|Skip left side navigation and go to content||
Investigative Bronchoprovocation and Bronchoscopy in Airway Diseases
Report of a National Heart, Lung, and Blood Institute/National Institute of Allergy and Infectious Diseases Workshop
Published in Am J Respir Crit Care Med Vol 172. pp 807–816, 2005 Internet address www.atsjournals.org
William W. Busse, Adam Wanner, Kenneth Adams, Herbert Y. Reynolds, Mario Castro, Badrul Chowdhury, Monica Kraft, Robert J. Levine, Stephen P. Peters, and Eugene J. Sullivan
From the University of Wisconsin-Madison, Madison, Wisconsin; University of Miami School of Medicine, Miami, Florida; National Institute of Allergy and Infectious Diseases, Bethesda; National Heart, Lung, and Blood Institute, Bethesda; Washington University School of Medicine, St. Louis, Missouri; U.S. Food and Drug Administration, Rockville, Maryland; National Jewish Medical & Research Center, Denver, Colorado; Yale University School of Medicine, New Haven, Connecticut; and Wake Forest University Health Sciences, Winston Salem, North Carolina.
This workshop, sponsored by the National Heart, Lung, and Blood Institute and the National Institute of Allergy and Infectious Diseases of the National Institutes of Health, was held in Bethesda, Md., July 25 and 26, 2003.
Rationale: Basic and clinical research strategies used for many lung diseases have depended on volunteer subjects undergoing bronchoscopy to establish access to the airways to collect biological specimens and tissue, perhaps with added bronchoprovocation in asthma syndromes. These procedures have yielded a wealth of important scientific information. Since the last critical review more than a decade ago, some of the techniques and applications have changed, and untoward events have occurred, raising safety concerns and increasing institutional review scrutiny. Objectives and Methods: To reappraise these investigational methods in the context of current knowledge, the National Heart, Lung, and Blood Institute and the National Institute of Allergy and Infectious Diseases of the National Institutes of Health convened a working group to review these procedures used for airway disease research, emphasizing asthma and chronic obstructive pulmonary disease. Main Results: The group reaffirmed the scientific importance of investigative bronchoscopy and bronchoprovocation, even as less invasive technologies evolve. The group also considered the safety of bronchoscopy and bronchoprovocation with methacholine and antigen to be acceptable for volunteer subjects and patients, but stressed the need to monitor this closely and to emphasize proper training of participating medical research personnel. Issues were raised about vulnerable volunteers, especially children who need surrogates for informed consent. Conclusion: This review of investigative bronchoscopy and bronchoprovocation could serve as the basis for future guidelines for the use of these procedures in the United States.
Keywords: airway hyperresponsiveness; asthma; bronchoalveolar lavage; chronic obstructive pulmonary disease; lidocaine; methacholine; segmental allergen challenge
RATIONALE FOR WORKSHOP
Bronchoprovocation and fiber optic bronchoscopy have become an integral part of research involving human subjects, especially for the study of chronic pulmonary illnesses such as asthma, chronic obstructive pulmonary disease (COPD), and interstitial lung diseases. These investigative approaches have facilitated the acquisition of quantitative biological data that have advanced our understanding of disease mechanisms and now form the basis of many of the current concepts in pathogenesis, diagnosis, and treatment. Although considerable insight into the mechanisms and regulation of airway hyperresponsiveness (AHR) and inflammation has been gained from animal models, knowledge about the uniqueness of these processes in human airway diseases can only be studied and obtained in patients. Experience has found that the risk–benefit ratios of bronchoscopy and bronchoprovocation are relatively low, rendering these current procedures feasible for research involving patients and healthy control subjects, including children in special situations. Investigators, funding agencies, pharmaceutical companies, and regulatory organizations now accept bronchoprovocation and bronchoscopy as useful scientific tools and rely on the research generated by these procedures to determine mechanisms of altered lung function in humans.
Although investigative bronchoprovocation and bronchoscopy previously have been reviewed, there is now a need to revisit the application of these procedures to research. First, the most recent committee reviews were published over a decade ago (1, 2), and other helpful articles that addressed bronchoscopy and bronchoalveolar lavage (BAL) in adults and children with airway disease (3–6) did not focus on investigative uses. Second, the procedures have become more standardized, and the experience gained has been expanded so that a current review and reassessment of techniques for these procedures and their applications will facilitate the comparison of research findings among different laboratories. Third, there are a growing number of investigators who include bronchoscopy, bronchoprovocation, or both in their research protocols such that an update on the indications, techniques, and safety of these popular investigative tools is both timely and instructive. Finally, ethical considerations, regulatory requirements, and institutional review board diligence may not have been adequately covered in previous reviews, necessitating an assessment of such issues.
