H. Lee Moffitt Cancer Center & Research Institute

Introduction

In oncology, as in other areas of medicine, ideal practice provides both excellent disease-modifying therapy and high-quality, symptom-focused, supportive palliation to patients through all stages of disease. In practice, however, treatment is often separated into phases oriented toward goals that may appear mutually exclusive. In the initial active-treatment phase, reversing the course of disease is usually the primary goal, and such unfortunate sequelae as pain and anxiety, whether disease- or treatment-related, are tolerated and even accepted by patients, families, and physicians. Later in the course of illness, if the patient and providers come to accept that the probabilities of survival and return to acceptable levels of function are increasingly unlikely, a transition to supportive care may occur. Goals of treatment may then shift to relief of symptoms, psychosocial support, and other aspects of palliative care. Indeed, the changing needs of patients and families confronting advanced life-limiting illness may justify a palliative strategy that is distinct from "curative" treatment, ie, hospice, as the end of life approaches.

This paper reviews current concepts in estimating survivability for patients with cancer, as well as other medical conditions that may present either as comorbidities with malignancy or as primary noncancer diagnoses. Aspects of palliative treatment for selected noncancer diseases are also discussed.

Estimating Survivability

Because of the distinction in practice between active treatment and palliation at the end of life, defining selection criteria for end-of-life services assumes critical importance. Unless patients and families can hear that the likelihood of survival is declining so that informed decision-making can occur, aggressive treatment may proceed without regard to its potential futility, attendant physical and emotional distress, and high cost.

Survivability estimates are not precise predictions of prognosis, which provide a "rate and a date." Sophisticated computer-based algorithms such as APACHE (Acute Physiology, Age, and Chronic Health Evaluation)¹ or the SUPPORT prognostic model (Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment)² used patient data as input to provide individual prognoses in the format: "x% probability of surviving y months." However, the accuracy of these programs has been limited, and clinicians have been reluctant to stop potentially life-extending treatment on the basis of these probabilities.³ At any rate, many authorities believe that in individual cases, prediction of prognosis precise enough to drive treatment plans will remain an unattainable goal for the foreseeable future.⁴

Estimates of survivability, on the other hand, are often exercises in clinical judgment and intuition that can be done only by physicians with intimate knowledge of the patient.⁵ Fortunately, aid is available for physicians in the form of guidelines from the National Hospice Organization (NHO) that help to define when survivability may become questionable in the trajectory of selected malignant and noncancer diseases.

Prognostic estimates, although challenging, are necessary to determine hospice eligibility. In order to qualify for hospice, patients must be certified by their physician as "terminal," defined by law as "six-month life expectancy, assuming the disease runs its normal course." NHO’s guidelines for cancer diagnoses are in preparation,⁶ and the 1995 second edition of its Medical Guidelines for Determining Prognosis in Selected Non-Cancer Diseases⁷ has been adapted by the US Health Care Financing Administration (HCFA) as national policy for hospice. Although the NHO guidelines have not been tested for six-month predictive validity, HCFA has held hospices accountable for failing to predict six-month survival accurately, accusing programs of "fraud and abuse" when their patients live longer than this arbitrary period.

Clinicians may choose in individual cases to apply these guidelines, coupled with clinical judgment, as a basis for initiating advance-planning discussions. This is an important step because most patients tend to be overoptimistic in estimates of their own survival. Evidence indicates that many patients, even those with very advanced disease, begin to plan realistically only after they comprehend that their own survival may be limited. For example, Weeks et al⁸ report that patients with lung or colon cancer changed their planning behavior once they understood that their odds of surviving six months were less than 90%. It appears that what is needed is not an accurate prognosis estimate, but rather an acknowledgment of the possibility of dying. Only then can informed consent discussions be initiated in which the "forgotten option" — that of a well-managed end of life — may be introduced by the compassionate and skillful physician.

