Abstract
Noninvasive respiratory support at the end of life is controversial, although it is becoming increasingly common. Supplemental oxygen is widely prescribed for palliative care and may help with hypoxemic respiratory failure. Noninvasive ventilation has a well-established evidence-based role in the management of respiratory failure due to exacerbations of COPD and cardiogenic pulmonary edema. However, its role during palliative care is unclear, and evidence of support is limited. High-flow nasal cannula oxygen therapy is a new strategy for which there is evidence to support its use for hypoxemic respiratory failure. However, any benefit of the use of high-flow nasal cannula oxygen therapy in the palliative setting is unknown at this time. This review examined evidence relating to the use of noninvasive respiratory support at the end of life.
- conventional oxygen therapy
- end of care
- high-flow nasal cannula therapy
- hypoxemic respiratory failure
- noninvasive ventilation
- palliative care
Introduction
The origins of palliative care can be traced back to the 1960s.1 Palliative care is defined by the World Health Association as “an approach that improves the quality of life of patients (adults and children) and their families who are facing problems associated with life-threatening illness. Palliative care is designed to prevent and relieve suffering through the early identification, correct assessment and treatment of pain, and other problems.”2 The general objectives of palliative care include the following: relieving pain and other distressing symptoms, affirming life, regarding dying as a normal process, intending to neither hasten nor postpone death, offering a support system to help patients live as actively as possible until death, enhancing the quality of life, and may also positively influencing the course of an illness.2 Palliative care is often confused with hospice. Palliative care, however, can begin at diagnosis and at the same time as treatment, whereas hospice begins after treatment of the disease has stopped. In other words, all patients in hospice receive palliative care but not all patients who receive palliative care are in hospice.
In the same manner, not all patients who receive palliative care carry a do-not-intubate status. Although some patients who receive palliative care may have a do-not-intubate status, it is possible that other patients may be receiving palliative care without a do-not-intubate directive. The patient can request not to be intubated (which is considered invasive) but still have other noninvasive measures carried out to alleviate dyspnea at the end of life. These would include things such as conventional oxygen therapy, noninvasive ventilation and high-flow nasal cannula (HFNC) therapy.
The use of palliative care has steadily increased in the past decade. Chatterjee et al3 performed a retrospective cross-sectional study by using data from the National Inpatient Sample Database from 2009 to 2013. Of the 1,751,870 hospitalizations, palliative care was used for 10.2% of the patients on long-term mechanical ventilation (178,745). The utilization of palliative care also increased yearly, from 6.5% in 2009 to 13.1% in 2013 (P < .001). As mentioned, palliative care at the end of life is designed to help the patient feel more comfortable and perhaps improve communication abilities. Unfortunately, near the end of life, dyspnea and acute respiratory failure are common. In fact, it has been reported that the prevalence of severe dyspnea in patients who are terminally ill is ∼65% with heart failure, 70% with lung cancer, and 90% with COPD.4 Because of this, one of the major objectives of palliative care is the relief of dyspnea.
Supplemental oxygen administration has long been used as a perceived comfort measure in patients near the end of life, especially when hypoxemic respiratory failure is present. Noninvasive ventilation (NIV) has a well-established role in the management of respiratory failure due to COPD, cardiogenic pulmonary edema, and hypoxic respiratory failure in patients who are immunocompromised.5 Considerable controversy exists as to whether NIV should be used in attempts to relieve dyspnea at the end of life. Some clinicians argue that palliative use of NIV may actually improve immediate survival in some cases and that it may be a valuable tool to help the patient communicate with family and perhaps complete some end of life tasks.6,7 Other clinicians maintain that use of NIV in this setting actually causes harm and contributes to an unjust use of health-care resources, and may actually delay the dying process.8,9 High-flow oxygen (≥40 L/min) is a relatively new therapy that also could be offered to patients at the end of life to help relieve dyspnea from hypoxemia, but there is little evidence to support its use at this time.