On July 25 and 26, 2003, the National Heart, Lung, and Blood Institute and the National Institute of Allergy and Infectious Diseases convened a workshop bringing together clinical investigators, research support staff, ethicists, and representatives of federal agencies for the purpose of discussing and evaluating the use of bronchoprovocation and bronchoscopy in airway disease research. The goals of the workshop were as follows: (1) review the experience and status of bronchoprovocation and bronchoscopy in the study of asthma and COPD; (2) evaluate and provide a rationale for using these investigative tools in preference over less invasive methodologies; (3) review the technical requirements, ethical implications, and regulatory aspects of the procedures where necessary; and (4) offer broad directives and suggestions, but not formal recommendations, for their use and safe application in human research.
This article is intended to review the state of the art of investigative bronchoprovocation and bronchoscopy at the time of the workshop and thus provide information that might be useful for clinical investigators, research sponsors, and research volunteers, and to highlight any related issues or topics that might need future consideration. Subsequent publications, some of which are included in this report, have added support to the workshop’s conclusions.
The workshop was not intended to formulate consensus-based guidelines for investigative bronchoprovocation and bronchoscopy but rather to summarize the scientific contribution and safety of these tools that could be used for future guideline development in the United States.
BRONCHOPROVOCATION IN ADULTS: APPLICATIONS AND CONTRIBUTIONS
AHR, an exaggerated bronchoconstrictor response to a variety of stimuli, is a prominent characteristic of asthma and also found in COPD, cystic fibrosis, and allergic rhinitis. Although the mechanisms underlying AHR are not fully understood, it is believed to result, at least in part, from airway inflammation. Bronchoprovocation is a well-established method to detect and quantify AHR and to obtain insights into the mechanisms associated with this pathophysiologic abnormality, particularly when assessed in conjunction with procedures such as bronchoscopy and mucosal biopsy.
Pharmacologic agents, including acetylcholine, methacholine, histamine, cysteinyl leukotrienes, prostaglandins, and adenosine 5'-monophosphate, and physical stimuli such as exercise and isocapnic hyperventilation with cold, dry air, have been used to detect, quantify, and characterize nonspecific AHR in asthma. Experience has indicated that AHR varies with the clinical severity of asthma and, largely based on the observation that anti-inflammatory therapy can reduce AHR measures of airway responsiveness, has been used as an indirect physiologic marker of airway inflammation (7). However, it is important to appreciate the potential differences that may arise from values of airway responsiveness to direct airway smooth muscle constrictors like methacholine versus responses that may follow more indirect stimuli like cold air or adenosine 5'-monophosphate.
Inhalation of allergens by allergic patients, with or without asthma, is often used to define mechanisms underlying the development of airway inflammation. Such insight has been aided by the differences between the immediate bronchospastic response to allergen and the development of the late allergic response, which is characterized by airway inflammation and enhanced airway responsiveness (8). Inflammatory cell function and phenotype may be altered by allergen challenge to modulate allergic inflammation, to provide insight into altered airway function and thus facilitate correlations among the cells and mediators of this complex inflammatory process and altered pulmonary physiology. However, insights into the complex mechanisms of the late phases are also undergoing reevaluation, as well as the clinical significance of treatments that affect this component of the airway response to antigen.
Insight into Disease Mechanisms with Bronchoprovocation
The basic techniques and applications of bronchoprovocation, both for testing nonspecific airway responsiveness and specific allergen challenge, have been well defined (9). The challenges include provocative agents that induce bronchoconstriction directly or indirectly by the release of spasmogens from airway cells (10). An alternative to allergen/antigen bronchoprovocation, either through whole lung aerosol challenge or bronchoscopic segmental allergen challenges (see below), is natural seasonal allergen exposure in allergic individuals with well-defined seasonal rhinitis or asthma (11). In contrast to laboratory challenge procedures, seasonal exposures cannot be precisely controlled. One additional recent modification to antigen provocation has been the use of repetitive, low-dose airway challenges, in lieu of a single dose, to investigate possible enhancement or tolerogenic mechanisms involved in the modulation of allergic airway inflammation (12).
AHR to methacholine correlates in a general way with symptoms and severity of the disease (7) and is a risk factor for progressive airflow obstruction and an accelerated rate of decline in FEV1 in smokers (13). Increased airway responsiveness after allergen inhalation parallels the subsequent inflammatory reaction, suggesting that the associated allergen-induced inflammation has direct effects on mechanisms of airway responsiveness (14); the observation that allergen avoidance can decrease AHR is consistent with this concept (15). Inhaled glucocorticosteroids reduce airway inflammation and to some extent bronchial responsiveness, further supporting the linkage between these two processes (14). Certain therapeutic agents, such as inhaled glucocorticosteroids and leukotriene receptor antagonists, have differential effects on the early and late response seen after allergen challenge, suggesting selective mechanisms of action (16, 17). Furthermore, late-phase airway responses to allergen inhalation are associated with inflammatory markers that are reflected both in the circulation and the airways somewhat differently, suggesting distinct and interactive effects in these two systemic compartments.