Guidelines for Cancer

Many factors enter into the clinical assessment of patients with cancer who are under evaluation for end-of-life services. In general, the malignancy should be advanced, defined as stage IV with distant metastases, and progressive, with evidence of increased burden of disease and health care utilization. Usually conventional anticancer therapy has become ineffective, is being given for palliative reasons alone, or has been declined. In some cases, significant nonmalignant comorbid conditions make disease-directed therapy unrealistic. Frequently, the patient or family (or both) have chosen treatment goals focused on comfort and relief of suffering rather than life-prolonging therapy, although this decision-making process may not be complete at the time of referral to hospice or palliative care.

Functional status is a critical component of survivability estimates. In many studies, prognosis worsens in direct relation to decrements in patients’ ability to carry out normal activities of living. Functional status can be measured in many ways. A Karnofsky Performance Status (KPS) score of <=50 or an Eastern Cooperative Oncology Group (ECOG) Scale of >=3 often indicates that function has declined to the point where survivability may be in question. Data in this section, of course, refer to cancer patients rather than those with nonmalignant conditions.

Functional status carries therapeutic as well as prognostic weight. As disease worsens, hospice or palliative services provided in the home serve to optimize patients’ functioning, augmenting quality of life in ways that are often not possible with "curative" therapy. In fact, further active treatment may often lead to decrements in function that may not be acceptable to patients or families.

Additional factors influencing prognosis include (but are not limited to) weight loss, symptoms of advanced disease such as dysphagia, dyspnea or anorexia, hypercalcemia, malignant effusions, symptomatic brain metastases, obstructive jaundice, and carcinomatosis of the meningeal or pleural surfaces.

Disease-specific guidelines for common solid tumors and hematologic malignancies are beyond the scope of this review. They are available by request from the NHO.

Comorbid Medical Conditions

Primary diseases of the cardiovascular, pulmonary, and other systems may influence prognosis in cancer as well as complicate its treatment. Certain malignancies have risk factors in common with noncancer diseases, and so they often occur and must be treated in tandem. Lung cancer and emphysema, both related to tobacco abuse, are familiar examples. Less commonly seen but no less troublesome to manage in advanced stages are mesothelioma and emphysema, both related to asbestos exposure with cigarette smoking as a cofactor.

NHO guideline criteria for referral to hospice or palliative care are reviewed for heart and lung disease and dementia. Although relatively few cancer patients manifest advanced noncancer disease to the degree outlined here, many oncologists in practice are faced with decisions concerning general medical patients with end-stage nonmalignant disease who would benefit from palliative services. Since hospice first evolved as terminal care for patients with cancer pain and other symptoms, its evolution into noncancer disease is relatively new. In fact, many physicians are not aware that hospice services are available for patients with primary noncancer diagnoses.

Prognostic Considerations in Noncancer Disease

While many noncancer patients do not carry a primary diagnosis, they are nonetheless on a downward trajectory due to multiple organ system failure, alcohol or other substance abuse, the effects of aging, or other factors. As in cancer, functional status plays an important role in prognosis. As patients lose the capacity for self-care, survivability begins to decrease. Activities of daily living (ADLs), defined as feeding, bathing, dressing, transferring out of bed or chair, going to the bathroom, and maintaining urinary and fecal continence, are common indices of functional status. Nutritional status is significant as well; an unintentional loss of 10% of body weight over six months is associated with poor prognosis independent of diagnosis.⁹

The final common pathway for most patients with advanced cardiovascular disease is congestive heart failure. The primary problem in most cases is systolic dysfunction and symptoms secondary to fluid overload, although many elderly patients without correctable coronary lesions may present with diastolic dysfunction and congestive symptoms or intractable chest pain.

Accurate estimation of prognosis is challenging because clinical declines may be precipitous due to acute episodes of pulmonary edema, in contrast to the inexorable but more predictable course of cancer. To further confound prognosis as patients progress to the end stages of heart disease, fluid overload may become more refractory in some cases, whereas others die of acute and unpredictable ventricular arrhythmias. Also, in advanced disease, exacerbations respond unpredictably to emergent treatment with diuretics, vasodilators, and pressors.