Supplemental Oxygen
Supplemental oxygen is frequently prescribed for patients in palliative care to manage their dyspnea at the end of life, even if they are not hypoxemic.10,11 In a survey sent to 648 palliative care specialists, >70% of them responded that they had ordered palliative oxygen if the patient was dyspneic. Thirty percent of the ordering physicians prescribed palliative oxygen merely at the patient's request.12 Published guidelines allude to the use of palliative oxygen on an individualized basis, but there is no clear evidence of symptomatic benefit in the literature.13,14 An Ad Hoc Committee was commissioned by the ATS to summarize clinical approaches managing what they termed a “dyspnea crisis.” In this summary the use of supplemental oxygen is mentioned as a sidebar without any elaboration or support.15 In fact, there is growing evidence that supplemental oxygen in palliative care may be unbeneficial and unnecessary.16–20
Uronis et al16 conducted a systematic review to determine the effect of supplemental oxygen on the relief of dyspnea in subjects with cancer who were mildly hypoxemic or nonhypoxemic. The pooled data from this review indicated that oxygen was not effective at reducing the sensation of dyspnea in this patient population.16 Abernathy et al17 performed a double-blind, randomized controlled trial in 239 out-patient subjects who had a terminal illness and who were dyspneic and had a PaO2 > 55 mm Hg. The study took place at 9 sites in Australia, the United States, and the United Kingdom.17 The subjects received either oxygen or room air through a concentrator, which they wore at least 15 h per day.17 The group used a numerical rating scale to measure breathlessness; they found that oxygen provided no additional symptomatic benefit in the relief of dyspnea.17
Campbell et al19 performed a double-blind repeated-measures observation, with each subject serving as his or her own control. They used a respiratory distress observational scale in 32 subjects who they described as being near death. Repeated measurements were taken on each subject both on and off oxygen. The application of oxygen to the subjects who were near death was not supported in their results.19
There is other robust evidence that mere air movement across the nasal passages results in dyspnea relief, even if by placing a fan in front of a patient's face. Galbraith et al21 performed a randomized controlled crossover trial in 50 subjects to test the effectiveness of a handheld fan directed at the face to reduce the sensation of breathlessness for subjects who were breathless at rest. The primary outcome measure was a decrease in breathlessness of 1 cm or more assessed by a 10-cm vertical visual analog scale with end anchors of “no shortness of breath” and “shortness of breath as bad as can be.” The result was a significant decrease in breathlessness when using a handheld fan directed at the face (P = .003). The mechanism by which the fan reduced breathlessness remains unclear. Several investigators have undertaken studies in healthy volunteers or subjects with COPD.22,23 The results indicate that physical stimulation or exposure of nasal or oral mucosa receptors to cold air results in decreased breathlessness.22,23 Because air motion seems to be an operative factor in relief of breathlessness, a simple handheld or tabletop fan could be a helpful, inexpensive first step.
The use of oxygen in patients who receive palliative care is widely prescribed. However, the evidence that supports this practice is essentially nonexistent with regard to the relief of dyspnea unless there is some degree of hypoxemic reversibility. It, however, may be that merely applying oxygen may increase the comfort level of the patient (and perhaps family) through the extra attention given. Or it may the “fan effect” mentioned above where the simple gas flow to the face can provide comfort. It is important to bear in mind that the application of oxygen is not without cost, and, in some circumstances, it may stress clinical staff workload. When considering prescribing oxygen for a patient who is receiving palliative care, it is important to weigh potential benefit (if any) against the induced cost and effect on staffing.
NIV
NIV is the frontline therapy in the management of COPD exacerbations and cardiogenic pulmonary edema, and in patients who are immunosuppressed and with respiratory failure.24,25 The use of NIV on patients with a do-not-intubate and/or comfort measures only request is controversial and the debate rages on. Clinicians who support the use of NIV for patients with do-not-intubate and/or comfort measures only argue that (1) there may be an improved chance of survival (especially if the patient has a reversible component to his or her dyspnea or respiratory failure), (2) there may be an improvement in patient comfort near death, (3) it provides additional time for patients to visit with family, and (4) it provides additional time to fulfill end-of-life tasks before death. Clinicians who say NIV should not be used in do-not-intubate and/or comfort measures only situations claim that clinicians may be delaying the dying process, unjustly using health-care resources, and that it may actually harm the patient. In this section, we examine the evidence that relates to the use of NIV in patients with do-not-intubate and comfort measures only desires.
