INTRODUCTION

TOP ABSTRACT INTRODUCTION AIMS OF TREATMENT IN... RATIONALE FOR USING STEROIDS BENEFICIAL EFFECTS OF ORAL... INHALED STEROIDS FUTURE PERSPECTIVES REFERENCES

Chronic obstructive pulmonary disease (COPD) constitutes a huge health problem. With an increasing mortality it is expected to be the fourth leading cause of death worldwide in 2020 (1). Despite the enormous burden of the disease both in health economic and personal perspective, there is a lack of adequate treatment for this disease. COPD can thus be regarded as the "Cinderella" respiratory condition in medical practice. Treatment is often unsatisfactory and treatment influencing the progressive nature of the disease is, apart from smoking cessation and oxygen in hypoxemic patients, not available so far. One of the most controversial questions in the management of COPD is whether inhaled glucocorticosteroids (steroids) are of benefit to patients with COPD. Although they are currently used extensively, their prescription is not evidence-based. This is largely due to the lack of data on long-term treatment and the fundamental lack of understanding about the underlying nature of the condition.

	AIMS OF TREATMENT IN COPD

TOP ABSTRACT INTRODUCTION AIMS OF TREATMENT IN... RATIONALE FOR USING STEROIDS BENEFICIAL EFFECTS OF ORAL... INHALED STEROIDS FUTURE PERSPECTIVES REFERENCES

When treating patients with COPD, it is necessary to assess first which end terms of treatment are to be obtained. A reversal of the emphysematous component of the disease, which is largely induced by smoking, cannot be expected. The disease is slowly progressive, with a long preclinical stage. In an advanced phase, there is increased bronchial collapsibility and loss of lung elasticity, which are probably irreversible. The damage to lung tissue and airways is generally already in an advanced state before symptoms occur, hence failure can be anticipated if the aim is to reverse the damage to lung tissue. In addition, the extent of the damage can hardly be measured without invasive procedures. Therefore, most studies examining the usefulness of certain drugs in COPD have aimed at slowing down the progression of the disease, i.e., either to improve FEV₁ or reduce the ongoing decline in FEV₁. Survival in COPD correlates directly with the level of FEV₁ (2), and any treatment that slows the accelerated decline in FEV₁ in COPD will likely also reduce mortality. A further aim can be the reduction of the number of exacerbations. Exacerbations have a large impact on the health status of patients with COPD and provide a large part of the costs of this disease, at least when requiring hospitalizations. On average, patients with COPD have one to four acute exacerbations a year, which translates to a total of 15 to 16 million episodes a year in the United States. Thus, a reduction in the number of exacerbations may have a large impact on the economic burden (3). Finally, improvement of symptoms and quality of life and limitation of the impact of disease on daily life are important aims for improvement (4). However, as long as the treatment of COPD is only palliative in nature, these aims may well be the most important in treatment.

	RATIONALE FOR USING STEROIDS

TOP ABSTRACT INTRODUCTION AIMS OF TREATMENT IN... RATIONALE FOR USING STEROIDS BENEFICIAL EFFECTS OF ORAL... INHALED STEROIDS FUTURE PERSPECTIVES REFERENCES

Inflammation Underlying the Pathophysiology of COPD

Pathophysiological studies of patients with COPD show that there is a chronic inflammatory process in the airway wall and lumen of the large and peripheral airways and a loss of elastic tissue in the supporting alveolar structure of the outer wall of the small airways. An increased smooth muscle thickness in peripheral airways of patients with COPD has also been observed, contributing to the airway narrowing, and possibly to hyperresponsiveness in COPD (5). The inflammatory process in the small airways, being important both during the initial stage in the development of COPD (6, 7) and in established COPD, consists of an increased number of lymphocytes, mononuclear cells, and neutrophils, increased connective tissue deposition and epithelial metaplasia, and ulceration in the airway walls (5, 8, 9). Some studies have also found increased numbers of mast cells in patients with COPD (10). Investigations involving smokers with or without airflow limitation show that there is a larger percentage of neutrophils in patients with airflow limitation (11). A causal relation between the numbers of neutrophils and lung function decline has been suggested by Stanescu and coworkers (12). They found that a higher number of neutrophils in sputum, as measured at the end of the observation period, was associated with a more rapid decline in lung function in the previous 15 yr. However, whether the higher number of neutrophils was the cause of the lower lung function or the consequence of more severe disease remains to be established. Furthermore, it is necessary to realize that neutrophils have been found to be increased in bronchoalveolar lavage fluid and sputum of patients with COPD, yet they are not abundantly present in the airway wall (8, 9). Macrophages may be implicated in the development and progression of COPD as well, since macrophages of smoking patients with COPD show a higher activation level. Cigarette smoke may therefore be regarded as an inducer of inflammation.

