Contemporary Management of Chronic Obstructive Pulmonary Disease:  Scientific Review FREE

Chronic obstructive pulmonary disease (COPD) affects more than 5% of the adult population,¹ and it is the only major cause of death in the United States in which morbidity and mortality are increasing.² Although COPD is currently the 4th-leading cause of mortality and the 12th-leading cause of disability, by the year 2020 it is estimated that COPD will be the 3rd-leading cause of death and the 5th-leading cause of disability worldwide.^3- 4 In 1993, the total economic costs of COPD were estimated to be $24 billion in the United States alone; more than 60% of these costs were direct expenditures related to hospital-based care.⁴ While these figures are alarming, they most certainly underestimate the true health burdens of COPD because airflow obstruction is an important contributor to other common causes of morbidity and mortality, including ischemic heart disease, stroke, pneumonia, and lung cancer.^5- 8Chronic obstructive pulmonary disease is characterized by irreversible airflow obstruction, secondary to airway inflammation and emphysematous changes in the lung parenchyma.⁹ Airway hyperresponsiveness also is a common feature of COPD, affecting 60% to 80% of patients with COPD.¹⁰ In more than 80% of cases, cigarette smoking is causally linked to the development of COPD.⁹ Other risk factors include exposure to noxious gases, ambient pollution, and chronic respiratory infections.^11- 12 A diagnosis of COPD should, therefore, be considered in current or former smokers (or in never smokers with other risk factors) who present with cough, sputum production, or dyspnea, with spirometric evidence of irreversible airflow obstruction.⁹ The latter sign is defined as a postbronchodilator forced expiratory volume in 1 second (FEV₁) value of less than 80% of predicted, in association with an FEV₁to forced vital capacity ratio (FEV₁/FVC) of less than 70%.⁹ Smoking cessation is the single most important therapy for improving health outcomes in patients with COPD. Unfortunately, even in the best programs, less than one third of smokers become "sustained" quitters.^13- 14 Furthermore, once individuals develop demonstrable airflow obstruction, their symptoms (cough and/or dyspnea) may persist even after smoking cessation.⁹

Over the past 20 years, several novel therapies, such as inhaled corticosteroids, long-acting β₂-agonists, and anticholinergics, have been introduced for patients with COPD, with the aim of favorably altering lung function and improving patient symptoms. However, most of the studies that have evaluated these therapies have been short and powered to evaluate the impact on physiological end points, such as FEV₁ values. Relatively few studies have evaluated the long-term effects of these interventions on clinical end points, such as health-related quality of life, COPD exacerbations, hospitalizations, or mortality. Because the rate of descent in FEV₁ is an imperfect surrogate for these clinical end points,¹⁵ we conducted a systematic review of the literature to evaluate the effects of commonly used anti-COPD therapies on patient-centered outcomes, such as health-related quality of life, exacerbations associated with COPD, hospitalizations, and/or mortality.

We decided a priori to examine the published evidence for the following anti-COPD therapies: long-acting β₂-agonists, long-acting inhaled anticholinergics (tiotropium), combination therapy with short-acting β₂-agonists and short-acting anticholinergics, inhaled corticosteroids, combination therapy with inhaled corticosteroids and long-acting β₂-agonists, pulmonary rehabilitation, long-term administration of nocturnal noninvasive mechanical ventilation (NIMV), domiciliary oxygen therapy, and disease management programs (which include any combination of patient education, enhanced follow-up, and/or self-management sessions).¹⁶