A number of physiologic factors characteristic of asthma have either been defined or explored using techniques of bronchoprovocation. These include the “excessive” airway closure that is characteristic of asthma (i.e., lack of a plateau on the methacholine dose–response curve). In addition, there appears to be a lack of regulatory mechanisms to restore airway caliber after bronchoconstriction (i.e., the effect of a deep breath), and the loss of airway–parenchymal interdependence (18). Moreover, when used appropriately, bronchial challenge studies have been a helpful tool for developing therapy. That is, pharmacologic intervention of the early and particularly the late allergic response after allergen challenge have successfully predicted that both leukotriene D receptor antagonism (17) and blocking IgE with a specific antibody (19) would be useful for the treatment of asthma. Conversely, studies with allergen challenge have also successfully predicted that platelet-activating factor antagonism would not be clinically beneficial in asthma treatment (20). Therefore, the collective experience with bronchoprovocation in humans using nonspecific stimuli or allergens has provided important in vivo information about the pathogenesis and pathophysiology of asthma that also has contributed directly to the development of new therapeutic strategies and underlying association between inflammation and AHR.
The availability of a U.S. Food and Drug Administration (FDA)-approved preparation of methacholine (Provocholine, Methapharma) (21) provides a well-defined agent for clinical and research use that has now largely precluded the need for histamine in the United States. Bronchoprovocation with methacholine has been standardized and is considered to be acceptably safe when established procedures are followed. The most important contraindication to performing this and other whole lung inhalational tests is a low baseline FEV1, usually considered to be below 70% predicted, as noted in the manufacturer’s package insert for Provocholine. However, one study has reported that methacholine testing can be performed safely in subjects with FEV1 values ranging between 22 and 59% of predicted (22), and several clinical networks and studies have used methacholine bronchoprovocation safely in subjects with severe asthma (22) and COPD (23). However, the safety experience in such patients is still limited, and the magnitude of airflow obstruction under which this procedure can be used safely will need further evaluation; the data may be available through the Severe Asthma Research Program.*
* This program, sponsored by the Division of Lung Diseases, National Heart, Lung, Blood Institute, and initiated in September 2003, supports a collaborative multicenter study in humans to investigate the mechanistic basis for severe asthma and to identify novel targets for potential therapeutic intervention. The goals are to reduce morbidity and mortality in patients with severe asthma and to lessen the substantial health and economic burden attributable to this disorder.
For the other above-mentioned pharmacologic agents and physical stimuli used to assess nonspecific airway responsiveness, the methodologies have been less well standardized, and there is less experience with their safety profile (24–26). Likewise, the safety of inhalational challenge with allergens, approved by the FDA for human use as skin-test reagents but not for lung challenge studies, is less well established. However, serious adverse events have not been reported (27).
In summary, research bronchoprovocation appears to carry a low risk of untoward effects. However, the proven safety record of investigative methacholine challenge should not be extrapolated to other less frequently used provocative agents for which the same high level of experience is lacking. The introduction of new agents to bronchoprovocation protocols will require that safety, as well as biological relevance, be established before such agents are applied on a wider scale. Two tragedies involving research subjects have underscored the importance of these safety issues (28, 29). Because improving and monitoring the protection of research subjects is a high priority in clinical research (28), further details about preclinical data requirements as well as other ethical concerns are provided in Appendices E1 and E2 in the online supplement.
BRONCHOSCOPY IN ADULTS: APPLICATIONS AND CONTRIBUTIONS
BAL (30–32), endobronchial brush or forceps biopsy, and transbronchial biopsy (33) have emerged as the most widely used invasive research tools to assess inflammation and tissue remodeling in airway and interstitial lung diseases (Table 1) (34). Endobronchial biopsy specimens can also be prepared as explants for further study (35).
TABLE 1. CONTRIBUTIONS OF RESEARCH BRONCHOSCOPY TO THE UNDERSTANDING OF ASTHMA AND CHRONIC OBSTRUCTIVE PULMONARY DISEASE
Investigative bronchoscopy and related procedures provide clinically relevant in vivo information on the pathophysiology of asthma (36–43) and COPD (44–47). For example, profiles of infiltrating inflammatory cells and markers of their level of activation have been reported to correlate with physiologic parameters in asthma (42, 47–49) and COPD (44, 46, 47). Thus, much of our recent knowledge about the histopathology of mild asthma, the similarities and differences of airways in patients with allergic rhinitis versus those with allergic asthma, and the histologic subtypes of severe asthma in patients has come from bronchoscopic studies. Because the complexity of these inflammatory processes and the attendant changes in tissue architecture cannot be fully reproduced in animal models, research bronchoscopy in humans has been required to provide direction and insight into mechanisms of airway disease not available through other means.