The inflammatory processes in the airways may serve as a reason for instituting antiinflammatory therapy in patients with COPD. The pathogenic role of inflammation is circumstantially supported by the observation that quitting cigarette smoking, which induces inflammation, results in a significant slowing of the deterioration in lung function.

Response of Inflammation to Steroid Treatment

There has been a debate that because the inflammation in the airways of patients with COPD is neutrophilic, the disease would be less amenable to steroid therapy. Systemic steroids are known to increase peripheral neutrophil counts, which may reflect an increased survival time due to an inhibitory action on neutrophil apoptosis (13). It has not been investigated, however, whether this is also the case for lung neutrophils. Thompson and coworkers (14), in accordance with this assumption, did not find a reduction in the numbers of neutrophils in bronchoalveolar lavage after treatment with inhaled steroids, although the total cell count was reduced. However, the observation that increased numbers of neutrophils are present during acute exacerbations of COPD and acute exacerbations do respond to oral steroids might suggest that neutrophilic inflammation does not per se reflect unresponsiveness to (oral or inhaled) steroids. Indeed, two articles present evidence of some beneficial effect of inhaled steroids in high doses. Llewellyn-Jones and coworkers (15) observed a reduction in the chemotactic activity of sputum after 8 wk of treatment with 1,500 µg of fluticasone in 17 patients with COPD. Moreover, they found a beneficial effect on the proteinase/antiproteinase balance. Notwithstanding the reduction in neutrophil chemotactic activity, the neutrophils themselves were not likely affected since myeloperoxidase content in sputum was comparable before and after 8 wk of treatment. The authors did not investigate whether the number of neutrophils was reduced by fluticasone treatment as well. A study by Confalonieri and coworkers (16) suggested that this would be the case since a daily dose of 500 µg of inhaled beclomethasone for 2 mo induced a reduction in the number of neutrophils in sputum. The study by Keatings and coworkers (17) did not find such a reduction. However, the time span of investigation was only 2 wk, and a longer duration of treatment could have had a better effect. These short-term studies suggest that there might be a beneficial antiinflammatory effect of inhaled steroids in COPD. However, whether these short-term antiinflammatory effects are associated with long-term beneficial effects on lung function remains to be investigated.

Some studies have suggested that eosinophilic inflammation in patients with COPD identifies individuals who benefit from steroid treatment. Chanez and coworkers (18) assessed the response to prednisone (1.5 mg/kg for 15 d), a positive response being defined as an increase in FEV₁ of at least 12% from baseline values and an absolute increase of 200 ml measured at the end of treatment. Twelve of 25 patients responded to treatment. In comparison with nonresponders, responders had a significantly larger number of eosinophils and higher levels of eosinophil cationic protein (ECP) in bronchoalveolar lavage fluid, while the number of EG2-positive cells in airway wall biopsies tended to be higher. The responders had a thicker reticular basement membrane than did the nonresponders. This suggests that the inflammation in this subgroup of responders resembles more or less that of asthma (18). Pizzichini and coworkers (19) investigated smokers with severe airflow limitation with respect to treatment response to 2 wk of 30 mg of prednisone daily. Eight of 18 patients had sputum eosinophilia ( 3% of the sputum cell count). Only in these patients did prednisone improve effort dyspnea, quality of life, and FEV₁. This is in contrast to the observations of Keatings and co-workers (17), who investigated a small number of patients and confirmed the presence of increased ECP levels in induced sputum in some patients with COPD, yet the levels did not diminish on a 2-wk course of oral or inhaled steroids. Thus, the relevance of increased ECP levels is unclear and the latter study did not show an FEV₁ response to steroids, despite the increased levels of ECP. Future studies must confirm whether only COPD patients with eosinophilic inflammation respond to long-term inhaled steroids and possibly represent a subgroup of individuals who have an asthmatic component.