For each of these therapies, we conducted a literature search using MEDLINE. We limited our search to English-language articles published from January 1, 1980, to May 1, 2002, reporting studies of adults (>19 years of age) in randomized clinical trials. To identify only studies in COPD, we used the following terms: obstructive or chronic obstructive or bronchitis or pulmonary emphysema or airway obstruction or emphysema or mediastinal emphysema or subcutaneous emphysema or COPD or lung diseases, obstructive* or pulmonary diseases, obstructive. Detailed search terms for each of the therapies and search results are available on the author's Web site (Table 1 e at http://www.mrl.ubc.ca/sin). To supplement this search, we examined the Cochrane Database of Systematic Reviews of Effectiveness as well as bibliographies of published articles and contacted experts in the field. Although we wanted to include only those studies with follow-up times of at least 6 months, we found insufficient numbers of such studies for most of the interventions, so a follow-up time of 3 months was used as the threshold for inclusion (with the exception of pulmonary rehabilitation programs, for which 6 weeks was used as the threshold). Given the controversy over the value of scoring systems for the quality of randomized controlled trials,¹⁷ we did not use a scoring system to adjudicate the quality of the trials included in this review, but we did restrict our analysis to trials that were randomized, placebo-controlled, had blinded ascertainment of end points, had complete or near complete follow-up data, and had baseline characteristics that were well balanced between treatment and control groups. We discarded studies reporting only physiological variables, such as changes in FEV₁, because the correlation between spirometric changes and long-term clinical outcomes in COPD has been shown to be weak.¹⁵ Crossover trials were included in this review. We restricted analysis of health-related quality of life to 2 well-standardized and validated instruments in COPD: St George's Respiratory Questionnaire (SGRQ)¹⁸ and Chronic Respiratory Questionnaire (CRQ).¹⁹ These instruments quantify the extent of physical and psychological impairments related to COPD and allow investigators to determine the (beneficial) effects of specific interventions on the functional status of patients with COPD.¹⁸

Where possible, for each end point we combined the results from individual studies to produce summary effect estimates (risk ratios). Heterogeneity of results across individual studies was checked using the Cochran Q test. If significant heterogeneity was observed (P<.05), we used the Dersimonian and Laird random-effects model to combine the results; otherwise, a fixed-effects model was used. As part of a sensitivity analysis for the latter situation, we used a random-effects model to determine the robustness of the data. In all cases, the results obtained from the random-effects model and fixed-effects model were similar. Continuous variables were merged using weighted mean difference technique. The use of the standardized mean difference technique produced similar results. All analyses were conducted using Review Manager version 4.1 (Revman; The Cochrane Collaboration, Oxford, England). For completeness, we also present a general overview of other anti-COPD therapies, such as smoking cessation, oral theophyllines, surgical procedures, and vaccine therapies.

Smoking cessation is the only therapy proven to slow the accelerated decline in lung function related to COPD.¹⁴ While the rate of FEV₁ decline is approximately 60 mL per year in smokers, it is only approximately 30 mL per year among ex-smokers.¹⁴ Moreover, smoking cessation reduces all-cause mortality rates by approximately 27% (95% confidence interval [CI], 1%-47%), driven largely by very significant reductions in cardiovascular mortality (relative risk [RR] compared with continued smokers, 0.54; 95% CI, 0.32-0.92).²⁰ However, smoking cessation is difficult to achieve and even more difficult to sustain over the long term.²¹ A single physician recommendation for smoking abstinence is associated with an approximate 5% long-term abstinence rate.²² If abstinence is followed by cessation counseling, education, and nicotine replacement therapy (or treatment with antidepressants such as bupropion and nortriptyline), the long-term abstinence rates can be as high as 25% in patients with early COPD.^23- 24 Thus, physicians should make every attempt to convince their patients to stop smoking. A detailed evaluation of various smoking cessation methods and programs is beyond the scope of this article, but they are available elsewhere.^25- 28 Although smoking cessation improves the natural history of COPD, once COPD becomes clinically apparent many sustained quitters remain symptomatic and their airways are persistently inflamed.²⁹ Thus, in symptomatic patients with COPD, additional therapies are indicated.⁹

Although little evidence is available for the usefulness of influenza and pneumococcal vaccinations for patients with COPD per se, they have been demonstrated in the general elderly population to reduce all-cause, pneumonia, and cardiac hospitalizations^30- 31 and deaths³¹ by 30% to 40% with only minor excess risks to recipients.³⁰ Since most patients with COPD are elderly and also are at increased risk of hospitalizations and mortality from various cardiovascular conditions and pneumonia,³² influenza and pneumococcal vaccination also should be instituted for most patients with COPD.