Interest in the use of NIV for palliative care began in the 1990s. Meduri et al26 studied NIV in 11 subjects who were in respiratory failure, 9 subjects who were hypercapnic, and 2 subjects who were hypoxemic. They concluded that mechanical ventilation via oronasal face mask provided an effective, comfortable, and dignified method of supporting subjects with end-stage disease and acute respiratory failure.26 After this study, interest in NIV during palliative care waned but picked up again in the 2000s. The studies in the 2000s that examined the use of NIV during palliative care are listed in Table 1.26–34 Chu et al27 performed a single-center prospective observational study of the 1-y outcome of 37 subjects with COPD and with a do-not-intubate code who developed acute hypercapnic respiratory failure. Because the investigators did not have a truly matched control group for the subjects with a do-not-intubate directive, the comparison they used was of subjects with COPD and without a do-not-intubate directive who received NIV during the same study period (N = 43).27 This group received the same standardized medical treatment and were intubated if NIV failed.27 One-year survival in the do-not-intubate group was 30% versus 65% in the non–do-not-intubate group; however, subjects in the do-not-intubate group tended to be older and had more severe disease and worse premorbid conditions.27
Studies That Examined the Use of NIV During Palliative Care
Levy et al28 carried out a prospective multi-center cohort trial in 114 subjects who were receiving NIV for acute respiratory failure and who had a do-not-intubate order. The results were that the subjects with a do-not-intubate directive and with congestive heart failure or COPD had a relatively good chance of surviving to discharge (50% for congestive heart failure and 75%, COPD), whereas those with pneumonia, cancer, or other primary diagnoses had an ∼25% chance of surviving. Cuomo et al29 studied 23 consecutive subjects who had acute respiratory failure complicated by a solid cancer judged to be in an advanced stage and who were receiving palliative care. They found that subjects with acute respiratory failure and end-stage solid malignancies had an ICU and 1-y mortality rate of 39% and 87%, respectively.29 As expected, the mortality rates in this population were high, but they also found that a consistent subset of subjects was successfully treated with NIV if the cause of respiratory failure was reversible.29 A later study, by Schettino et al,30 examined all subjects with do-not-intubate status who received NIV for a 1-y period in a single-center prospective observational trial. The main findings of this study were that NIV was successful in reversing acute respiratory failure and in preventing hospital mortality in subjects with do-not-intubate directives and with the primary diagnosis of COPD or cardiogenic pulmonary edema (≤40% hospital mortality), and NIV was not useful in preventing hospital mortality in do-not-intubate subjects with hypoxemic acute respiratory failure (>80%).30
Meert et al31 looked at the use of NIV in 18 cancer subjects (17 with solid tumors and 1 with hematologic malignancy) with life-support techniques limitation (no invasive mechanical ventilation, hemodialysis or cardiopulmonary resuscitation) in acute respiratory failure. Eleven subjects had hypoxemic respiratory failure, whereas 7 had hypercapnic respiratory failure.31 They put the subjects on pressure support (10–18 cm H2O) with a PEEP of 4–8 cm H2O and found that 78% of the subjects were discharged from the ICU and that 56% were able to leave the hospital.31 Of note, 75% of the subjects who specifically had lung cancer survived their hospitalization. The group concluded that NIV seemed to be an effective ventilation support for cancer subjects with limited life support options.31
Bourke et al32 assessed the effect of NIV on quality of life and survival in 92 subjects with amyotrophic lateral sclerosis in a randomized controlled trial. At enrollment and at intervals of 2 months thereafter, the subjects underwent assessment of symptoms and measurement of sitting and supine vital capacity, maximum inspiratory and expiratory pressures, and sniff nasal inspiratory pressure.32 They were randomly assigned to receive NIV (S/T mode) or standard care if they presented with either orthopnea with maximum inspiratory pressure of <60% of predicted or symptomatic hypercapnia during a visit.32 The results were that, in subjects with amyotrophic lateral sclerosis without severe bulbar dysfunction, NIV improved survival (219 versus 171 d, P = .006) with maintenance of, and improvement in, quality of life.32
Fernandez et al33 performed a single-center retrospective cohort study of all subjects who received NIV in a 2-y period (N = 233). Of that group, 36 had do-not-intubate orders and 199 did not. They found that overall hospital survival was 66%. Survival was better in the subjects without do-not-intubate orders (74%) than in the group that had do-not-intubate orders (26%), both during hospitalization and after 6 months (64% with no do-not-intubate order versus 15% with do-not-intubate order). In both groups, the presence of COPD was associated with a better prognosis during hospitalization, but there was a decline in survival over time that resulted in no medium-term benefit.
