Deception in Bronchodilator Inhaler Use

Deception in Bronchodilator Inhaler Use

patient had atypical angina. The remaining two patients did not have chest pain. After cardiac catheterization, including provocative tests for corona...

51KB Sizes 3 Downloads 29 Views

patient had atypical angina. The remaining two patients did not have chest pain. After cardiac catheterization, including provocative tests for coronary vasospasm using acetylcholine, all but one patient without chest pain were treated with calcium antagonists (diltiazem, four patients; amlodipine, three patients; verapamil, two patients; and nisoldipine, one patient), although treatment was withheld in one of these patients (receiving nisoldipine) because of hypotension. In addition, two patients with typical angina were given the potassium channel opener nicorandil and long-acting nitrates, because their anginal attacks did not subside with calcium antagonists alone. A third patient was treated with both a calcium channel blocker and a long-acting nitrate for the same reason. The one patient who had no angina and did not receive calcium antagonists took the class Ia antiarrhythmic drug cibenzoline because of paroxysmal atrial fibrillation. Ultimately, this patient was given the calcium antagonist (nilvadipine) because of hypertension. Cibenzoline was given to another patient in order to prevent paroxysmal atrial fibrillation and to a third patient in whom nonobstructive HCM converted to obstructive HCM during follow-up.11 The latter patient received a ␤-adrenergic blocking agent (metoprolol) in addition to a calcium antagonist (amlodipine) and cibenzoline. ␤-Blockers were given to three other patients (bisoprolol, two patients; propranolol, one patient) who had palpitations or dyspnea on effort, although the ␤-blocker (propranolol) was discontinued in one patient because of recurrent anginal attacks. After treatment, the occurrence of chest pain decreased markedly in 10 patients with typical or atypical angina, and did not appear in 2 patients who had no chest pain. During the follow-up period (mean, 61 ⫾ 24 months), two patients died. One patient who did not have angina but had several episodes of paroxysmal atrial fibrillation died suddenly; and the remaining patient, who had typical angina, died of colon cancer. An autopsy was not performed for either patient. Koji Kodama-Takahashi, MD Yawatahama General Hospital Ehime, Japan Yuji Shigematsu, MD Mareomi Hamada, MD Kunio Hiwada, MD Ehime University School of Medicine Ehime, Japan Yukio Kazatani, MD Ehime Prefectural Central Hospital Ehime, Japan Keisuke Matsuzaki, MD Eiki Murakami, MD Kinki Central Hospital Hyogo, Japan Correspondence to: Koji Kodama-Takahashi, MD, The Department of Internal Medicine, Yawatahama General Hospital, 1– 638 Ohira, Yawatahama-shi, Ehime 796-8502, Japan; e-mail: [email protected]

References 1 Kodama K, Shigematsu Y, Hamada M, et al. The effect of coronary vasospasm on the direction of ST-segment deviation in patients with both hypertrophic cardiomyopathy and vasospastic angina. Chest 2000; 117:1300 –1308 2 Hodgson JM, Marshall JJ. Direct vasoconstriction and endothelium-dependent vasodilatation: mechanisms of acetylcholine effects on coronary flow and arterial diameter in patients with nonstenotic coronary arteries. Circulation 1989; 79: 1043–1051