Bronchoscopy can be used in conjunction with allergen challenge to correlate changes in pulmonary function with inflammatory cell recruitment into the airways (50, 51). Allergen can be delivered by aerosolization into the whole lung or via instillation through the bronchoscope into an isolated airway (i.e., segmental allergen challenge) (52–55). Segmental allergen challenge tends to better localize the site of allergen delivery, and higher doses of allergen can be used to induce greater localized inflammation with less overall bronchoconstriction. In addition, multiple segments of the airways can be challenged at the same time with different doses of antigen, or pharmacologic agents can be added in an attempt to block the in situ inflammatory response.
The limitations of segmental allergen challenge include inter and intrasegment variabilities in the inflammatory response to allergen and the fact that this model does not necessarily mimic allergen-induced asthma exacerbations (43).
Research BAL and bronchoscopic biopsy procedures have not been standardized, and the reproducibility of reported findings often is not known. For example, it is not always clear how well a specific airway biopsy specimen reflects the histology of other airways and what the level of histologic similarity is between multiple biopsies in asthma and COPD. The ideal number of biopsies recommended in the previous workshop was 3 (1), but several studies have obtained 9 to 10 endobronchial biopsies at the same procedure without complications (56, 57).
Insight into Disease Mechanisms
In asthma, biological events associated with the initiation, propagation, and resolution of an acute inflammatory reaction in the airways have begun to be clarified with this research approach (51, 58, 59). Those studies were performed in the context of an acute allergen challenge, although events after amechanical injury have also begun to be explored. The role of infectious agents, including viruses (e.g., rhinovirus) and atypical pathogens (i.e.,Mycoplasma pneumoniae and Chlamydia pneumoniae), in the pathogenesis and exacerbations of asthma has been studied using bronchoscopic methods (60, 61). Direct measurements of peripheral airway resistance and responsiveness as well as the characterization of airway surface liquid in thermally induced asthma also have been investigated by bronchoscopy (62–64). Moreover, the application of specialized techniques, such as bronchoscopic endobronchial ultrasonography, has been important for some of these findings (65). Finally, using appropriate clinical protocols designed to take into account intersubject variability and intrasubject reproducibility of airway sampling (54), bronchoscopic techniques have provided direct measurement of the effect of different drugs and therapy on airway inflammation and airway structural changes that occur in asthma and COPD (32, 66, 67). Recently, transbronchial lung biopsy has been performed in patients with asthma and demonstrated a significant inflammatory component in the lung periphery (42). Thus, the use of bronchoscopy in airway research has had broad application and has provided significant new information, especially on the nature of airway tissue inflammation in airway disease (68–70).
A number of reports have suggested that bronchoscopy, including BAL and bronchial forceps and brush biopsy, can be safely performed for research purposes in patients with asthma and COPD (34, 71–74). Repeated procedures have also been reported to be safe, at least in patients with mild asthma (59). Experience is more limited with transbronchial lung biopsy (42); thus far, serious adverse events have not been reported. However, because a death has been reported in a normal subject undergoing research bronchoscopy (28, 75), safety considerations generally need to be constantly reassessed (Table 2).
TABLE 2. SAFETY OF RESEARCH BRONCHOSCOPY IN OBSTRUCTIVE LUNG DISEASE
Definition of abbreviations: BAL = bronchoalveolar lavage; COPD = chronic obstructive pulmonary disease; PEFR = peak expiratory flow rate.
Bronchoscopy. Previous guidelines have suggested that an FEV1 less than 60% constitutes a contraindication to performing research bronchoscopy (1). However, bronchoscopy in adults with asthma has been performed safely when the FEV1 is lower, such as less than 50% predicted post-bronchodilator, and in patients with COPD (30, 72, 76) when the FEV1 is less than 25% predicted prebronchodilator. Furthermore, in a single report, Martin and colleagues (22) demonstrated that bronchoscopy was safely performed in subjects with asthma with an FEV1 less than 30% predicted. However, more experience needs to be gained before research bronchoscopy can be assumed to be safe in patients with such severe airflow obstruction. Premedication with atropine and bronchodilators can be given, or omitted, depending on the procedures to be performed (34) and the number of bronchoscopies a research subject may safely undergo over time.
Segmental allergen challenge. Segmental allergen challenge in patients with allergic airway disease is generally well tolerated. There have been reports of increased AHR, wheezing, and decrements in lung function in relationship to this procedure (76); however, the changes in AHR can be more pronounced when the segmental challenge is followed by BAL and biopsy, and has lasted up to 72 hours (76). Jarjour and colleagues (77) reported no significant difference in lung function measured 2 hours after a segmental allergen challenge in subjects with asthma as compared with subjects with allergic rhinitis. When BAL and biopsy were performed after whole lung allergen challenge, no further change in AHR was documented (78). Thus, the experience to date suggests that these challenge procedures are well tolerated.