Finally, Bülow and Thorsager34 performed an observational, single-center, retrospective follow-up study on ICU subjects who were receiving NIV and who had a do-not-intubate order. During a 1-y period, 38 subjects were identified: 11 died in the ICU, 16 died on the ward, and 11 survived the hospital stay.34 Five of the survivors died within 6 months, 1 died after 2.7 y, 1 died after 3.3 y, and 4 were still alive after 5 y.34 The results of these studies seemed to indicate that it may be reasonable to at least offer NIV in situations in which invasive ventilation is not an option. This is especially true in patients with some degree of reversible respiratory failure, such as patients with COPD or cardiogenic pulmonary edema.26–34
There have been several guidelines released from various organizations that relate to palliative care and the relief of dyspnea near the end of life. The International Consensus Conference on Intensive Care Medicine issued a statement that the “the use of NIV may be justified in selected patients who are do-not-intubate but have a reversible cause of respiratory failure and may provide patient comfort and facilitate physician-patient interaction in the assessment of the reversibility of the respiratory failure.”35 In 2008, Lanken et al13 authored an official statement for the American Thoracic Society regarding palliative care for patients with respiratory diseases and critical illnesses; in it, they advocate that NIV be used when severe dyspnea is present. This was based on a case series presented by Shee and Green,36 in which NIV was deemed helpful in reducing dyspneic symptoms; however, in 5 of the 10 cases, the patients had a diagnosis of COPD, in which NIV seemed to have a definitive role.
A consensus statement from the American College of Critical Care Medicine was also released in 2008.14 This group recommended the use of NIV during end-of-life care after careful evaluation of patient goals. Two situations were cited in which the group thought that NIV might be useful. First is the patient who refuses intubation but desires other clinical procedures with the goal of prolonging survival. Second is a palliative technique to alleviate dyspnea. The first guidelines that focused on a categorical approach to NIV delivery came from a task force from the Society of Critical Care Medicine in 2007.5 They separated how NIV should be used according to 3 categories: (1) life support without preset limits, (2) life support with preset limits (do-not-intubate), and (3) comfort measures only.5 For each category, the group defined the primary goals of care, the main goals to communicate with the patient and family, determination of success, the end point for NIV, response to failure, and likely location of NIV.5 The end points for categories 2 and 3 are intolerance of NIV, the patient is not more comfortable, and the patient becomes unable to communicate.5 The task force also recommended that this information should be communicated with the patient and family before the therapy is initiated so that they are prepared in the event of NIV failure.5 Also, the goals of care should be reassessed on an ongoing basis because the risks, benefits, and treatment preferences can change over time.
More recently, Nava et al37 examined the palliative use of NIV in subjects with solid tumors and at the end of life. The group compared NIV with standard oxygen therapy in a randomized feasibility study.37 They found that dyspnea (measured by the Borg scale) was reduced in the NIV group compared with the oxygen group; morphine use was also lower in the NIV group as well.37 NIV was even more effective in reducing dyspnea if the subject had a component of hypercapnic respiratory failure.37
Azoulay et al38 looked at NIV in subjects who were terminal and with a do-not-intubate directive. This was a prospective observational study of 134 subjects with a do-not-intubate directive; 25% of the subjects survived to day 90.38 Those survivors reported no decrease in the health-related quality of life through telephone interviews.38 Baxter et al39 examined the family caregiver and health-care professional experiences when NIV is used at the end of life in 20 subjects with motor neuron disease. They concluded that the use of NIV in these situations was beneficial and not perceived as adversely affecting the subject's end-of-life experience.39
The most recent study that examined the use of NIV for subjects with do-not-intubate and/or comfort measures only was performed by Wilson et al40 in which they did a systematic review and meta-analysis of 33 studies. They found that a significant proportion of subjects with do-not-intubate orders survived to hospital discharge (56%) and at 1 y (32%).40 Survival was better for subjects with COPD (68% to hospital discharge) and cardiopulmonary edema (68% to hospital discharge) but worse for subjects with pneumonia (41% to hospital discharge) and malignancy (37% to hospital discharge).40 The takeaways from these studies are that the use of NIV during palliative care is not infrequent and that approximately half of the subjects with a reversible aspect of respiratory failure (COPD or cardiopulmonary edema) survive to hospital discharge.37–40
In 2017, the most recent clinical practice guidelines41 were developed by clinicians from the European Respiratory Society and the American Thoracic Society with regard to the use of NIV in acute respiratory failure that stems from different etiologies. One of the guideline sections addresses the use of NIV in patients with acute respiratory failure who receive palliative care.41 Based on available evidence, the guideline committee recommended “offering NIV to dyspneic patients for palliation in the setting of terminal cancer or other terminal conditions” (conditional recommendation, moderate certainty of evidence).41
The possible benefits to the use of palliative NIV are improved survival and relief of dyspnea, perhaps to enable the patient to communicate verbally, and it may buy additional time for family to visit or to take care of end-of-life matters. Specific goals of the therapy, however, need to be well established and its definitive end points (eg, no relief of dyspnea, the patient does not tolerate NIV, the patient's status deteriorates) must be decided by the patient, family, and clinicians. In addition, alternative measures in the event of NIV failure need to be identified before the initiation of NIV. There are pitfalls that clinicians need to be aware of when considering palliative NIV. Unreasonable persistence with NIV may also unduly delay the dying process. Also, the patient and family must be made aware of the risks associated with the use of NIV. These risks include that survival may not be achieved, the mask and settings may be uncomfortable, and the possibility of facial necrosis. Clinicians need to be aware of the patient's goals at the end of life and how NIV may have symptomatic benefit as well as the potential to interfere with those goals.
