Respiratory therapists treat dyspnea daily with vast resources ranging from nasal cannulae to mechanical ventilators; from relaxation and breathing exercises to pharmacological agents. Dyspnea presents in up to 70% of patients in the last 6 weeks of life, according to a National Hospice Study.1 It is one of the more common symptoms that bring terminally ill patients to the hospital. There is a process of treating patients in a curative or restorative manner and then a process of providing palliative care. The relationship of the two processes varies greatly. In a perfect world, patients would receive both concurrently. Clearly, any treatment that palliates discomfort benefits the patient.

Models of Care

The World Health Organization (WHO)2 presents interesting models of the relationship between curative/restorative care and palliative care. The American Thoracic Society (ATS),3 in its policy statement on palliative care in respiratory and critically ill patients, takes these models further. All models begin with the onset of disease. The models have two different endpoints. Two end with the patient’s death. The other two extend beyond the patient’s death to family bereavement. All models can exist in the same health care facility. Acute care hospitals are geared to restorative/curative care. Often palliative care is considered only when acute, curative care is no longer possible and medical futility is reached. This represents one model: Curative care ends and palliative care begins. Another model, with patient death as the endpoint, shows a linear relationship between curative care and palliative care. The other two models have curative/restorative care paralleling palliative care but to different degrees. Both take divergent paths as patient death approaches.

What makes so many models exist relates to more problems with terminal dyspnea, those being etiology, definition, and recognition of medical futility. Some etiologies are the natural progression of a chronic disease process. Heart failure and chronic obstructive pulmonary disease are two such examples. Other etiologies are acute, treatable sequelae of disease processes such as a malignant pleural effusion. Pleural effusions bring the added troublesome symptom of pain as sensory fibers are located in the parietal pleura. Some identify terminal dyspnea as the terminal event in a chronic disease process for which palliative care is initiated. Others see the onset of dyspnea as the beginning of the end, so to speak, for some disease processes such as cancer. Dyspnea can herald a change from intensive treatment and curative/restorative care for a patient to palliative care. Note that palliative care does not mean “do not treat” acute problems. Palliative care is comfort care. Hospice care does not provide curative or restorative therapies or treatments for any underlying conditions.

Curative or Palliative Care?

In hospitals, some intensive care treatments are so encompassing that patient comfort can be overlooked, but the transition from intensive care to palliative care or the introduction to palliative care can be a seamless one when barriers to this transition/addition are recognized and addressed. For example, a 70-year-old male with severe COPD is in the CCU awaiting evaluation for aortic valve replacement. He tolerates various amounts of time off mechanical ventilation on a trach collar on a daily basis. After several weeks, a speech pathologist is consulted to perform a bedside swallow evaluation. The speech pathologist asks the respiratory therapist to join in the evaluation at the bedside to consider the patient’s risk for aspiration and the presence of the trach tube. The final determination was that with several patient safety protocols in place, the patient could have pudding, applesauce, and thickened liquids. This provided a great deal of comfort and satisfaction for an acutely ill patient. How often is such an evaluation done in an intensive care unit?

Another barrier to palliative care in an ICU is medical futility. Patients vary greatly in their recognition and acceptance of medical futility. Some patients who have been ill for a prolonged period of time battling a particular disease become sick and tired of being sick and tired. Other patients seek curative treatment to the very end. Neither position is being judged; they are presented solely for the degree to which patients recognize medical futility. Medical futility is not a concept the health care team and family arrive at together. Medical futility is often identified by physicians first, then by allied health professionals, and last by the families. Families and loved ones should be involved to the extent of the patient’s wishes. Ideally, advance directives will identify health care surrogates who will use substituted judgment in the event the patient can no longer participate in health care decision-making processes.

Communicating with patients and families regarding end of life issues requires certain skills. The ATS policy statement even outlines them. Candor and compassion are vital during this process. Curtis et al provide a good discussion of missed opportunities during family conferences and end of life care.4 Patients and their families need assistance putting palliative care plans into activities of daily living. Some interventions are pharmacological. Some involve health care providers in the home. Palliative therapies for dyspnea include treating the underlying cause when possible. These can be pneumothorax, pleural effusion, anemia, a reversible component of obstructive airway disease, secretion management, congestive heart failure, pain management, and relieving anxiety. Other strategies to relieve terminal dyspnea include using a fan to blow cool air on the patient’s face, opioids, anxiolytics, and noninvasive ventilation when tolerated.

Oxygen Versus Air

Other strategies involve the use of oxygen. Oxygen in the home conjures up fears in the minds of many a family member. “What if the oxygen tank falls?” “What if there is a fire?” The list is long. The guidelines for home oxygen in the hypoxemic patient are clear-cut. The patient must have a Pao2 <55 mm Hg or a Sao2 <87%. Guidelines are also clear for palliative use of oxygen. The physician prescribes “palliative” oxygen. That should be the end of the discussion, but at times it is not. Since the etiology of terminal dyspnea is often multifactoral, will oxygen be enough in hypoxemic patients? Bruera et al, in a 1993 study in Lancet, found that hypoxemic patients did, in fact, feel relief from their dyspnea when using oxygen as opposed to air.5

