient is unable to cooperate with the mask or to understand what is happening owing to substantial cognitive impairment. Instead, high-flow nasal cannula oxygenation instead of NIV may be indicated and deserves more research.

Patient-centered NIV use in seriously ill older adults with terminal respiratory failure entails careful patient selection and use in conditions that are supported by evidence to yield favorable outcomes, which are largely situations that have a potential for reversibility (ie, acute hypercapnic respiratory failure in the setting of COPD or acute cardiogenic pulmonary edema associated with CHF).⁵ Beyond that, the evidence is elusive. The increasing use of NIV without a reciprocal reduction in the use of IMV among older adults with advanced cancer and dementia also illustrate the ongoing struggles to integrate early palliative care before the end of life in seriously ill older adults. Given the limited and variable access to specialist palliative care, more values-based discussions by frontline clinicians with their older patients long before a terminal hospitalization, including conversations about NIV and IMV use, could fill these gaps. Broad NIV use at the end of life, even as a means of palliative ventilatory support, could have unintended consequences in certain subgroups, such as older adults with advanced cancer and dementia. Future research on terminal respiratory failure in these vulnerable populations and their bereaved caregivers should explore in more detail the feasibility, acceptability, and palliative benefits of NIV, especially in comparison to high-flow nasal cannula oxygenation. Although the use of this newer modality may be limited in smaller centers due to equipment availability, the recent proliferation of high-flow nasal cannula oxygenation could alter some of the trends seen in the current analysis and radically change the evidence for NIV use in subgroups of seriously ill older adults with terminal respiratory failure.