When Should NIV Not Be Used in Palliative Care
There are situations in which NIV should not be used in palliative care. If the patient has some sort of facial deformity or injury that does not allow for a comfortable mask fitting, then something other than NIV should be considered. If NIV fails to relieve dyspnea, then it should be removed. If the patient loses consciousness, then it would be more dignified for the patient to have NIV discontinued. If the patient wants to communicate with loved ones at the end of life, then NIV may be inappropriate if it interferes with communication. Clear end points (eg, more comfort, relief of dyspnea, improved communication) should be established before any NIV attempts so that futile NIV therapy is avoided.
HFNC
HFNC oxygen therapy is a relatively new strategy for treating hypoxemic respiratory failure and dyspnea. Whereas traditional nasal cannula flows generally range from 2 to 6 L/min, HFNC flows run in the neighborhood of 40–60 L/min. This higher flow flushes carbon dioxide from the upper airway, thereby decreasing anatomic dead space and creating low-level CPAP. HFNC has some appeal over NIV in that it can be less claustrophobic, produces less skin breakdown, and does not impede eating or talking.
There is clinical evidence that HFNC is useful in the treatment of hypoxemic respiratory failure and may be superior to NIV and traditional oxygen therapy for this patient population.42 However, there is no evidence at the present time that HFNC is superior to NIV for palliative care. There are studies that attempted to compare HFNC with NIV in subjects with do-not-intubate orders. Peters et al43 examined 50 subjects with a do-not-intubate directive and with hypoxemic respiratory failure who received HFNC.43 HFNC provided adequate oxygenation, and many subjects did not need to escalate to NIV.43
In a retrospective single-center study,44 subjects with interstitial lung disease and hypoxemic respiratory failure and do-not-intubate orders had their records reviewed to compare the use of HFNC and NIV. The subjects who received HFNC had a similar survival rate to those who received NIV, but it was deemed that the HFNC was better tolerated than NIV.44 Another study examined the use of HFNC and NIV for the relief of persistent dyspnea in subjects with advanced cancer.45 Twenty-three subjects were randomized to receive either HFNC or NIV and then their level of dyspnea was assessed by using a numeric rating scale and the Borg scale.45 The investigators concluded that both HFNC and NIV alleviated dyspnea and improved physiologic parameters, and were safe.45 HFNC is a relatively new technology that provides adequate oxygenation in patients with hypoxemic respiratory failure. However, its role in palliative care is undetermined at the present time, and further research is needed. Clinicians should be cautious about non–evidence-based use of HFNC in patients without a realistic path to recovery.
Choice of Interface
Management of dyspnea presents a challenge because there is no roadmap to guide therapy. The typical recommendation is to relieve dyspnea by treating the underlying cause. However, in patients who receive palliative care, this is often not successful or simply not possible. The focus then shifts to symptomatic relief of dyspnea. In terms of noninvasive respiratory support, as discussed, choices include supplemental oxygen, NIV, and HFNC. Factors that determine which therapy to use include the presence of hypoxemia, presence of hypercarbia, patient comfort, and patient preference. Supplemental oxygen is the therapy most often used. It can be delivered via nasal cannula, simple oxygen mask, air-entrainment mask, or a non-rebreather mask. Nasal cannula is most often used for comfort reasons and is generally used when no hypoxemia or mild hypoxemia is present.
If a patient in palliative care is experiencing moderate-to-severe hypoxemia, then a simple oxygen mask, an air-entrainment mask, non-rebreather mask, or HFNC may be used because these allow for higher FIO2 delivery. HFNC therapy probably has the most appeal these days due to the comfort afforded by the use of a nasal cannula. NIV can be delivered via nasal pillows, nasal mask, oronasal mask, or full face mask. The choice of interface may depend partly on the presence of hypercarbia. If hypercarbia is present, then the NIV masks are preferable because they can facilitate bi-level pressure ventilation to help remove carbon dioxide. The type of mask and settings for these patients are variable and most likely will be dependent on clinician assessment and patient comfort because it is important for the patient to tolerate both the mask and the clinician settings. If both hypercarbia and hypoxemia are present, then the use of NIV may be preferable because it has the ability to alleviate both conditions to some extent. If neither hypercarbia nor hypoxemia are present but the patient in palliative care still is experiencing dyspnea, then the choice of therapy should be dictated by patient preference and comfort.