NHO guideline criteria for end-stage heart disease require patients to be (1) symptomatic at rest or with minimal exertion, with dyspnea or recurrent chest pain (New York Heart Association [NYHA] class IV) and (2) already optimally treated with oral diuretics and vasodilators, usually angiotensin-converting enzyme (ACE) inhibitors. In patients with primarily systolic dysfunction, ejection fractions should be less than 20%. Frequently, these patients have undergone trials of intravenous inotropes such as dobutamine or milrinone and either no longer respond well or have refused further aggressive or invasive treatment.

As in heart disease, prognosis is difficult in end-stage pulmonary disease. Once again, precipitous exacerbations are the rule. Even in end-stage disease, long periods of stability may be seen between admissions to the hospital for intubation and mechanical ventilation.

As in other nonmalignant conditions, survival is difficult to predict in advanced dementia. Patients who have lost most cognitive function may survive for long periods with meticulous care. Ironically, care this devoted is poorly supported. Resource allocation for home care of patients with Alzheimer’s disease is so inadequate that many families may become bankrupt both financially and emotionally as the disease progresses toward the end stages. For this reason, supportive options such as hospice and palliative care are often welcomed.

Indeed, palliative approaches become more appropriate from a medical standpoint as dementia advances. Aggressive measures to reverse the medical complications of severe dementia, which are usually infectious, become less effective as the disease reaches its last stages. Extensive workups for fever are frequently inconclusive, and although antibiotics may cause patients to defervesce, infections often recur after short intervals since it is frequently impossible to correct the underlying cause. Pneumonia and sepsis, for example, often recur shortly after antibiotic therapy because nocturnal aspiration of oral secretions cannot be prevented in demented bed-bound patients with blunted swallow reflex.

NHO criteria for dementia include advanced disease and the onset of medical complications. Dementia is sufficiently advanced when patients lose the functions of independent ambulation and the ability to hold meaningful conversation. Common medical complications include aspiration pneumonia, urosepsis, decubitus ulcers of stage III or IV that progress despite therapy, fevers recurrent after antibiotics, and significant weight loss in patients with or without feeding tubes. Nutritional deficiency can occur in patients with feeding tubes when even devoted caregivers begin to feel ambivalent about force-feeding severely demented patients.

Guidelines have also been published for other diagnoses such as HIV and AIDS, liver disease, renal failure, stroke, coma, and amyotrophic lateral sclerosis. These are available on request from the NHO.

Prognosis in patients with CHF remains unfavorable and poor despite aggressive therapy with diuretics, vasodilators and, most recently, beta-blockers.¹⁰ In the Veterans Administration Heart Failure Trial I,¹¹ cumulative mortality at four years in patients with only mild to moderate heart failure was 49.7%, a mortality rate worse than that of many cancers.

Symptoms in CHF are predominantly due to circulatory congestion, which results from a decrease in left ventricular systolic wall motion that can be segmental (in the case of ischemic heart disease) or global due to cardiomyopathy (such as that resulting from overexposure to anthracyclines). Although diastolic dysfunction due to decreased left ventricular compliance and consequent filling difficulties is responsible for a large proportion of cases of mild to moderate severity, most end-stage CHF is due to systolic dysfunction.

Although nonpharmacologic measures such as reduction of salt intake are helpful, management of end-stage CHF requires administration of adequate diuretics to relieve circulatory overload. Almost all patients require doses of loop diuretics such as furosemide in once- or twice-daily doses as high as 240 mg per day. Adequacy of diuresis can be easily assessed non-invasively at the bedside by testing for hepatojugular reflux. With the patient sitting upright at approximately 30°, the right upper quadrant is compressed — which may reveal an enlarged and tender liver due to congestion — and if elevated left ventricular filling pressure is present, increased jugular venous distention will be seen. This indicates the need for a higher diuretic dose.