What is the benefit of oxygen in terminal dyspnea in the nonhypoxemic patient? Abernethy et al addressed this very issue in Lancet.6 Here patients were randomized to oxygen or air in a double-blinded study. Concentrators were set up in the home to deliver oxygen at 2 LPM or air at 2 LPM. Patients used the medical gas for 15 hours a day. Neither the patients and families nor the researchers knew what medical gas the patient received. A period of 7 days was used to evaluate dyspnea based on a survey of palliative care physicians. Patients were asked to use a 10 point dyspnea scale twice daily—once upon awakening and once when going to bed. The scale ranged from breathlessness only on strenuous exercise to getting breathless while dressing. Patients were not enrolled in the study if they had refractory dyspnea, Pao2 greater than 54.9 mm Hg. (Note: Since the study was published in Lancet, kPa was the unit of pressure used to measure oxygen levels. These values have been converted in this review.7) Exclusion criteria included long-term oxygen use, history of hypercarbic respiratory failure, anemia, Paco2 >50.4 mm Hg, or a cardiac or pulmonary event within 7 days prior to enrollment.

Breathlessness did not vary in either group (air or oxygen) at any time in the study. Diaries recording symptoms were begun 2 days before the introduction of medical gas to each patient. Breathlessness decreased in both groups after initiating medical gas therapy compared to before when all patients were breathing room air. Quality of life surveys completed by all patients did not vary in either group. An improvement in perceived quality of life was reported by both groups after 3 days into the study. This increase was a significant 87% improvement.

The use of medical air in this patient population is not merely a placebo effect but an intervention. Since the effects of the medical gases, both oxygen and air, relieved dyspnea and none of the patients were hypoxemic, it leads one to question the effect of gas flow in the nasal passages on dyspnea. Cool air provided by a fan is known to relieve dyspnea. Can the mechanisms be somewhat related? Is the effect related to relieving anxiety? As with any good study, the answers lead to more questions.

While all patients reported benefit within the first 3 days, many reported relief within the first 24 hours. If a nonhypoxemic patient is to benefit from medical gas delivery, it can be evaluated early. Thus, turning the home into a medical equipment supply store can be avoided if this treatment becomes a source of anxiety for the patient rather than a source of relief.

Remember that dyspnea is a subjective symptom. Patients often report less dyspnea than do family members. Family member distress can contribute to patient anxiety, and educating the family to address their loved one’s level of dyspnea, not their own level of comfort with that dyspnea level, is the goal of therapy.

Monitoring for Improved Care

Once one decides to monitor successful management of terminal dyspnea, a quality monitor needs to be assessed. Quality indicators at end of life assist in developing improvements in health care delivery to this patient population and their families. Such indicators are, in part, how the curative/restorative versus palliative care models evolved. The original pain scale monitoring is such an example of how monitoring can lead to improved care. Levels of pain were assessed with vital sign monitoring. The original results were many pain levels but no corresponding interventions. The patient with a pain level of 5 was documented, for example, but there the issue ended. Circumstances like these have led to more clinical pathways and more aggressive pain-management interventions. The same can be true of terminal dyspnea. While we assess it, how do we address it? The design of a quality monitor can include the interventions introduced to abate the level of distress. Both the patients and families are educated to use specific interventions. The use of a medical gas does not always need to be oxygen. This can affect how the medical gas is delivered, how it needs replacement for refill or maintenance, and how comfortable the patient and family are with such a device in the home.

Many institutions require an assessment of hypoxemia for the prescription of oxygen yet allow its use in nonhypoxemic terminally ill patients anyway. A 3-month follow-up is required whenever oxygen is prescribed. Using air might allow some patients to be less dyspneic and allow a home care company, institution, or physician to more easily meet regulatory issues while caring for their patients.

Angela T. Lorenzo, MS, RRT, RPFT, is clinical supervisor, respiratory care services, VA New York Harbor Healthcare System, and adjunct faculty, respiratory care, Long Island University, Brooklyn, NY. For further information, contact [email protected]


  1. Cleary JF, Carbone PP. Palliative medicine in the elderly. Cancer. 1997;80:1335-47.
  2. World Health Organization. Cancer Pain Relief and Palliative Care: Report of a WHO Expert Committee. Technical Report Series No. 804. Available at: Accessed January 7, 2011.
  3. Lanken PN, Terry PB, Delisser HM, et al. American Thoracic Society. An official American Thoracic Society clinical policy statement: palliative care for patients with respiratory diseases and critical illnesses. Am J Respir Crit Care Med. 2008;177:912-27.
  4. Curtis JR, Engelberg RA, Wenrich MD, et al. Missed opportunities during family conferences about end-of-life care in the intensive care unit. Am J Respir Crit Care Med. 2005;171:844-9.
  5. Bruera E, de Stoutz N, Velasco-Leiva A, Schoeller T, Hanson J. Effects of oxygen on dyspnoea in hypoxaemic terminal-cancer patients. Lancet. 1993;342:13-4.
  6. Abernethy AP, McDonald CF, Frith PA, et al. Effect of palliative oxygen versus room air in relief of breathlessness in patients with refractory dyspnoea: a double-blind, randomised controlled trial. Lancet. 2010;376:784-93.
  7. Pressure Conversion. Available at: Accessed January 6, 2011.