Summary
Noninvasive respiratory support at the end of life is controversial. Although supplemental oxygen is widely prescribed, there is little evidence of benefit. Perhaps the extra clinical attention eases anxiety for the patient and family but that is unproven. Perhaps the mere flow of gas helps relieve breathlessness. The use of NIV at the end of life is also controversial. There does, however, seem to be some benefit in patients with a component of reversible respiratory failure, such as COPD and cardiopulmonary edema. HFNC is a relatively new therapy that may have a place in palliative care, but no evidence to support its use exists yet. HFNC does have some appeal over NIV in terms of facial comfort. Regardless of which noninvasive therapy questions need to be considered by patients, families, and clinicians, include the following (1) how long will it prolong life, (2) will it restore or maintain alertness, (3) will the therapy increase or relieve suffering? The patient, his or her family, and the clinicians must establish clear goals and end points for whichever therapy is chosen. It is clear that more research is needed with regard to noninvasive respiratory support at the end of life.
Discussion
Strickland:
Thank you very much. This is a topic that I'm fairly passionate about as well. In my experiences in working with patients who require this at the end of life to help with symptoms knowing we're not going to change their physiologic outcomes, knowing we are not bridging to recover from whatever it is they have, my experience lies less with trouble with the family and the patient and more with the clinicians. During my time as an ethics consultant, I had several patients who were using NIV as a palliative measure at the end of life, knowing we weren't bridging, and the patients all very clearly acknowledged “I know this is not going to make me better.” The families clearly acknowledged “we know this will not prolong the life of my loved one, we know they will not recover from this.” The respiratory therapists (RTs), unfortunately, were very frustrated because it was “not doing anything” and that it was taking up their time, and they were wasting their time. And, that was a very frustrating thing for me as an RT to get my head around; you're here to provide comfort and care to the patient in whatever way that patient needs the comfort and care. My question comes from the perspective of educating our clinicians and how do we help our RTs understand that yes, indeed, this is an intervention that is not going to change the trajectory of this patient's prognosis and yet it's still an important care provision?
Davies:
That's a great point, I've seen that too when the RTs have a big workload, and it's almost like they're bothered by having to take care of this type of patient. However, I don't think it's just RTs, I think this situation applies to nurses and physicians as well. Another issue may be at shift changes when many different clinicians are involved in a patient's palliative care. Some clinicians may, for whatever reason, be less vested in the patient's care, and this can lead to confusion for the patient and family as to what level of care to expect. We have a palliative care team, but you're right, the education in terms of how each multidisciplinary clinician functions on the team is lacking in many instances. In our institution, it's almost impossible to train all the RTs and nurses in palliative care because of the sheer numbers, the turnover rate, and the fact that palliative care can take place in a number of different locations. Does anybody have a designated palliative care team that's very active in their institution?
Scott:
On the clinician's side of this, I think there can be considerable moral distress related to placing somebody on NIV for end-of-life care. Some clinicians say that NIV is comfort care, but it's the RT who is actually applying the mask to the face of someone who doesn't look well. We also know from years of experience that many people find NIV uncomfortable, so we have the pretense of making the patient “more comfortable,” which may not actually be the case. I think with the improvement of interfaces and the expanding role of HFNC, we have more options than we used to, but we still have a long way to go. It's a very tough situation for the RT at the bedside.
Davies:
It is, and I'll add another point to that. How old are your RTs? It's a very real that possibility RTs in their late teens or early 20s are dealing with a dying patient. It's highly likely they have never been involved in this type of situation before. Their heightened anxiety probably affects how they react to the patients. I've seen that in code situations, and I'm sure it happens in palliative care.
Strickland:
I'll reiterate what John [Davies] said earlier, that this is not a blanket type of intervention. Not every patient at the end of life who is experiencing these types of symptoms will respond well to NIV. They might, and it's not necessarily a bad idea to try, but also recognize that some patients will adapt and feel better and some won't.