3 Vita JA, Treasure CB, Nabel EG, et al. Coronary vasomotor response to acetylcholine relates to risk factors for coronary artery disease. Circulation 1990; 81:491– 497 4 Zeiher AM, Drexler H, Saurbier B, et al. Endotheliummediated coronary blood flow modulation in humans: effects of age, atherosclerosis, hypercholesterolemia, and hypertension. J Clin Invest 1993; 92:652– 662 5 El-Tamimi H, Mansour M, Wargovich TJ, et al. Constrictor and dilator responses to intracoronary acetylcholine in adjacent segments of the same coronary artery in patients with coronary artery disease: endothelial function revisited. Circulation 1994; 89:45–51 6 Maseri A, Davies G, Hackett D, et al. Coronary artery spasm and vasoconstriction: the case for distinction. Circulation 1990; 81:1983–1991 7 The TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI) trial: phase 1 findings. N Engl J Med 1985; 312:932–936 8 Yasue H, Matsuyama K, Matsuyama K, et al. Responses of angiographically normal human coronary arteries to intracoronary injection of acetylcholine by age and segment: possible role of early coronary atherosclerosis. Circulation 1990; 81:482– 490 9 Kodama K, Hamada M, Kazatani Y, et al. Clinical characteristics in Japanese patients with coexistent hypertrophic cardiomyopathy and coronary vasospasm. Angiology 1998; 49: 849 – 855 10 Dimitrow PP, Krzanowski M, Nizankowski R, et al. Verapamil improves the response of coronary vasomotion to cold pressor test in asymptomatic and mildly symptomatic patients with hypertrophic cardiomyopathy. Cardiovasc Drugs Ther 1999; 13:259 –264 11 Hamada M, Shigematsu Y, Ikeda S, et al. Class Ia antiarrhythmic drug cibenzoline: a new approach to the medical treatment of hypertrophic obstructive cardiomyopathy. Circulation 1997; 96:1520 –1524

Deception in Bronchodilator Inhaler Use To the Editor: Although there are some limitations in the study design, the article by Simmons et al1 (August 2000) is very provocative, and the results strongly suggest that noncompliance can be an issue even within a monitored clinical trial. The authors attempted to identify demographic features that would distinguish the medication “dumpers” from medication “nondumpers.” The dumpers, though, did not appear differ significantly from the nondumpers in characteristics such as age, gender, race, body habitus type, education level, smoking history, features of asthmatic disease, or study visit follow-up rate. Areas the authors did not explore, however, were the relationship between the investigators and the patients, and what the patients’ expectations of the study were. Only two sites and two investigators participated, suggesting that effects intrinsic to the research center and the investigators were not a factor. It is not known, however, whether the patients were being treated by the investigator outside of the trial and whether the patients and investigators had a long-standing clinical relationship. Did dumpers and nondumpers differ by these variables? It may be useful to identify patient incentives for entering the study. It is not known whether dumpers more likely to have inadequate insurance and needed the trial to obtain medical care. Nor is it known whether dumpers more likely to have participated in other clinical trials. Furthermore, it is also not known whether patients felt free to reveal problems with compliance and the features of the drug product that contributed to problems with compliance. Ultimately, this type of information is critical to improve our understanding of compliance in clinical CHEST / 119 / 4 / APRIL, 2001


trials.2– 4 It could lead to more accurate drug labeling and stimulate the development of pharmaceutical products that are truly effective and functional. Elizabeth Koller, MD Saul Malozowski, MD, PhD Steven Johnson, PharmD Rockville, MD

pliant patients could in fact be related to discomfort with discussing compliance issues with study personnel. We can only speculate, however, on what the motivation for dumping may be. Because clinical trials are generally not designed to study questions of this type, future studies that focus on the motivations for compliance and deceptive practices among participants in clinical trials are needed. Michael S. Simmons Donald P. Tashkin, MD, FCCP UCLA School of Medicine Los Angeles, CA

Correspondence to: Elizabeth Koller, MD, 5600 Fishers Lane, Parklawn Building, Room 14B04, Rockville, MD 20857

