Correspondence

Prevention of Death in COPD

N Engl J Med 2007; 356:2211-2214May 24, 2007

Article

To the Editor:

Calverley et al. do not sufficiently emphasize some aspects of their study on the use of salmeterol and fluticasone in patients with chronic obstructive pulmonary disease (COPD) (Feb. 22 issue).1 Their study, called the Towards a Revolution in COPD Health (TORCH) trial, showed that treatment with fluticasone alone actually increased mortality at the end of 3 years, although the increase was not significant. This finding contrasts markedly with retrospective analyses and meta-analyses showing a substantial reduction in mortality from all causes by about 25% associated with the drug.2,3 This discrepancy between the results of a well-conducted, randomized, controlled trial and historical analyses highlights how misleading the latter may be. The net effect of therapy with inhaled corticosteroids for patients who have COPD may be detrimental in view of the increased episodes of pneumonia associated with such agents.1,4 Another important result of the TORCH study was the failure of inhaled corticosteroids, even when combined with salmeterol, to reduce the annual decline in lung function. The lack of effect of inhaled corticosteroids on mortality and disease progression may reflect resistance to the antiinflammatory effects of corticosteroids in COPD.5

Peter J. Barnes, D.M., D.Sc.
National Heart and Lung Institute, London SW3 6LY, United Kingdom
p.j.barnes@imperial.ac.uk

Dr. Barnes reports being a member of advisory boards at and receiving research funding from AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, Novartis, and Pfizer. No other potential conflict of interest relevant to this letter was reported.

5 References

1. 1

   Calverley PM, Anderson JA, Celli B, et al. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. N Engl J Med 2007;356:775-789
   Full Text | Web of Science | Medline

2. 2

   Sin DD, Wu L, Anderson JA, et al. Inhaled corticosteroids and mortality in chronic obstructive pulmonary disease. Thorax 2005;60:992-997
   CrossRef | Web of Science | Medline

3. 3

   Macie C, Wooldrage K, Manfreda J, Anthonisen NR. Inhaled corticosteroids and mortality in COPD. Chest 2006;130:640-646
   CrossRef | Web of Science | Medline

4. 4

   Kardos P, Wencker M, Glaab T, Vogelmeier C. Impact of salmeterol/fluticasone propionate versus salmeterol on exacerbations in severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2007;175:144-149
   CrossRef | Web of Science | Medline

5. 5

   Barnes PJ. Reduced histone deacetylase in COPD: clinical implications. Chest 2006;129:151-155
   CrossRef | Web of Science | Medline

To the Editor:

The TORCH trial is described as a parallel-group study primarily comparing the combination of salmeterol and fluticasone with placebo but also comparing the combination with each component alone. The same data can be analyzed according to a factorial design1 to estimate the main effect of each drug with adjustment for the other (Table 1Table 1Analysis of Main Effects of Treatment with Salmeterol and Fluticasone on Mortality.).

The factorial analysis indicates that the effect of the combination therapy on mortality was entirely due to salmeterol, and this effect (a reduction in mortality of 19%) was highly significant (P=0.004) on the basis of data from 3054 patients who received the drug and 3058 who did not. By contrast, the fluticasone component had no effect on mortality among 3067 patients who received the drug and 3045 who did not. The results of the factorial analysis support the conclusions of the editorial by Rabe2 accompanying the TORCH report.

Considering that the combination of salmeterol and fluticasone is superior to either drug alone in reducing exacerbations of COPD and improving health status and lung function, the reduction of mortality associated with salmeterol alone should be balanced against the more favorable pattern of symptomatic effects of the combination of the two drugs, with allowances made for the increased frequency of pneumonia.

Carlo La Vecchia, M.D.
Istituto di Ricerche Farmacologiche Mario Negri, 20157 Milan, Italy
lavecchia@marionegri.it

Leonardo M. Fabbri, M.D.
University of Modena e Reggio Emilia, 41100 Modena, Italy

Dr. La Vecchia has been asked by GlaxoSmithKline to review and comment on the TORCH study, and financial aspects are under discussion. Dr. Fabbri reports receiving consulting and lecture fees from Altana Pharma, AstraZeneca, Boehringer Ingelheim, Chiesi Farmaceutici, GlaxoSmithKline, Merck, Novartis, Roche, and Pfizer and grant support from Altana Pharma, AstraZeneca, Boehringer Ingelheim, Menarini, Miat, Schering–Plough, Chiesi Farmaceutici, GlaxoSmithKline, Merck, UCB, and Pfizer. No other potential conflict of interest relevant to this letter was reported.