Scott:
You also mentioned monitoring patients who are on NIV at the end of life. I would like to say that more frequently than I'd like to admit, I've been asked to draw a blood gas on a patient who was placed on NIV for the purposes of comfort. Of course, this prompts conversations about whether sticking someone for an arterial blood gas is the right thing to do. But, it's part of the stress that RTs deal with during this situation. It's 2:00 am and somebody's on NIV for comfort at the end of life and we're asked to stick them with a needle.
Davies:
That's a good example of perhaps a resident new to the patient coming on and not being familiar with the palliative goals that were established.
Volsko:
I think that really goes back to your initial comments about the goals of therapy. The goals are inclusive of the patients, family, the physician leading the care, and the entire team. When I looked around the room, I saw 3 participants here who have palliative care teams, all of whom come from children's hospitals. Although palliative care is an important service line in our venue, it is also important in the adult acute care setting. A palliative care service really does wrap in the entire team, so, before we even start a therapy or the plan of care is executed, the entire team, including the RT, the nurses, the physicians, and the family, all understand the direction that we are going and are all in agreement, which can affect patient and family satisfaction as well as patient outcomes.1 At our institution, we also allow those young RTs who may have difficulty with death and a dying patient the opportunity to opt out.
Davies:
Yes, in an ideal world that's the way to do it. Unfortunately, getting all those people together is like herding cats with a hockey stick. The reality is that it is a monumental task and I applaud your institution.
* Jackson:
The other challenge, and I've faced this a lot, whether it be NIV or HFNC, is, if we start them in the hospital setting and then their wish is to go home, the challenge to then be able to move to home care is to understand that not everything needs to be an HCPCS (Healthcare Common Procedure Coding System) code. We need to look at the big picture with these patients and find ways to take care of them outside the acute care arena without worrying about the expenses related to the devices. That's been our biggest challenge. We've moved to HFNC but still being able to get them home can be problematic. Home is where they want to die, they don't want to die in the hospital. I think that's another avenue that's not well explored, yet one that we need to figure out.
Davies:
I agree because, in most instances, you're transferring care to an interim group that really doesn't know the patient, plus you're putting the burden on the family to take care of the patient as well.
† Kallstrom:
You started off by talking about oxygen, are there any clinical practice guidelines or guidance about giving oxygen to patients in hospice?
Davies:
I doubt it. There are some recommendations but no guidelines that I'm aware of.
Strickland:
I don't think so.
Davies:
In my opinion, there is absolutely no consensus on how to prescribe oxygen to the patient in palliative care.
† Kallstrom:
I went through this with my parents, and it was killing me to not have oxygen on them, but the care team said this is not our practice, this is not the way we do it. You're right, it would have made me feel better but that's about it.
Davies:
Right, and oxygen is the one therapy that would be fairly easy to come up with guidelines for. I will say that, during my research for this presentation, I couldn't find any evidence that shows that oxygen provides symptom relief.
Hill:
That was a really nice discussion, John, and the issue of defining clear goals is just so important. Sometimes survival is one of the goals in that, if you have a patient with COPD or cardiogenic pulmonary edema who has a do-not-intubate/do-not-resuscitate order, he or she may still desire to survive the hospitalization. You showed data from our study that demonstrated that the patients with those diagnoses had a better than even chance of surviving hospitalization if they used NIV. There was a task force convened by the Society of Critical Care Medicine on NIV in patients with do-not-intubate/do-not-resuscitate who I'm sure you looked at when you put this together. Randy Curtis was chair and first author on the publication.2 The document pointed out that goals might include survival or just amelioration of a specific symptom, like dyspnea, and how important it is to pay attention to those goals. There has been increasing use of HFNC to palliate these patients because of its greater tolerability than NIV. The study by Epstein et al3 you presented is relevant here. I think the most important finding was that 85% of the subjects tolerated it well. That's not something you'll see with NIV. I find myself more and more stopping NIV when I see these patients and switching them to HFNC. Still, we need more studies in these patients to better understand how best to use these modalities to palliate these patients.
Davies:
I agree. It looks like the patient is wearing a normal nasal cannula as opposed to having a mask strapped on his or her face, and so I think it's comforting from that perspective. Maybe HFNC will help with a reversible hypoxemic respiratory failure element but, certainly, patient and family perception of someone having a cannula versus a full face mask makes them less anxious.
Kacmarek:
Our patients with comfort measures only are almost exclusively managed with HFNC. We now rarely apply NIV to that group, those who choose do-not-resuscitate and/or do-not-intubate, it depends on the particular diagnosis, but, for patients with comfort measures only, it's almost exclusively HFNC because they tolerate it so much better, they can eat, they can drink, they can communicate, they can have a more-normal dying process than having a mask strapped to their face.