Topical anesthesia. Topical anesthesia is needed for bronchoscopy and can be a source of increased risk. Previously, an upper dose limit of lidocaine of 400 mg was recommended (1), but a recent report suggested 600mg or 9 mg/kg as a safe limit for adults (79). Despite these recommendations, a safe upper limit for topical lidocaine used in bronchoscopy has not been firmly established, and instillation of lidocaine into the airways is not an FDA-approved route of administration. It has been suggested that the death of the above-mentioned volunteer subject undergoing bronchoscopy could have been related to lidocaine usage (28).
BAL. This is also generally well tolerated, although changes in lung function have been reported (76). BAL does not significantly alter airway responsiveness, airflow limitation, and/or airway inflammation when BAL is incorporated into other procedures, such as bronchial biopsy, and segmental or whole lung allergen challenge (71, 76–78, 80). Nonetheless, cough, wheezing, and post-bronchoscopy fever have been associated with BAL. Hypoxemia is less common, particularly because the use of supplemental oxygen has become a standard practice (80).
Bronchial biopsy. Bronchial forceps biopsy has become a routine investigative procedure, and a number of studies have addressed subject safety, often in the setting of combined biopsy and BAL. Djukanovic and colleagues (71) showed that endobronchial biopsy (up to four biopsies taken) with a BAL volume of 160 ml in 20 subjects with mild to moderate asthma (FEV1 values of 88 +/- 18% predicted) resulted in a fall in FEV1 of 26 +/- 17%, which correlated with AHR before the procedure. Healthy control subjects were also assessed, and their mean decrement in FEV1 was 9 +/- 4.7%. In addition, hypoxemia was found in asthma (mean fall of oxygen saturation, 3%; range, 1–17%) but not in normal control subjects. This report differs from that of Van Vyve and colleagues (80) who studied 50 subjects with asthma and 25 control subjects undergoing bronchoscopy with four endobronchial biopsies and a BAL volume of 250 ml. These investigators did not administer a Beta2-agonist before the procedure, but the subjects with asthma received it after the procedure; no supplemental oxygen was administered, unless an oxygen desaturation of less than 80% for more than 1 minute was observed. Decrements in arterial oxygen saturation were observed to a similar extent in subjects with asthma and control subjects, with a mean decrease in oxygen saturation of approximately 4 to 5%; the decrease in FEV1 was similar in subjects with asthma (20% mean) and healthy control subjects (17% mean) (80).
Although these studies suggest that bronchoscopy induces transient changes in airway function and gas exchange in healthy subjects and subjects with asthma, asthma control, as determined by peak expiratory flow rate, symptom score, and medication use, appears not to be lost after bronchoscopy involving biopsy and BAL (56).
The safety of bronchoscopy with endobronchial biopsy and BAL also has been assessed in COPD by Hattotuwa and colleagues (72). Fifty-seven patients with COPD whose FEV1 ranged from 25 to 75% predicted (mean, 44.5% predicted) underwent either bronchoscopy with endobronchial biopsy and BAL (68 procedures) or endobronchial biopsy alone (30 procedures). Eleven patients had mild disease, 28 were considered moderate, and 18 were considered severe according to British Thoracic Society guidelines (81); all were considered to have stable disease. In these 98 procedures, five adverse events occurred, including bronchospasm that required hospitalization (one subject), pneumothorax (one subject with severe disease), and hemoptysis (three subjects, but no hospitalization was required). The overall incidence of adverse events requiring hospital treatment was 2 and 3.1% for minor hemoptysis.
Information about the safety of transbronchial biopsy in asthma is limited and has been obtained primarily from studies at one center in the United States involving 49 subjects undergoing 72 bronchoscopies (42, 57, 82). The procedure was deemed safe by those investigators, although one subject experienced a 10% pneumothorax, which resolved with supplemental oxygen (42).
Airway brushing. Airway brushing during bronchoscopy is generally considered safe and well tolerated (83). Bleeding can occur but is rare, and cough is self-limited. Although the previous workshop (1) recommended a maximum of three brushings per bronchoscopy, recent studies have reported that subjects can tolerate more extensive brushings, with up to 24 brushings performed during a single bronchoscopy (84).
BRONCHOPROVOCATION AND RESEARCH BRONCHOSCOPY IN CHILDREN: APPLICATIONS AND CONTRIBUTIONS
Fiber optic bronchoscopy is routine in children and is used to evaluate stridor and recurrent pneumonia and exclude foreign bodies or infectious etiologies, and is an established, useful procedure in clinical medicine (85, 86). For example, Godfrey and colleagues (87) evaluated 200 consecutive bronchoscopes in children and found that bronchoscopy yielded abnormal findings in approximately two-thirds of the cases. Information from these procedures contributed to clinical management in approximately 90% of patients.
Because concepts of asthma pathophysiology and inflammation in adults cannot be extrapolated to children (74, 88, 89), research fiberoptic bronchoscopy in combination with bronchoprovocation has been recently extended to children and adolescents. However, only a limited understanding exists about the underlying pathology of wheezing phenotypes in children (90). BAL analysis in children with asthma or persistent wheezing has suggested that there is an increase of eosinophils and neutrophils compared with normal control subjects and those with chronic cough (90, 91). Furthermore, bronchoscopic studies and lavage fluid analyses suggest that a pure virus-associated wheeze is not just an eosinophilic disease in children (89, 92).