When venous pressure remains high despite increasing doses of loop diuretics and sodium restriction, the addition of metolazone 2.5 to 5 mg per day given one hour prior to furosemide may result in diuresis that in fact can be unexpectedly brisk. Since pronounced kaliuresis may result, serum potassium levels should be monitored and potassium replacement provided if serum levels fall below 4 mg/dL. For many patients on furosemide alone, however, administration of a potassium-sparing diuretic such as spironolactone or amiloride can obviate the need for oral potassium supplements, and in fact some older patients do not require potassium supplements at all.

Overdiuresis causing volume depletion can be as harmful to comfort and quality of life as fluid overload can be. Excessive fatigue, hypotension, or azotemia with normal jugular venous pressure may call for a reduction in diuretic therapy. Since diuretic requirements can vary with diet and activity level, patients should be counseled to weigh themselves each morning and adjust their diuretic dose to maintain weight within a range in which their symptoms of pulmonary congestion — dyspnea, orthopnea, or paroxysmal nocturnal dyspnea — are minimal. Using control of peripheral edema as an endpoint, however, can result in overdiuresis since this finding may be a result of peripheral venous insufficiency rather than CHF.

Vasodilators, including ACE inhibitors, relax both arterial and venous smooth muscle, thus reducing venous capacitance (preload) and resistance to left ventricular ejection (afterload). These drugs are important for palliation in CHF because of beneficial effects on CHF symptoms and quality of life. Their survival benefits are also well documented.¹² Another clinical advantage of ACE inhibitors in CHF is their conservation of potassium through inhibition of aldosterone secretion, often preventing the hypokalemia caused by diuretic administration and obviating the need for potassium supplements.

Low initial doses of ACE inhibitors are recommended (eg, 2.5 mg of enalapril a day) particularly if the patient’s serum sodium is less than 135 mg/dL, which, in the absence of other causes of hyponatremia, indicates a high level of plasma renin activity.¹³ Moderate asymptomatic hypotension or azotemia (eg, serum creatinine less than 2.5 mg/dL) is acceptable with ACE inhibitors and is an indication to reduce the diuretic dose if jugular venous pressure is normal. Although symptomatic hypotension or progressive renal insufficiency may force discontinuation of ACE inhibitors, this is seldom necessary. In fact, evidence indicates that ACE inhibitors are often underused and underdosed in patients with advanced CHF.¹⁴ Target doses of 10 mg of enalapril twice a day are tolerated well by the majority of NYHA class IV patients.¹⁵

Although the benefits of digoxin have been questioned,¹⁶ evidence indicates that withdrawal of the drug can worsen CHF symptoms even in patients treated with ACE inhibitors.¹⁷ Patients who are already taking the drug should probably continue it at doses low enough to avoid digitalis toxicity, which can be difficult to diagnose in patients already seriously ill with cancer, heart disease, or both.

Recent studies of beta blockers, such as carvedilol, for CHF indicate that their use should be restricted to the early stages of CHF, where they may have a beneficial effect on mortality.¹⁸ From the standpoint of symptom relief, although these drugs may confer benefit in patients with mild to moderate disease,¹⁹ they may cause significant worsening of CHF in patients with severe, unstable disease and are therefore contraindicated pending further study.

Implanted defibrillators, effective in reducing mortality in patients with mild to moderate heart failure who have survived an episode of ventricular tachyarrhythmia, often interfere with palliative care for patients with end-stage heart failure. Consideration should be given to turning off these devices after suitable discussion with the patient and family. They may prolong or even add to the discomfort these patients experience as they approach the end of life.

As heart failure progresses, whether or not it accompanies terminal cancer as a comorbid condition, opioids should be used for palliation of dyspnea. The physiologic benefits of opioids in CHF are twofold. First, subjective improvement in breathlessness results from opioid action on midbrain centers that trigger the sensation of dyspnea in response to arterial hypoxemia. Second, morphine and other opioids lower vasomotor tone in the large veins of the extremities, increasing capacitance and decreasing preload, therefore relieving congestive symptoms. As in control of pain, oral administration is effective in most cases, starting at low doses and titrating upward until dyspnea is relieved. Slow-release preparations can then be substituted every 12 hours.