MacIntyre:
I'd like to take a moment to clarify the term HFNC. In our discussions here, I think what we're referring to are high-flow devices such as the Optiflow (Fisher & Paykel, Irvine, CA) or Vapotherm (Vapotherm, Exeter, NH) systems in which we are talking about 30-60 L/min of humidified flow, with physiologic effects beyond just supplementing oxygen. Unfortunately, I find that, at Duke, and I don't know about other places, there's a notion that simple wider-bore nasal cannulas that can deliver 8-12 or 15 L/min are an HFNC. This creates obvious confusion, and I think we need to make very clear that whenever we're talking about HFNC in these discussions, we're talking about the 30-60 L/min devices and we're not talking about 8 L/min through a slightly larger traditional nasal cannula.
Benditt:
This is a real difficult area, I was on the task force with Randy [Curtis], and I think you were too Nick [Hill].2 To me, this is an area in which the art of medicine is very important in the sense that being at the bedside with the patient is so critical. Back in the day when we started NIV in a patient with COPD, it was novel and the physician would go to the bedside and do the mask fitting with the therapist. I think this is one in which same principle applies. It's sometimes difficult to be present with the patient long enough due to work hours and workloads but it is important to identify if it's not helping so it can be stopped. Second, I think it shouldn't be used in diseases that are not hypercarbic respiratory failure, so COPD, the neuromuscular patients I'll be talking about. I certainly don't think every patient who's near the end of life should be offered NIV, but I do think it can be very beneficial in some cases but it requires a soft touch and a gentle approach that is part of palliative care. You can't write an order and then leave the room.
Davies:
I do believe merely the extra attention provided the team at the end of life is of help psychologically to both the patient and family.
Hill:
You mean we actually should go to the bedside rather than try to figure it out at the computer screen?
* Jackson:
I'd just like to add to Neil's [MacIntyre] comment. We've started to change the mindset of our physicians and our nurses, and we no longer allow them to call the green nasal cannula device an HFNC. We tell them it's a mid-flow cannula. If it's anything between 8-15 L/min, it's going to be a mid-flow cannula and amazingly enough it's picking up. I heard nurses the other day doing a report, and they said this patient is on a mid-flow cannula. Just another thought process, I'm not saying we have to change the way we look at it all together but that's how we're handling it so that people know what we're talking about.
MacIntyre:
I think that's a wonderful idea.
# Branson:
I have a couple of comments about this HFNC issue. One, and to Neil's point, what's high flow depends on whether you're in the neonatal ICU or pediatric or adult arena, so flow alone isn't going to do it. The other I think is really important, for years, we have preached that high-flow oxygen in COPD is bad. So it's not just flow, it's the high-flow low FIO2oxygen distinction that I think is really important when we talk about this that it's not 100% oxygen at 60 L/min, it's 30 or 40%, or at the stated flow. I can remember when everybody got a bubble humidifier at 2-3 L/min, and, then we did studies and found there's no difference between that and using nothing. But, does anybody besides me wonder how much better patients might do on 3 L/min if they had a high-flow humidifier in terms of dryness of the airway and discomfort? I know it would be expensive, but I really think that's something we could consider. My final observation is the most inappropriate application I see now is this kind of mid-flow approach in which a patient receives 8 L/min nasal oxygen and therapists get a bigger cold bubble humidifier. That larger cold humidifier does nothing to add humidity. In respiratory care, the terms we call equipment often is a little misleading, I've already heard today a panel member calling an HFNC by a brand name. We have to be careful and specific to say what we mean and mean what we say.
* Jackson:
If I have my way, those larger cold bubble humidifiers will be gone very soon from our facility for adults. You're right, all they do is blow bigger bubbles.
Footnotes
- Correspondence: John D Davies MA RRT FAARC, Duke University Health System, Box 3911, DUH, Durham, NC 27710. E-mail: john.davies{at}duke.edu
Mr Davies discloses a relationship with Teleflex Medical.
Mr Davies presented a version of this paper at the 57th Respiratory Care Journal Conference, Noninvasive Respiratory Support in Adults, held June 14-15, 2018, in St Petersburg, Florida.
↵* Julie A Jackson RRT RRT-ACCS, invited discussant, Fisher and Paykel.
↵† Thomas J Kallstrom RRT MBA FAARC, Executive Director, American Association for Respiratory Care.
# Richard D Branson MSc RRT FAARC, Editor in Chief, Respiratory Care.
- Copyright © 2019 by Daedalus Enterprises