	BENEFICIAL EFFECTS OF ORAL STEROIDS

TOP ABSTRACT INTRODUCTION AIMS OF TREATMENT IN... RATIONALE FOR USING STEROIDS BENEFICIAL EFFECTS OF ORAL... INHALED STEROIDS FUTURE PERSPECTIVES REFERENCES

Oral steroids for treatment of stable COPD have been investigated in several studies, focusing on improvement of the level of airway obstruction (20, 21). Results from studies with a controlled double-blind design are conflicting, but in a meta-analysis it was demonstrated that they do have a small effect: Callahan and coworkers (21) calculated that approximately 10% of patients with stable COPD have a 20% or greater improvement in baseline FEV₁ after a 2-wk treatment with oral steroids. With longer duration of oral steroid therapy, this number may be higher. Thus, Weir and coworkers (22) found that FEV₁ continued to improve after 14 d of steroid treatment in some subjects with COPD. Oral steroids do not change airway hyperresponsiveness (23, 24) and bronchodilator response to cumulatively applied doses of a -agonist or anticholinergic, nor do they alter the protection provided by either drug against histamine (25). A response to oral prednisolone occurs as frequently in patients with physiologic features of emphysema as in those without such features (26).

Several studies have shown that oral steroids in patients with an acute exacerbation are effective in reducing symptoms and improving lung function (27, 28). A double-blind study of 27 patients with COPD with an exacerbation showed that a 10-d tapering dose-course of oral steroids improved arterial PO₂, alveolar-arterial oxygen difference, and lung function to a larger extent than placebo (28). Moreover, it reduced the severity of symptoms and the number of treatment failures as well. This suggests that patients with COPD are "steroid responders" during acute exacerbations, but not necessarily during the stable state of the disease.

Only two long-term, yet retrospective, studies of oral steroids in COPD are available (29, 30). In those studies, we showed a favorable effect of prednisolone on FEV₁ over a 20-yr period. At a dose of 10 mg/d or more, FEV₁ remained stable or even increased. In the first study (29), allergic patients were not systematically excluded, and this might have influenced the results. Therefore, a similar study was conducted in patients without any sign of allergy (30). Comparable results were obtained. A clinically important finding was that a change in the decline of FEV₁ could be observed only after at least 6 to 24 mo of therapy. This is in striking contrast with asthma, where steroids have an almost instantaneous effect.

	INHALED STEROIDS

TOP ABSTRACT INTRODUCTION AIMS OF TREATMENT IN... RATIONALE FOR USING STEROIDS BENEFICIAL EFFECTS OF ORAL... INHALED STEROIDS FUTURE PERSPECTIVES REFERENCES

Short-term Studies

There are now approximately 10 randomized, placebo-controlled short-term studies (14, 25, 31) showing that inhaled steroids administered over a period of 3-12 wk generally do not change the level of airways obstruction, as assessed by FEV₁ and peak expiratory flow (PEF), or airway hyperresponsiveness to histamine in patients with COPD (Table 1). Nevertheless, several studies showed that some individuals had a substantial improvement in lung function, although it is not yet clear which characteristics predict the responsive individual. For instance, the response to a short course of oral steroids was not predictive (37). We assessed the effect of 6 wk of treatment with 1,600 µg of budesonide on hyperresponsiveness to adenosine 5'-monophosphate (AMP) in 44 patients with COPD (34). A bronchoconstrictor effect after inhalation of AMP is thought to occur largely owing to mast cell activation. Since mast cell numbers are increased in the airway wall of patients with COPD (10), we argued that beneficial effects of inhaled steroids might have been missed by using only histamine as the bronchoconstrictor agent. However, both the PC₂₀ methacholine (the provocative concentration of methacholine causing a 20% fall in FEV₁) and the PC₂₀ AMP did not change significantly with inhaled steroids. An interesting observation was that levels of serum interleukin 8 (IL-8), a cytokine that predominantly attracts neutrophils to sites of inflammation, were significantly reduced during this treatment. This might thus suggest that neutrophil attraction is reduced with inhaled steroids via an IL-8 pathway, as was already shown in a previous study (16).