For patients whose failure has become refractory to ACE inhibitors and high-dose diuretics, opioid doses may be increased and the route of administration changed to rectal or parenteral if necessary. However, sedatives such as barbiturates or benzodiazepines should be added for severe terminal dyspnea or agitation, which may be seen for instance in patients with end-stage heart failure due to valvular disease. Opioid toxicity can be troublesome if these agents are used alone in high doses.

Although guidelines have been available for reversible bronchospasm (asthma) for some time, consensus is only now emerging to guide therapy of COPD induced by cigarette smoking.²⁰ Although management principles are different for the two diseases, they often are confused in practice.

Airflow limitation in COPD occurs as a result of destruction of elastic tissue in respiratory bronchioles proximal to the alveoli. In late-stage disease, obstruction to expiration occurs as these small airways collapse with the onset of positive intrathoracic pressure as the muscles of the chest wall and diaphragm first contract. In contrast, the reversible bronchospasm of asthma results from inflammation of small airways with infiltrates composed predominantly of lymphocytes and eosinophils, combined with increased smooth muscle tone that causes decreased airway diameter. Although airway inflammation may also be responsible for some of the bronchospasm of COPD, the infiltrates, if any, are composed of neutrophils.

These findings have implications for bronchodilator therapy in COPD, particularly for the use of corticosteroids. Whereas these agents are a therapeutic mainstay in asthma, corticosteroids do not significantly influence airway inflammation and the clinical course of COPD in the great majority of patients. Only about 10% of COPD patients benefit by bronchodilation from systemic or inhaled corticosteroid therapy.²¹ Only a two-week trial of prednisone at a dose of 0.5 to 1 mg per day can demonstrate corticosteroid responsiveness. Pre- and post-pulmonary function tests should reveal an improvement in post-bronchodilator FEV₁ of at least 20% to 30% before the clinician can be confident that these agents are causing bronchodilation.²² They can then be administered chronically in alternate-day regimens to avoid toxicity such as myopathy, which can cause problems with ambulation in patients who are often already debilitated. Corticosteroid-related deficits in functional status are difficult to justify, considering that 90% of patients with COPD do not respond to these drugs.

Inhaled quaternary anticholinergic agents such as ipratropium bromide, administered by metered-dose inhaler or nebulizer, produce greater bronchodilation in COPD than that seen with beta-2 agonists.²⁵ In addition, the sensitivity of the adrenergic system decreases with age, necessitating the use of increasing doses of beta agonist, whereas no such age-related decrement in responsiveness is seen in the cholinergic system’s response to ipratropium.²⁶ Ipratropium is poorly absorbed after inhalation. Consequently, few cardiovascular or atropine-like side effects are seen.²⁷ Starting doses of 4 to 6 puffs of ipratropium by a metered-dose inhaler or, if a mucolytic effect is desired, by a nebulizer with saline may obviate the need for combination bronchodilator therapy.²⁸

The use of theophylline in both asthma and COPD has diminished since it produces minimal bronchodilation but places patients at significant risk for toxicity.²⁹ It may have an adjunctive role in patients who have persistent major airflow obstruction with maximal use of inhaled bronchodilators. If it is used, the dose should be conservative, with a target serum level of 5 to 12 mg/L, since most of the bronchodilator effect occurs in this range.³⁰ Postulated nonbronchodilator benefits of theophylline such as central respiratory stimulation or strengthening of respiratory muscles can be demonstrated in the laboratory, but their clinical significance is uncertain.