Although bronchoprovocation can be performed with acceptable safety in children, as noted in several large studies, including the Childhood Asthma Management Program (93), information on applications and safety of research bronchoscopy thus far has been derived primarily from clinical experience. Diagnostic flexible bronchoscopy in children has become routine for the evaluation of stridor and pulmonary infections. Bush and Pohunek (94) demonstrated with 278 endobronchial biopsies obtained from children that there were no complications other than minor bleeding. Another study evaluated the safety of bronchoscopy in severe or difficult-to-control asthma as determined by medication requirements (68). This 3-year, prospective, observational study was conducted in two tertiary pediatric respiratory centers specializing in the management of severe asthma. Bronchoscopy was performed in 38 children with mild to severe asthma, with FEV1 values that ranged from 44 to 104% predicted, and in 35 nonasthmatic control subjects; rigid bronchoscopy was performed in 10 children, after a course of prednisolone. Perioperative complications occurred in one patient undergoing flexible bronchoscopy (oxygen desaturation) and two undergoing rigid bronchoscopy (desaturation and bronchospasm); four patients with asthma did report an increase in symptoms 1 week after bronchoscopy.
The largest published study with pediatric bronchoscopy is by de Blic and colleagues (6), who described their experiences in 1,328 pediatric patients undergoing bronchoscopy in a clinical setting. Of these patients, 3% had the diagnosis of asthma, and an additional 30% had airway disease (recurrent wheezy bronchitis, persistent cough, or bronchiectasis). Most of the procedures (92.8%) were performed in conscious patients with sedation, and 7.2% were performed under deep sedation with an endoscopic facemask. Minor, expected complications consisting of cough and epistaxis occurred in 46 of 1,328 subjects. Major complications were rare (n = 22) and included oxygen desaturation less than 90% (n = 16), coughing (n = 4), bronchospasm (n = 1), and pneumothorax (n = 1). Thus, the authors concluded that flexible bronchoscopy is a safe procedure in children, with complications occurring in less than 2% of the procedures.
The risk of the procedure lies not just with the bronchoscopy but also with deep sedation and/or the anesthesia. Therefore, if the child is being anesthetized for another purpose, it is legitimate to seek consent for a BAL and/or an endobronchial biopsy, provided these are obtained by an experienced bronchoscopist and no contraindication exists (i.e., coagulopathy or respiratory compromise from a lung disease). The additional use of lavage specimens for research purposes in children with asthma, recurrent wheeze, or cough has also been reported (91, 95). To understand the biology of the developing airways, information from normal control subjects would be optimal to compare with subjects with asthma. In infants with cystic fibrosis, BAL analysis has suggested that inflammation and infection occur early in life (96–98). Furthermore a study demonstrated that BAL in healthy children undergoing elective surgery was safe (90).
ALTERNATIVES TO INVESTIGATIVE BRONCHOPROVOCATION AND BRONCHOSCOPY
Because of the importance of airway inflammation in the pathogenesis of asthma and COPD, investigators have sought to develop techniques to assess airway inflammation by noninvasive approaches, such as analysis of induced sputum (99, 100), exhaled gases, and breath condensates (101). The utility of these alternative research tools is increasingly being recognized (30, 102). For example, the distribution of inflammatory cells found in induced sputum correlates reasonably well with the cell spectrum obtained by lavage, particularly with respect to eosinophils (103). Likewise, the exhaled level of nitric oxide appears to be a sensitive indicator of airway inflammation and decreases rapidly with the administration of glucocorticosteroids, including low-dose inhaled glucocorticosteroids (104). Recently, lung imaging with hyperpolarized helium and computed tomographic scanning has provided quantitative data on some aspects of airway remodeling (105). However, a major limitation of currently available noninvasive alternatives to bronchoscopy is their failure to provide tissue specimens for histologic, immunologic, and molecular analyses. In addition, the wide variety of detachable lung cells found in expectorated sputum limits their utility for ex vivo studies.
NEED FOR NEW STANDARDS FOR INVESTIGATIVE BRONCHOPROVOCATION AND BRONCHOSCOPY
In discussing research applications of bronchoprovocation and bronchoscopy, the workshop participants suggested that previous recommendations may have to be revisited and expanded. More procedures are being coupled together; use of existing agents to challenge the airways has been modified; and new agents are likely to be introduced. Evidence to support continued use of research modalities, as discussed, was substantiated by published citations and by personal experience of the participants, many of whom are leaders in this field of respiratory research. The procedures reviewed were judged to be acceptably safe for normal subjects and patients, as documented in published reports. But it is also apparent that this field of airway research is dynamic and growing such that a framework for considering new adaptations or incorporating new pharmacologic agents or allergens in research protocols might be needed. Therefore, in addition to the general conclusions and suggestions offered by workshop participants (Tables 3 and 4), detailed information about the regulatory requirements for the unapproved use of an approved drug or the use of a new drug or substance would be helpful to include for reference. Thus, information was prepared by the participants from the Division of Pulmonary and Allergy Drug Products of the FDA (Appendix E1).