Bronchodilators. Airflow obstruction is present in all patients with COPD.⁹ Accordingly, bronchodilators are used almost universally to provide symptomatic relief for patients with COPD.⁹ Traditionally, inhaled short-acting β₂-agonists and short-acting anticholinergics most commonly were used either on an as-needed basis for rescue care or on a regular basis to prevent or to reduce symptoms.³³ Although these medications produce only a modest improvement in lung function in patients with COPD, studies have demonstrated that they reduce symptoms and improve exercise tolerance.^34- 36 However, no evidence shows that these medications have any effect on the rate of decline in lung function¹⁴ or survival.³⁷

Because β₂-agonists and anticholinergics produce bronchodilation through different pathways, combination products were introduced in the hopes of achieving greater bronchodilation (and, in turn, better symptom control) than with monotherapy. In 3 trials (1399 patients with advanced COPD, mean FEV₁ approximately 1 L in all 3 studies) with follow-up of 3 months or longer, combination therapy with a short-acting β₂-agonist and an anticholinergic resulted in 32% (95% CI, 9%-49%) fewer COPD exacerbations compared with monotherapy with a short-acting β₂-agonist.^38- 41 However, combination therapy was not superior to monotherapy with ipratropium (Table 1). Mortality rates over the 3-month follow-up were low and similar among patients treated with combination therapy, short-acting β₂-agonist, or ipratropium monotherapy. We did not identify any published studies wherein combination therapy was compared directly with a placebo.

Long-acting β₂-agonists were introduced several years ago to achieve longer and more predictable improvements in lung function than what was possible with short-acting β₂-agonists.⁴² Nine placebo-controlled clinical trials (4198 patients with moderate to severe COPD and followed up for ≥3 months) demonstrated a 21% reduction (95% CI, 10%-31%) in COPD exacerbation rates (Figure 1).^43- 51They also improved health-related quality of life of patients with COPD (SGRQ, 2.8-unit improvement; 95% CI, 1.6-4.1 vs placebo; CRQ, 4.3-unit improvement; 95% CI, 1.6-7.0). The RR for all-cause mortality compared with placebo was 0.76 (95% CI, 0.39-1.48). The effect on FEV₁ was variable. In the 2 trials that evaluated FEV₁ changes over at least 1 year, long-acting β₂-agonists nonsignificantly increased trough FEV₁, on average, by 82 mL (95% CI, −26 to 190 mL per year) compared with placebo.^52- 53

Five clinical trials of tiotropium (3574 patients with moderate to severe COPD) uniformly have demonstrated a beneficial effect in reducing exacerbation rates compared with either placebo (RR, 0.74; 95% CI, 0.62-0.89) or with ipratropium bromide (RR, 0.78; 95% CI, 0.63-0.95) (Figure 2).^55- 59 Tiotropium also improved health-related quality of life relative to placebo (SGRQ, 2.9-unit improvement; 95% CI, 1.5-4.3). However, no convincing data to date have shown that they are superior (or inferior) to long-acting β₂-agonists in reducing exacerbation rates or improving health status in patients with moderate to severe COPD (RR for exacerbations vs long-acting β₂-agonists, 0.93; 95% CI, 0.80-1.08). Moreover, the current trials are too short and underpowered to evaluate the effects of these drugs on all-cause mortality. Long-acting anticholinergics have a powerful effect on FEV₁. In the 2 trials that had 1-year follow-up, the trough FEV₁ increased by 121 mL (95% CI, 102-141 mL per year) compared with placebo or ipratropium monotherapy.^55,59 In the 2 trials that compared long-acting anticholinergics with long-acting β₂-agonists over 6 months, long-acting anticholinergics had a more favorable effect on trough FEV₁ (37 mL; 95% CI, 12-61 mL).^56- 57