References 1 Simmons MS, Nides MA, Rand CS, et al. Unpredictability of deception in compliance with physician-prescribed bronchodilator inhaler use in a clinical trial. Chest 2000; 118:290 –295 2 Enstrom I, Pennert K, Lindholm LH. Durability of improvement achieved in a clinical trial: is compliance an issue? J Fam Pract 2000; 498:634 – 637 3 Li BD, Brown WA, Ampil FL, et al. Patient compliance is critical for equivalent clinical outcomes for breast cancer treated by breast-conservation therapy. Ann Surg 2000; 231: 883– 889 4 Crim C. Clinical practice guidelines vs actual clinical practice: the asthma paradigm. Chest 2000; 118(2 suppl):62S– 64S To the Editor: The correspondents raise some interesting questions regarding our report1 (August 2000) on compliance in the Lung Health Study, as well as the importance of future investigations in this field. The 10-center Lung Health Study was not initially designed for the purpose of investigating compliance issues, as they correctly point out. While the ancillary study on which our article is based was designed to study compliance at the two centers at which electronic medication monitors were used, some constraints and limitations were necessarily posed by the design and organization of the parent study. The patients’ motivation for entering our study cannot be determined with certainty, but no medical treatment was promised or provided except for the interventions specified in the study protocol. Possible incentives included the smoking cessation program (one half of all participants) and bronchodilator inhalers (one third of all participants received active bronchodilator inhalers). Because most participants did not have any serious or disabling symptoms (asthmatics were excluded) and no medical treatment was promised, we believe that the bronchodilator inhaler was not a motivating factor for most subjects to enter the study. We feel the promise of pulmonary function testing and the possibility of assistance with smoking cessation were the primary motivations for patient participation. The nature of the relationship between study participants and the investigators was not described in our article but, in fact, patients of study investigators were excluded from the study to avoid conflict of interest. There are no data on the patients’ medical insurance coverage or their previous participation in other clinical trials, although concurrent participation in non-Lung Health Study clinical trials was not permitted. Study participants were asked about their bronchodilator inhaler use at each clinic visit. When poor compliance was revealed, they were queried about possible causes, including difficulties in scheduling or remembering their inhaler use, as well as any problems they may have had with the medication itself. Because the intent was to improve compliance with the study medication, free discussion was encouraged and we believe participants were comfortable discussing the use of their inhalers with clinic personnel. A notable exception, of course, would be the “dumpers” who denied problems with compliance. The motivation behind dumping in this interesting subset of noncom1292

Correspondence to: Michael S. Simmons, UCLA, Division of Pulmonary and Critical Care, Los Angeles, CA 90095-1690; e-mail: [email protected]

Reference 1 Simmons MS, Nides MA, Rand CS, et al. Unpredictability of deception in compliance with physician-prescribed bronchodilator inhaler use in a clinical trial. Chest 2000; 118:290 –295

Pneumothorax Is Chest Tube Clamp Necessary Before Removal? To the Editor: I read with interest the study, “Pneumothorax: Experience With 1,199 Patients” (May 2000).1 While the discussion part of the article was thorough and informative, the statement that the pleural drain should be removed without clamping after the cessation of air leak needs further exploration. As the choice of therapeutic procedure varies according to the number of episodes, the size of pneumothorax, and the particular group of investigators, chest tube removal should also be dependent on the underlying lung disease, the presence or absence of bronchopleural fistula, the history of pneumothorax, and the severity of the initial episode. Primary spontaneous pneumothorax has been viewed as lowmortality nuisance.2 Chest tubes may be removed as soon as the lung expands and air leak ceases. In the presence of underlying lung disease or when pneumothorax has been associated with bronchopleural fistula, it is always imperative to clamp the tube for a variable duration before it is removed. Many physicians also agree that the chest tube should be clamped before it is finally removed.3– 6 Similar is the recommendation in the Light textbook on pleural diseases.7 An analysis on pneumothorax management performed by Baumann and Strange8 also verifies the above view, where at least 67% of the responders suggested chest tube clamping for a period varying from 4 to 24 h. Another 27% in their study preferred to wait for at least 24 h after the air leak stopped before the chest tube was removed, although they did not prefer a clamp. Similar was the authors’ approach in the present study,1 but they ultimately suggested not to wait or clamp the tube. In my experience, it is not uncommon to observe a relapse of pneumothorax after removal of chest tube as soon as the lung expands or air leak ceases. I would also like to mention an Indian study9 performed by my senior colleagues, in which they observed that the early removal (6 h) of the chest tube after complete expansion of the lung results in a greater relapse rate than removal after 48 h. The onset of breathlessness at a variable time after a clamp would definitely indicate more prolonged chest tube drainage. Communications to the Editor