Long-term Studies

Effects on lung function. There have now been six long-term randomized, placebo-controlled studies published as full articles or abstracts (39) and three reported at international meetings (45) assessing the long-term effect of inhaled steroids in COPD (Table 2). Two studies with a 6-mo follow-up have obtained conflicting results. The first did not show an effect of 800 µg of inhaled budesonide in 40 patients who did not respond to a 2-wk treatment with oral prednisone, when effects were compared with those 40 patients with COPD who received placebo (44). Assessments of other outcome parameters including daily symptoms, exercise capacity, and quality of life, led to the same conclusion. There were many dropouts, making the power of the study too small to provide sufficient evidence that there was no effect. Numbers of withdrawals and reasons for withdrawal were comparable in both treatment arms. However, these data are compatible with the Copenhagen study, in which only those individuals were included who did not respond to oral corticosteroids (to exclude patients with asthma as far as possible) (47). In that study also no significant effect on lung function (FEV₁) was observed within the first 6 mo of treatment with 1,200 µg of budesonide.

The second 6-mo study (43) compared inhaled fluticasone (500 µg twice daily) with placebo in current or ex-smokers with COPD (139 patients in the placebo group and 142 in the fluticasone group). The authors concluded that fluticasone might be of clinical benefit in patients with COPD, based on PEF, FEV₁, and FVC. The authors tried to find an explanation why some individuals responded well and others less or not at all to inhaled fluticasone. Patient age, sex, baseline FEV₁ and bronchodilator reversibility, smoking habit, and serum cortisol level did not predict the responder. An interesting observation in a subanalysis was that individuals who had COPD of more than 10 yr duration had a better response than those with a recent presentation of disease. The investigators could not explain this by baseline FEV₁, which was comparable between the groups. Thus, either these patients perceived their airway obstruction earlier, or they had, as the authors suggested, a component of chronic asthma.

In a Dutch multicenter trial we studied a mixed group of patients with airway obstruction (39), treated with 800 µg of inhaled beclomethasone for 3 yr. In a subgroup with COPD (defined as chronic cough and/or sputum production, no asthma attacks, all current or ex-smokers, an FEV₁ < 95% of predicted, and no response to bronchodilator greater than 15%), a small, nonsignificant improvement in FEV₁ of 4.4%, and a nonsignificant improvement in PC₂₀, was seen. Another study looked at the effect of 1 yr of inhaled steroids in patients with COPD (48). Patients in that study were selected from a former study on the basis of their rapid decline in lung function over the past 2 yr when they did not use inhaled steroids. The patients served as their own controls when 800 µg of inhaled beclomethasone was instituted after the first 2 yr. There was a large mean fall in FEV₁ of about 160 ml/yr in the first 2 yr. After institution of beclomethasone, patients improved their prebronchodilator FEV₁ significantly, by 160 ml, in the first 6 mo of treatment, but the fall in FEV₁ continued in the next 6 mo by a mean of 70 ml (Figure 1), which was not significantly different from the decline in FEV₁ during the first 2 yr. This study thus suggests that in the study by Paggiaro and coworkers (43) the eventual deterioration in FEV₁ may have been missed owing to a relatively short observation time. This initial improvement in FEV₁ followed by the "normal" course of disease, i.e., a slow but ongoing deterioration, has now been shown in other studies as well. It is comparable to the findings of postbronchodilator FEV₁ in the EUROSCOP study of smoking individuals with early disease (45) and the ISOLDE study of patients with advanced diseases (46). These results have not been published so far.

View larger version (19K): [in this window] [in a new window]   Figure 1.   Course of pre- and postbronchodilator FEV1 during 4 yr of follow-up in COPD; intervention with 800 µg of inhaled beclomethasone dipropionate per day (48).

In contrast, a study of a small group of patients with COPD (without allergy) with a 2-yr double-blind follow-up did not show this initial increase in FEV₁ (40). One individual in the inhaled steroid group was responsive to an oral steroid course; otherwise all 38 patients in the inhaled steroid and placebo groups were nonresponsive. That study did not find a different slope in FEV₁ between those treated with 1,600 µg of budesonide daily (median value, 30 ml/yr) and placebo (median value, 60 ml/yr). Airway hyperresponsiveness also failed to change with steroid treatment. The large intersubject variability in FEV₁ slopes may have obscured potential effects on FEV₁. Beneficial effects were limited to symptoms (Figure 2) and withdrawal rate due to pulmonary problems.