Inhaled Corticosteroids With and Without Long-Acting β₂-Agonists. The use of inhaled corticosteroids remains one of the most contentious issues in COPD pharmacotherapy.^60- 61 In 6 placebo-controlled trials (1741 patients) with at least a 6-month follow-up period, inhaled corticosteroids led to a 24% reduction in COPD exacerbations (95% CI, 20%-28%) (Table 2).^62- 69 Importantly, this beneficial effect was modified by disease severity, as measured by FEV₁ (Figure 3).^62- 66,68Whereas the study that had the highest mean FEV₁ value failed to demonstrate a beneficial effect of inhaled corticosteroids,⁶⁸ trials that had a mean FEV₁ of less than 2.0 L (or <70% of predicted) almost uniformly demonstrated a positive effect of inhaled corticosteroids on exacerbations, regardless of the duration of the study or the specific formulation used.The pooled RR for COPD exacerbation among trials enrolling patients with a mean FEV₁ of 2.0 L or less was 0.75 (95% CI, 0.71-0.80) compared with an RR of 0.96 (95% CI, 0.77-1.20) in a trial with an FEV₁ greater than 2.0 L. The effect of inhaled corticosteroids on mortality is uncertain. However, a trend was observed toward reduced mortality in patients randomized to inhaled corticosteroid therapy (Table 2). In a sensitivity analysis, we added the inhaled corticosteroid and placebo data from 2 trials that contained 3 treatment groups (inhaled corticosteroids, inhaled corticosteroids/long-acting β₂-agonists, and long-acting β₂-agonists) to the original steroid analysis.^52,54 The results were materially unchanged (RR for exacerbation, 0.76; 95% CI, 0.73-0.80; RR for mortality, 0.75; 95% CI, 0.57-1.00). Inhaled corticosteroids also decelerated the rate of decline in health status (SGRQ, 1.4-unit improvement relative to placebo; 95% CI, 0.6-2.1; Table 2).

The reporting of adverse effects related to corticosteroid use varied across the studies. Six studies reported the incidence of oral thrush, and its risk was increased among users of inhaled corticosteroids (RR, 2.98; 95% CI, 2.09-4.26)^{53- 54,64- 65,67,69}; 4 studies reported incidence of dysphonia (RR, 2.02; 95% CI, 1.43-2.83)^{53,64- 65,69}; 3 studies reported incidence of bruising (RR, 1.62; 95% CI, 1.18-2.22)^53,64,67; and 2 studies reported the risk of cataract (RR, 1.05; 95% CI, 0.84-1.31).^64,67 Bone mineral density data from the femoral neck and lumbar spine were reported by the Lung Health Study and EUROSCOP.^67,70 Over 3 to 4 years of follow-up (N = 972), a net reduction of 1.57% (95% CI, 2.40%-0.74%) and 1.07% (95% CI, 1.86%-0.28%) was reported in the bone mineral density of the femoral neck and lumbar spine, respectively, compared with placebo. No excess risk of fractures was reported within a 3-year follow-up period (RR, 0.70; 95% CI, 0.36-1.37).^67,70 The life-time risk of fractures in those patients who continue to use inhaled corticosteroids for longer than 3 to 4 years is unknown.

Inhaled corticosteroids have a modest effect on FEV₁. In the first 6 months of therapy, they increased baseline trough FEV₁, on average, by 45 mL (95% CI, 22-69 mL) in the 6 largest trials.^64- 69 However, after the initial 6 months, the rate of FEV₁ decline was unaffected by inhaled corticosteroid therapy (5 mL/y; 95% CI, −1 to 11 mL/y relative to placebo) in the 4 trials reporting this outcome over that time frame.^64,67- 69