Supplemental oxygen reduces mortality in COPD. In patients with COPD and cor pulmonale, long-term 24-hour oxygen increases life expectancy by six to seven years,³¹ although this benefit may be reduced in patients who continue to smoke. More relevant to palliative considerations, oxygen relieves the discomfort and anxiety of dyspnea associated with hypoxemia and should be offered to any dyspneic patient with oxygen saturation below 88% to 90%.

Even "weak" opioids such as codeine or hydrocodone, available in nonrestricted cough suppressants, can exert a beneficial effect on dyspnea. However, it is preferable in cases of severe shortness of breath to administer morphine or oxycodone every four hours as an oral solution or in the form of tablets, starting at a dose of 2.5 to 5 mg for morphine or approximately half that for oxycodone. Other strong opioids can be used in case of toxicity from these agents. If dyspnea returns prior to the next dose, it is usually better to increase the dose rather than decrease the dosage interval to avoid the necessity of frequent nighttime administration. The dose may be increased every one or two days until dyspnea is relieved. Occasionally, high doses are required but are often well tolerated if the increase is gradual. Once the dose is stabilized, slow-release preparations can be substituted every 12 hours. Nebulized opioids can relieve breathlessness in COPD and are especially helpful in patients with severe cough associated with lung cancer, particularly if they are already trained in the use of a nebulizer.

Advance planning, preferably initiated by the physician and carried on by a hospice or palliative care team, is mandatory for patients who have been intubated in the past and who state they wish to avoid this treatment for the next exacerbation. Considerable reassurance can be provided through careful explanations of the effects of opioids and sedatives to relieve dyspnea and anxiety, thus facilitating a relatively comfortable end of life. Advance directives, prominently displayed, can prevent unwanted aggressive measures that are often reflexively provided by paramedics and emergency personnel to patients with exacerbations.

Many physicians believe that once therapeutic options have been exhausted, nothing more can be done for patients with advanced disease. Unfortunately, this attitude deprives patients, their families, and the physicians themselves of the opportunity to find comfort and to experience the benefits of care that, even if provided in the patient’s home, may be as intensive as that provided in the hospital. The key to a well-managed end of life is careful attention to changing goals of treatment from curative to supportive early enough in the disease process to allow patients and families the time to benefit from palliative care. When events are managed well in alignment with patient and family preferences, all participants may come to remember the end of life as a time that can be challenging, but is often extraordinarily meaningful.

4. Thibault G. Prognosis and clinical predictive models for critically ill patients. In: Field MJ, Cassel CK, eds. Approaching Death: Improving Care at the End of Life. Committee on Care at the End of Life, Division of Health Care Services, Institute of Medicine. Washington, DC: National Academy Press; 1997:358-362.

5. Boult C, Boult L, Pacala JT. Systems of care for older populations of the future. J Am Geriatr Soc. 1998;46:499-505.

6. National Hospice Organization Standards and Accreditation Committee, Medical Guidelines Task Force. Medical Guidelines for Determining Prognosis in Selected Cancer Diseases. 1999. In press.

7. National Hospice Organization Standards and Accreditation Committee, Medical Guidelines Task Force. Medical Guidelines for Determining Prognosis in Selected Non-Cancer Diseases. 2nd ed. Arlington, Va: National Hospice Organization; 1996.

8. Weeks JC, Cook EF, O’Day SJ, et al. Relationship between cancer patients’ predictions of prognosis and their treatment preferences. JAMA. 1998;279:1709-1714.

9. Murden RA, Ainslie NK. Recent weight loss is related to short-term mortality in nursing homes. J Gen Intern Med. 1994;9:648-650.

10. Chatterjee K. Heart failure therapy in evolution. Circulation. 1996;94:2689-2693.

11. Cohn JN, Archibald DG, Ziesche S, et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure: results of a Veterans Administration Cooperative Study. N Engl J Med. 1986; 314:1547-1552.

12. Johnson G, Carson P, Francis GS, et al. Influence of prerandomization (baseline) variables on mortality and on the reduction of mortality by enalapril. Veterans Affairs Cooperative Study on Vasodilator Therapy of Heart Failure (V-HeFT II). Circulation. 1993;87 (suppl):VI32-39.