2 References

1. 1

   Peto R. Clinical trial methodology. Biomedicine 1978;28 Spec No.:24-36. Peto R. Clinical trial methodology. Biomedicine 1978;28 Spec No.:24-36.;
   

2. 2

   Rabe KF. Treating COPD -- the TORCH trial, P values, and the dodo. N Engl J Med 2007;356:851-854
   Full Text | Web of Science | Medline

To the Editor:

The TORCH study investigators report that inhaled fluticasone at high doses may reduce exacerbation rates in COPD but that the drug may cause more harm than good overall. The finding that pneumonia occurred more frequently in the two groups receiving fluticasone (in an average of 19% of the patients) than in the placebo group or the salmeterol-only group (in an average of 13%) is worrisome. The number needed to harm is 17 (P<0.001). This could be explained by immunosuppression due to systemic activity.

How do Calverley et al. explain how they derived the numbers needed to treat? They state that “the number needed to treat to prevent an exacerbation in 1 year was 4, and the number needed to treat to prevent a hospitalization was 32.” This statement relates to placebo as compared with combination therapy. This finding is not immediately apparent from the data provided. For example, the number needed to treat for exacerbations seems counterintuitive, given a reduction in the annual rate of exacerbation from 1.13 to 0.85, an absolute difference of 0.28. Likewise, the annual hospitalization rate was reduced from 0.19 to 0.16, a difference of 0.03.

Martin Duerden, M.B., B.S., M.R.C.G.
Meddygfa Gyffin, Conwy, North Wales LL32 8LT, United Kingdom

To the Editor:

By assigning a group of symptomatic patients with COPD to a placebo group without the use of long-acting bronchodilators, Calverley et al. violated paragraph 29 of the Declaration of Helsinki, which states that “the benefits, risks, burdens and effectiveness of a new method should be tested against those of the best current prophylactic, diagnostic, and therapeutic methods.” These patients had some degree of airway reversibility. Comparing the post-bronchodilator response with the percent of the predicted value for forced expiratory volume in 1 second (FEV₁) rather than with the prebronchodilator FEV₁ minimized the amount of reversibility. It is no wonder that so many symptomatic patients dropped out of the placebo group.

Yizhak Kupfer, M.D.
Sidney Tessler, M.D.
Maimonides Medical Center, Brooklyn, NY 11219
stessler@maimonidesmed.org

To the Editor:

In the editorial accompanying the report on the TORCH study, Rabe refers to a 25% prevalence of venous thromboembolism in patients hospitalized with a severe exacerbation of COPD, citing a study by Tillie-Leblond et al.1 An instinctive response of defensive medical practice may now be to order a computed tomographic scan with a pulmonary-embolism protocol for every patient with a severe COPD exacerbation on the assumption that a 25% return must be cost-effective.

Tillie-Leblond et al. actually report that “25% of COPD patients with severe unexplained breathlessness have been shown to have pulmonary embolism.” The subjects were selected because of “the absence of a lower respiratory tract infection” — in other words, an exacerbation of COPD “of unknown origin.” Experience tells us that the prevalence of venous thromboembolism with exacerbations of COPD is low — only 3.3% in a recent study of 123 consecutive patients.2 The discussion in that report echoes that of Robin and McCauley,3 who bemoaned the ingrained belief that a low partial pressure of arterial oxygen is of positive predictive value for the diagnosis of venous thromboembolism.

Niall Keaney, M.B., Ph.D.
Sunderland Royal Hospital, Sunderland SR4 7TP, United Kingdom
niall.keaney@chs.northy.nhs.uk

3 References

1. 1

   Tillie-Leblond I, Marquette C-H, Perez T, et al. Pulmonary embolism in patients with unexplained exacerbation of chronic obstructive pulmonary disease: prevalence and risk factors. Ann Intern Med 2006;144:390-396
   Web of Science | Medline

2. 2

   Rutschmann OT, Cornuz J, Poletti P-A, et al. Should pulmonary embolism be suspected in exacerbation of chronic obstructive pulmonary disease? Thorax 2007;62:121-125
   CrossRef | Web of Science | Medline

3. 3

   Robin ED, McCauley RF. The diagnosis of pulmonary embolism: when will we ever learn? Chest 1995;107:3-4
   CrossRef | Web of Science | Medline

Author/Editor Response

Barnes is concerned that inhaled corticosteroids increase mortality in COPD. Although in our study, numerically more patients died in the group that received fluticasone alone than in the placebo group (hazard ratio, 1.06), the difference was not significant (P=0.53). Space limitations precluded a discussion of differences between our data and those from database analyses, but we agree that a large, randomized, controlled trial such as ours provides a more robust test of the effect of inhaled corticosteroids on survival than do such analyses. The rate of decline in lung function cannot be inferred from the spirometric data in our study.