Three clinical trials (2951 patients) demonstrated that combination therapy with inhaled corticosteroids plus long-acting β₂-agonists is associated with lower exacerbation rates compared with monotherapy with long-acting β₂-agonists (RR, 0.80; 95% CI, 0.71-0.90) or with placebo (RR, 0.70; 95% CI, 0.62-0.78).^52- 54 A trend was observed toward decreased COPD exacerbations compared with inhaled corticosteroid monotherapy, but it did not achieve statistical significance (Table 3).^52- 54 The beneficial effects of inhaled corticosteroids and long-acting β₂-agonists appeared, therefore, to be additive, not synergistic. The effect of this combination on mortality is uncertain (RR vs placebo, 0.52; 95% CI, 0.20-1.34). Combination therapy, however, was effective in improving trough FEV₁ compared with placebo (101 mL/y; 95% CI, 76-126 mL/y), long-acting β₂-agonists (34 mL/y; 95% CI, 11-57 mL/y), and inhaled corticosteroids (50 mL/y; 95% CI, 26-74 mL/y).

Respiratory muscle fatigue and dynamic hyperinflation commonly are observed in patients with severe COPD.^71- 73 Even at rest patients with COPD work harder than patients without COPD because they have to overcome dynamic lung hyperinflation and airflow obstruction, which limit their tidal volume.^72- 73 Long-term NIMV therapy potentially should unload the inspiratory muscles of respiration and help restore depleted energy stores and partially reverse respiratory muscle fatigue.⁷⁴ Moreover, nocturnal NIMV may improve central ventilatory drive and responsiveness to chemical and mechanical stimulation by lowering nocturnal PCO₂ levels and improving daytime sleepiness, which help the body to reset the ventilatory thresholds to more normal levels.⁷⁵ Despite these compelling physiological arguments, the long-term use of NIMV cannot be recommended at this time because there is insufficient clinical trial evidence for its efficacy (N = 142; RR for exacerbation/hospitalization vs usual care, 0.87; 95% CI, 0.67-1.12) (Table 4e at http://www.mrl.ubc.ca/sin).^76- 79

Patients with advanced COPD experience marked dyspnea and exercise intolerance related in part to generalized muscle weakness and dystrophy, cardiac impairments (eg, cor pulmonale), and nutritional deficiencies.⁸⁰ Pulmonary rehabilitation programs were developed to address some of these adverse physiological changes in patients with COPD.⁸⁰Although the contents of pulmonary rehabilitation vary from center to center, most programs contain 4 major components: exercise training, education, behavioral modification, and outcome assessment.⁸⁰ The training intensity of the exercise program is heterogeneous. In general, most aerobic training is targeted at 60% to 90% of the predicted maximal heart rate for about 30 minutes.⁸⁰ Although most programs emphasize endurance training, some centers also incorporate strength and respiratory muscle training. The 19 clinical trials (1447 patients) we identified support the notion that pulmonary rehabilitation improves the health status of patients with moderate to severe COPD (mean FEV₁, 1.0 L; mean age, 66 years) and increases their exercise tolerance beyond that achieved by standard care alone, at least during the time patients are engaged in the rehabilitation program (Table 5e at http://www.mrl.ubc.ca/sin; Figure 4).^81- 98 Although some heterogeneity was observed in the results (likely arising from the variations in the components of pulmonary rehabilitation program across the studies), pulmonary rehabilitation, on average, improved CRQ scores (dyspnea domain) by 4.1 units and SGRQ (total) scores by 4.4 units. Because clinically relevant thresholds for CRQ and SGRQ are considered to be score changes of 0.5 or higher per question for CRQ⁹⁹ and 4.0 units or higher for SGRQ,¹⁰⁰ the overall effect of pulmonary rehabilitation on the health status of patients with COPD was both clinically and statistically significant. These gains were maintained over the ensuing 3 to 18 months of follow-up (2 trials reported long-term CRQ [dyspnea domain] data, 1.11; 95% CI, 0.31-1.90^85- 86; and 2 trials reported long-term SGRQ data, −5.55; 95% CI, −8.76 to −2.33).^81,86 Pulmonary rehabilitationdid not have any significant effect on mortality (RR, 0.90; 95% CI, 0.65-1.24)^{85- 86,88- 89,92- 93,95,98} or on hospitalization rates (RR, 0.99; 95% CI, 0.56-1.75).⁸⁶