View larger version (40K): [in this window] [in a new window]   Figure 2.   Mean (± SEM) symptom scores per treatment group before treatment and in the first and second year of treatment (40). Bud = 1,600 µg budesonide per day; bud + pred = 1,600 µg of budesonide per day; and 5 mg of prednisolone. Plac = placebo.

Effects on exacerbations. There are some indications that exacerbations may be reduced in frequency by inhaled steroids. The number of exacerbations per patient was lower in the fluticasone group in the study by Paggiaro and coworkers (43), but not significantly so (p = 0.067). Nevertheless, significantly fewer patients in the fluticasone group had severe and moderately severe exacerbations compared with placebo treatment. In the study by Renkema and coworkers (40), five patients dropped out of the placebo group owing to pulmonary problems, whereas none dropped out of the inhaled steroid group. Moreover, symptoms improved significantly more in the group treated with inhaled steroids. The number of exacerbations was the same in the years without and with inhaled steroids in the Dompeling study (a mean of 1.8 exacerbations per year in both periods [48]) and in the placebo and inhaled steroid arms of the Renkema study (40), i.e., 1 and 2, respectively, 2 and 2.5 exacerbations per year in the first and second year of follow-up. Finally, the number of exacerbations was reported not to decrease in the EUROSCOP study (45), but the participants had few exacerbations to begin with given the mild nature of disease. In contrast, the ISOLDE study (46) found a significant reduction in the number of exacerbations with steroid treatment in patients with severe COPD.

Effects on health status and symptoms. Paggiaro and co-workers (43), in their study with a follow-up of 6 mo found an improvement in symptom scores and walking distance in addition to improvement in lung function. Renkema and coworkers (40) observed a beneficial effect on symptoms (Figure 2) as well. Finally, the ISOLDE study (46) showed that health status improved continuously with inhaled steroids, despite the fact that lung function did not improve further after the first 6 mo and even deteriorated again. This requires further study as to the underlying mechanism of the better improvement in health status with inhaled steroids than with placebo. Other long-term studies have not addressed health status as an outcome parameter.

Dosing considerations. The dose of the inhaled steroid might have an effect on the outcome of the studies as well. This remains a matter that is open to debate, since the studies were performed with different types of drugs, i.e., beclomethasone, budesonide, and fluticasone propionate. With a daily dose of 800 µg of beclomethasone or budesonide there occurred a positive initial response in the study by Dompeling and coworkers (48) and in the EUROSCOP study (45), yet no effect in the study by Bourbeau and coworkers (44). Moreover, the Copenhagen study, where a higher dose of corticosteroids was used, i.e., 1,200 µg of budesonide during the first 6 mo, did not find a positive effect either (47). In the study by Paggiaro and coworkers (43) and in the ISOLDE study (46), both showing an effect of inhaled steroids, 1,000 µg of fluticasone was used.

Results might be different owing to a class effect, but positive effects were found with the inhalation of fluticasone (43, 46), with budesonide (45), and with beclomethasone (48). The severity of airflow limitation cannot explain the results either, since positive and negative responses have been shown in patients with advanced airflow limitation, as well as in patients with mild airflow limitation. A last consideration may be smoking habits, since the EUROSCOP study suggested in a subanalysis that a lower number of pack-years might be associated with a better response to the treatment. The data are not yet available to allow interpretation of the other large and long-term studies.

	FUTURE PERSPECTIVES

TOP ABSTRACT INTRODUCTION AIMS OF TREATMENT IN... RATIONALE FOR USING STEROIDS BENEFICIAL EFFECTS OF ORAL... INHALED STEROIDS FUTURE PERSPECTIVES REFERENCES

It would be extremely attractive to assess which factors determine the response to treatment with inhaled steroids in patients with COPD. This will need a large cohort with variability in the parameters under study. Since these data will not be available for long if data are not grouped, a formal meta-analysis would be of great clinical importance. Finally, further investigation is required to determine why steroid treatment seems to be effective only during the first 6 mo. The effects of inhaled steroids on airway wall and tissue remodeling have thus far not been investigated in COPD. This is of prime interest, since this might explain the observed effects in the long-term studies.