La Vecchia and Fabbri have undertaken an interesting post hoc analysis suggesting that the salmeterol component has a substantial effect on mortality. Factorial analysis assumes that each treatment has the same additive effect in the absence and presence of the other treatment. This was not the case for the TORCH trial. Our data show the clear clinical superiority of combination treatment with salmeterol and fluticasone, including fewer exacerbations and better health status.

Kupfer and Tessler suggest that our study was unethical. During the trial design, there was concern about the safety and efficacy of the component treatments, questions that our data have resolved. Patients in the placebo group received regular short-acting bronchodilators, but our results show that this treatment will not be an appropriate standard of care in the future. The degree of bronchodilator reversibility in our patients was similar to that in other large COPD trials, which failed to show any relationship between clinical outcomes and reversibility.1-3

Duerden suggests that treatment with inhaled corticosteroids may do more harm than good. More patients with pneumonia were reported in groups that received inhaled corticosteroids, although there was no disproportionate mortality from pneumonia among patients receiving inhaled corticosteroid monotherapy, nor did the overall rate of hospitalization for COPD differ from that in the placebo group. We agree that more data are needed to better understand this finding. Large data sets will be needed, since pneumonia was relatively infrequent, as compared with other serious outcomes (about 1000 cases of pneumonia in 780 patients vs. 13,000 exacerbations of COPD in 4000 patients). We calculated the number needed to treat as the number of patients required to prevent one exacerbation, according to published methods.4 This number is not the same as the number needed to treat to prevent one patient from having an exacerbation, a number more appropriate for a binary event such as mortality. Although attractive in clinical practice, the number needed to treat should be viewed with caution, since it depends on the background event rate in the population under study.

Peter Calverley, M.D.
University Hospital Aintree, Liverpool L9 7AL, United Kingdom
pmacal@liverpool.ac.uk

Julie Anderson, M.A.
GlaxoSmithKline, Greenford UB6 0HE, United Kingdom

Bartolome Celli, M.D.
Caritas St. Elizabeth's Medical Center, Boston, MA 02135-2997

4 References

1. 1

   Calverley PMA, Burge PS, Spencer S, Anderson JA, Jones PW. Bronchodilator reversibility testing in chronic obstructive pulmonary disease. Thorax 2003;58:659-664
   CrossRef | Web of Science | Medline

2. 2

   Calverley PMA, Pauwels R, Vestbo J, et al. Combined salmeterol and fluticasone in the treatment of chronic obstructive pulmonary disease: a randomised controlled trial. Lancet 2003;361:449-456[Erratum, Lancet 2003;361:1660.]
   CrossRef | Web of Science | Medline

3. 3

   Tashkin D, Kesten S. Long-term treatment benefits with tiotropium in COPD patients with and without short-term bronchodilator responses. Chest 2003;123:1441-1449
   CrossRef | Web of Science | Medline

4. 4

   Halpin DMG. Evaluating the effectiveness of combination therapy to prevent COPD exacerbations: the value of NNT analysis. Int J Clin Pract 2005;59:1187-1194
   CrossRef | Web of Science | Medline

Citing Articles (5)

Citing Articles

1. 1

   Fabiano Di Marco, Marco Guazzi, Giuseppe Francesco Sferrazza Papa, Marco Vicenzi, Pierachille Santus, Paolo Busatto, Federico Piffer, Francesco Blasi, Stefano Centanni. (2012) Salmeterol improves fluid clearance from alveolar-capillary membrane in COPD patients: A pilot study. Pulmonary Pharmacology & Therapeutics
   CrossRef

2. 2

   Donald P Tashkin. (2010) Impact of tiotropium on the course of moderate-to-very severe chronic obstructive pulmonary disease: the UPLIFT ^® trial. Expert Review of Respiratory Medicine 4:3, 279-289
   CrossRef

3. 3

   Juan José Soler-Cataluña, Miguel Ángel Martínez García. (2009) Impacto de los estudios de eficacia y mortalidad (TORCH y UPLIFT) en el tratamiento broncodilatador de la EPOC. Archivos de Bronconeumología 45, 14-20
   CrossRef

4. 4

   Pierre Ernst, Samy Suissa. (2008) Pneumonia in elderly patients with chronic obstructive pulmonary disease. Current Infectious Disease Reports 10:3, 223-228
   CrossRef

5. 5

   S. Nitschmann, G. Schultze-Werninghaus. (2007) Auswirkungen der COPD-Therapie auf das Patientenüberleben. Der Internist 48:12, 1450-1452
   CrossRef