Supplemental oxygen is effective in prolonging survival of patients with COPD whose resting PaO₂ is lower than 60 mm Hg at sea level (2 trials with 290 patients, RR, 0.61; 95% CI, 0.46-0.82) (Table 6e at http://www.mrl.ubc.ca/sin).^101- 102 However, in trials in which the resting mean PaO₂ was 60 mm Hg or higher, no survival advantage was related to oxygen therapy (2 trials with 211 patients, pooled RR, 1.16; 95% CI, 0.85-1.58).^103- 104 Supplemental oxygen therapy appeared to have a mild beneficial effect on the mean pulmonary arterial pressure and CRQ (dyspnea domain) scores.^105- 108

Disease management is an approach to coordinate resources across the health care system with the aim of fostering continuity of care and increasing patients' knowledge and control over their chronic disease.¹⁰⁹ Because the care of patients with COPD frequently requires multiple caregivers, including physicians (both generalists and specialists), nurses, physiotherapists, pharmacists, and nutritionists, a process to promote integration and seamless care may improve clinical outcomes in COPD. However, the efficacy of disease management programs in COPD remains uncertain (Table 4).^110- 117 On average, these programs appear to improve health status of patients but may not meaningfully impact hospitalization rates. However, because of marked heterogeneity in the content of the programs and their effects, these data need to be interpreted cautiously and further study is required.

The effects of oral theophyllines on exacerbation and mortality in COPD are uncertain and few well-conducted randomized trials are available powered on these end points.¹¹⁸ However, theophyllines appear to have some beneficial effects on FEV₁ as well as on the arterial contents of oxygen and carbon dioxide of patients with moderate to severe COPD.¹¹⁸ Oral theophyllines, however, increase the risk of nausea by approximately 7 fold. Three well-conducted clinical trials (1321 patients) demonstrated that lung volume reduction surgery (LVRS) improves health-related quality of life and exercise tolerance of patients with an FEV₁ less than 30% of predicted.^119- 121 However, even among carefully selected patients, LVRS did not modify all-cause mortality rates over 5 years.¹¹⁹ The short-term mortality was higher among patients who received LVRS than among those who were treated medically.^119,121 In those patients with FEV₁ less than 20% predicted, LVRS increased the risk of mortality by approximately 4 fold (beyond medical therapy).¹²² Accordingly, for most patients with COPD, LVRS cannot be recommended at this time. Lung transplantation should be reserved for patients with very advanced COPD (and without major comorbid conditions) and whose projected survival is less than 2 to 3 years.¹²³ Although lung transplantation may improve functional status and exercise tolerance of patients with COPD, no well-conducted studies have been performed to demonstrate survival benefits.¹²⁴

Chronic obstructive pulmonary disease is common and associated with immense health and economic burdens.⁹Airflow obstruction is a cardinal feature of COPD, and therapies that produce bronchodilation have been demonstrated to improve patients' health status and reduce exacerbations. Long-acting β₂-agonists and tiotropium both reduce exacerbation rates by approximately 25% in patients with moderate to severe COPD. Long-acting β₂-agonists induce bronchodilation by relaxing smooth muscle cells through activation of the adenylate cyclase pathways, which in turn increases intracellular concentrations of cyclic adenosine monophosphate.⁴² In addition, in vitro experiments have suggested that these compounds may have certain anti-inflammatory properties.¹²⁵ Anticholinergic bronchodilators, on the other hand, induce bronchodilation by attenuating vagal tone in the airways.¹²⁶ While both ipratropium and tiotropium are nonselective antagonists with similar binding affinity for the muscarinic receptors, tiotropium dissociates more than 100 times more slowly from the receptor complex than ipratropium, making the former more potent and longer acting than the latter.¹²⁶ In most situations, tiotropium may be used once daily, while ipratropium generally is given 4 to 6 times per day.