The tradition of using volunteer normal subjects and patients for medical research is well established and is essential for advancing new knowledge about diseases, and this has been accompanied by a constantly evolving and challenging range of ethical issues. In particular, as new indications arise for the investigation of younger subjects with airway diseases, who need to be well protected and usually require surrogate consent, special ethical considerations must be evaluated. Perhaps the creation of a registry to track the safety and outcome of volunteer subjects should be considered. Participants in the workshop strongly endorsed the continuation of clinical research in airway diseases, but some cogent ethical issues were raised that were beyond the scope of the workshop. Some of these are presented for reference in Appendix E2.
CONCLUSIONS AND SUGGESTIONS
Workshop participants concluded that bronchoprovocation and research bronchoscopy were among the important technical developments in pulmonary research during the past 25 years, and have contributed significant insights about the pathogenesis of asthma and COPD (Table 3). Continued use of these procedures was affirmed by the workshop, including the introduction of novel techniques as they become available (Table 4).
These procedures will continue to provide further opportunities for longitudinal observations, including studies of the natural history of disease, an area of particular importance that should include infants and children at high risk for the development of asthma. As new targets for therapy are identified through gene profiling and other applications of genomic and proteomic research, the study of airway-associated cells and tissues from subjects with asthma and control subjects ex vivo will be important in defining the roles of specific mediators, cytokines, and chemokines, as well as neural and inflammatory pathways. Furthermore, the overall safety profile of these research procedures has been favorable, albeit with some precautions and limitations acknowledged.
In the future, imaging modalities, such as virtual bronchoscopy, may either replace bronchoscopy (for some indications) or provide complementary information. These approaches likely will include ultrasound techniques and transbronchial lung biopsies. In addition, nasal brushing or lavage is an alternative to airway tissue obtained in a less invasive manner (106). Airway mapping to identify segmental or airway generational differences relevant to airway pathophysiology, including temperature, oxygen tension, concentration of gas phase molecules, such as nitric oxide and pH, is also promising (107–109). Finally, evanescent spectroscopy, microarray gene analysis, and a variety of other in vivo and ex vivo assay techniques will enhance the benefit of fiberoptic bronchoscopy as a research tool.
Expanded use of research bronchoscopy in pediatric subjects is promising and should lead to a better understanding of the complex airway processes that occur early in life. Although clinical phenotypes have been described in early childhood for transient or early-onset wheezing and persistent wheezing, which can reduce lung function, there is still little understanding of the underlying pathophysiology of these disorders (110). The application of investigative bronchoscopy in children is still in its early stages and poses special opportunities and requirements before this area of study is fully explored (Appendix 2).
TABLE 3. CONCLUSIONS OF THE WORKSHOP
Definition of abbreviation: COPD = chronic obstructive pulmonary disease.
TABLE 4. RECOMMENDATIONS OF THE WORKSHOP
For definition of abbreviation, see Table 3.
Conflict of Interest Statement:
W.W.B. has received consultancy fees for the past 3 years from the following companies, with a total consultancy fee for these 3 years as indicated: Bristol Myers Squibb ($2,000), Dynavax ($3,000), Hoffman LaRoche ($2,000), Schering ($3,000, 2002–2003), and Fujisawa ($3,000). He has also served on advisory boards in various capacities over the past 3 years (2001–2003) with the following reimbursements: GlaxoSmithKline (GSK; $8,500), Aventis ($2,000), Schering ($4,000), Pfizer ($4,000, 2004), and AstraZeneca ($2,000). He has also received honorarium for speaking or other educational activities in the past 3 years for Merck ($7,000, 2003), GSK ($2,500, 2003), and Aventis ($2,500, 2003). He has received industry-sponsored support for research from GSK ($750,000, 2002 and 2003) and for participation in multicenter trials: Fujisawa ($250,000, 2002 and 2003), GSK ($500,000, 2001–2003), Aventis ($200,000, 2001–2003), Hoffman LaRoche ($120,000, 2002), Genentech/Novartis ($100,000 in 2002/2003), and Merck ($100,000, 2003). A.W. serves on the GSK COPD Global Expert Panel and received $6,000 in honoraria over the past 3 years. He was also the principal investigator on the GSK academic research grant (airway blood flow in COPD). K.A. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. H.Y.R. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. M.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. B.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. M.K. is a consultant for Genentech ($3,000/year for 2002– 2004) and Merck ($2,000/year for 2001–2004). She has been reimbursed by Genentech ($8,000/year for 2002–2004), Merck ($5,000/year for 2001–2004), Novartis ($3,000/year for 2002–2004), and GSK ($2,500/year for 2001–2003) as a speaker. She was also the principal investigator on a grant sponsored by Genentech from 2000–2003, with the total grant award of $5,300. R.J.L. is a member of the Bioethics Committee of Eli Lilly Corporation; for this he received $5,000 in 2004, $8,194 in 2002, and $4,475 in 2001. S.P.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. E.J.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