13. Levine TB, Franciosa JA, Vrobel T, et al. Hyponatraemia as a marker for high renin heart failure. Br Heart J. 1982;47:161-166.

14. Visser FC, Visser CA. Current controversies with ACE inhibitor treatment in heart failure. Cardiology. 1996;87:23-28.

15. Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). The Consensus Trial Study Group. N Engl J Med. 1987;316:1429-1435.

16. Bigger JT Jr, Fleiss JL, Rolnitzky LM, et al. Effect of digitalis treatment on survival after acute myocardial infarction. Am J Cardiol. 1985;55:623-630.

17. Packer M, Gheorghiade M, Young JB, et al. Withdrawal of digoxin from patients with chronic heart failure treated with angiotensin-converting enzyme inhibitors. RADIANCE Study. N Engl J Med. 1993;329:1-7.

18. Bristow MR, Gilbert EM, Abraham WT, et al. Carvedilol produces dose-related improvements in left ventricular function and survival in subjects with chronic heart failure. MOCHA Investigators. Circulation. 1996;94:2807-2816.

19. Packer M, Colucci WS, Sackner-Bernstein JD, et al. Double-blind, placebo-controlled study of the effects of carvedilol in patients with moderate to severe heart failure. The PRECISE Trial: Prospective Randomized Evaluation of Carvedilol on Symptoms and Exercise. Circulation. 1996;94:2793-2799.

20. Chapman KR. Therapeutic approaches to chronic obstructive pulmonary disease: an emerging consensus. Am J Med. 1996; 100:5S-10S.

21. Callahan CM, Dittus RS, Katz BP. Oral corticosteroid therapy for patients with stable chronic obstructive pulmonary disease: a meta-analysis. Ann Intern Med. 1991;114:216-223.

22. Ferguson GT, Cherniack RM. Management of chronic obstructive pulmonary disease. N Engl J Med. 1993;328:1017-1022.

23. Braun SR, McKenzie WN, Copeland C, et al. A comparison of the effect of ipratropium and albuterol in the treatment of chronic obstructive airway disease. Arch Intern Med. 1989;149:544-547.

24. Gross NJ, Bankwala Z. Effects of an anticholinergic bronchodilator on arterial blood gases of hypoxemic patients with chronic obstructive pulmonary disease. Comparison with a beta-adrenergic agent. Am Rev Respir Dis. 1987;136:1091-1094.

25. Gross NJ. Ipratropium bromide. N Engl J Med. 1988;319: 486-494.

26. Chapman KR. The role of anticholinergic bronchodilators in adult asthma and chronic obstructive pulmonary disease. Lung. 1990;168:295-303.

27. Pakes GE, Brogden RN, Heel RC, et al. Ipratropium bromide: a review of its pharmacological properties and therapeutic efficacy in asthma and chronic bronchitis. Drugs. 1980;20:237-266.

28. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. This official statement of the American Thoracic Society was adopted by the ATS Board of Directors, November 1986. Am Rev Respir Dis. 1987;136:225-244.

29. Bleecker ER, Britt EJ. Acute bronchodilating effects of ipratropium bromide and theophylline in chronic obstructive pulmonary disease. Am J Med. 1991;91:24S-27S.

30. Guidelines for the assessment and management of chronic obstructive pulmonary disease. Canadian Thoracic Society Workshop Group. CMAJ. 1992;147:420-428.

31. Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease: a clinical trial. Ann Intern Med. 1980;93:391-398.

From the Visiting Nurse Association and Hospice of Northern California, Santa Rosa, Calif.

Address reprint requests to Brad Stuart, MD, Medical Director, Visiting Nurse Association & Hospice of Northern California, 1110 North Dutton Ave, Santa Rosa, CA 95401.

No significant relationship exists between the author and the companies whose products are referenced in this article.