The current evidence suggests that these 2 classes of long-acting bronchodilators (long-acting β₂-agonists and anticholinergics) have similar efficacy, although only 2 trials (1830 patients) have directly compared them. Whether a combination of these 2 classes of long-acting bronchodilators would have an additive benefit on clinical outcomes (over monotherapy) is unknown. Until such data are published, this practice cannot be routinely recommended.

Inhaled corticosteroids also reduce exacerbation rates by approximately 25%. The mechanisms by which inhaled corticosteroids exert their beneficial effects on COPD outcomes are unclear. However, they attenuate airway hyperresponsiveness,^67,127 which may be an important predictor of COPD mortality,¹²⁸ and reduce some^129- 132 but not all^133- 134 components of airway inflammation and oxidative stress. Some caution should be exercised in using inhaled corticosteroids because they are associated with increased risk of certain adverse effects including thrush, oral candidiasis, and bruising. Inhaled corticosteroids also have some deleterious effect on bone mineral density, but the effect was very modest and was not associated with an excess risk of clinically evident fractures during these admittedly short-term trials. While the trials may have been too short to detect any changes, it is possible that inhaled corticosteroids may not increase fracture rates as, by reducing the frequency of COPD exacerbations by one quarter, they will spare patients from exposure to higher doses of systemic corticosteroids.

Theoretically, adding long-acting β₂-agonists to inhaled corticosteroid use may amplify the anti-inflammatory effects of corticosteroids¹³⁵ since long-acting β₂-agonists may enhance nuclear localization of glucocorticoid receptors in inflammatory cells,¹³⁶ making it easier for corticosteroids to effectively block cytokine expression in these cells. Long-acting β₂-agonists also may increase the effectiveness of corticosteroids in suppressing expression of adhesion molecules such as intracellular adhesion molecule 1.¹³⁷ Consistent with these observations, clinical studies suggest that by combining long-acting β₂-agonists and inhaled corticosteroids, exacerbation rates are lowered beyond that achieved by individual component therapy. However, the beneficial effects are additive (not synergistic) (RR, 0.70 vs placebo).

Supplemental oxygen therapy improves dyspnea scores, reduces pulmonary arterial pressure, and prolongs survival in those patients with COPD with a PaO₂ lower than 60 mm Hg. Pulmonary rehabilitation is effective in improving exercise tolerance and health status of patients with an FEV₁ less than 1.5 L. The relative improvements in patients' health status are maintained during the 18 months of follow-up in some studies, but its effects beyond that time frame are uncertain. No compelling evidence is available showing that pulmonary rehabilitation reduces frequency of hospitalizations or death in patients with COPD.

Although conceptually appealing, disease management programs have not yet been shown to improve clinical outcomes in patients with COPD (Table 4). However, this finding may reflect differences in the core components of disease management strategies across various studies. In general, studies that taught self-management skills and provided comprehensive educational services to patients had better outcomes than those that provided closer follow-up (Table 4). Well-designed comparative studies are needed to validate these initial observations.

In summary, consistent with the guidelines from the Global Initiative for Chronic Obstructive Lung Disease committee,⁹ smoking cessation is the cornerstone of chronic COPD management. Inhaled bronchodilator therapy on an as-needed basis should be considered for patients who experience occasional exacerbations. Regular bronchodilator therapy may be instituted for those with persistent symptoms. Combination therapy (with both a β₂-agonist and an anticholinergic) or the use of long-acting agents would appear to be the best approaches in this setting. In symptomatic individuals with moderate to severe disease, addition of inhaled corticosteroids with or without long-acting β₂-agonists and pulmonary rehabilitation should be considered. In patients with hypoxia at rest, supplemental oxygen therapy should be instituted. Although little evidence is available specifically examining the effects of influenza and pneumococcal vaccinations in the COPD population, they have been demonstrated to reduce hospitalizations and deaths in the general elderly population with only minor excess risks to recipients, and, thus, they are recommended for most patients with symptomatic COPD.