Correspondence and requests for reprints should be addressed to:
Herbert Y. Reynolds,
Regulatory Considerations Regarding Investigational use of Agents for Bronchoprovocation
Badrul A. Chowdhury, MD, PhD; Eugene J Sullivan, MD*
*This document is written by the authors in their private capacity and not in their capacity as employees of the US Food and Drug Administration (FDA). The opinions expressed are not intended to convey official FDA policy and are not binding on the regulated industry or the FDA. No support or endorsement by the FDA is intended or should be inferred
A bronchoprovocation study usually involves the administration of a drug to study subjects either by inhalation or by direct instillation of the drug into the airways via a bronchoscope. Currently, no agents are approved for direct instillation into the airway in the US. Methacholine (Provocholine®, Methapharm, Coral Springs, FL) is approved for airway provocation, but only by nebulizer. Thus, drugs that investigators may wish to use in bronchoprovocation studies are either not approved for direct delivery into the lung or not approved for human use at all. In such a circumstance, the use of the provocative agent in research would require the approval of an investigational new drug (IND) application by the US Food and Drug Administration (FDA). In the following sections, we discuss the regulatory framework of the IND process and the elements necessary to support a successful IND application.
Investigators planning to initiate bronchoprovocation studies using either unapproved agents, or drugs approved for other routes of administration (e.g., histamine, allergens), would generally need to file an IND with the FDA. In addition, they must also satisfy the Institutional Review Board (IRB) requirements of the institution at which the study will be conducted. IND and IRB review are complementary but independent processes that have been put into place to enhance the safety of subjects enrolled in clinical studies. Investigators should be familiar with the IND regulations in order to assure compliance with Federal law (21 CFR 312).
In certain circumstances, an investigator may be able to conduct a bronchoprovocation research study using an approved drug product without an IND. FDA regulations (see 21 CFR 312.2(b)(1)) provide that a study is exempt from IND requirement if all the following apply:
An IND application consists of several required forms, a clinical protocol, and other documentation to support the proposed human study. The bulk of the required scientific information is in the areas of chemistry, manufacturing and controls (CMC) information on the drug product, and any preclinical (animal) toxicology data to support the proposed human study. The investigator should also submit any previous human data that may support the proposed clinical study. Depending on the extent and quality of these prior human data as they pertain to safety, animal data may be unnecessary. The FDA uses scientific judgment to decide on the extent of the animal data that will be required. In certain circumstances, depending on the drug product under consideration and the scope of the clinical investigation, the FDA has the discretion to exempt the requirement of an IND. When an IND is submitted, the FDA will review the submitted data within 30 days and will either allow the proposed human study to proceed or place the study on a hold. A clinical investigation that is subject to an IND may not be initiated within this 30-day period (see http://www.fda.gov/drugs/developmentapprovalprocess/howdrugsaredevelopedandapproved/approvalapplications/investigationalnewdrugindapplication/ucm071073.htm for helpful information on IND submissions and necessary forms ).
Chemistry, Manufacturing, and Controls (CMC) Considerations
CMC attributes ensure that the purity and performance characteristics of the drug product are such that it is safe for use in bronchoprovocation studies. Included are adequate information on the amount and nature of any contaminants, impurities, and/or degradation products, the stability of the drug product over the intended duration of use, and the sterility of the drug product. Information should be submitted on the drug substance (the active moiety itself), the drug product components (the drug substance and all other ingredients in the formulation), the manufacturing process, and associated controls for each of these areas. The CMC requirements for investigational drugs for bronchoprovocation generally follow the requirements for new drug applications (1, 2), but for an investigational IND would be much less detailed. A specific requirement (21 CFR 200.51) is that all drug products intended for delivery into the lung, including drug products for bronchoprovocations studies, must be sterile.
Detailed information on the manufacturer and the supplier should be provided. Complete information regarding the source (manufacturer) and comprehensive characterization of the physical and chemical properties for the drug substance should be included in the IND application. This often will require obtaining permission from the supplier or manufacturer to refer to data that they may have previously supplied to the FDA in a Drug Master File. The purity of the drug substance and impurity profile should be characterized.
Use of excipients in a drug product for a bronchoprovocaton study should be limited as much as possible, because lung is sensitive to many foreign materials. Any excipient that has not been previously approved in inhalation products may need to be characterized as described above for drug substance.
Drug Product Manufacturing and Use:
The drug product should be made under aseptic techniques, and a sterile and pyrogen-free diluent should be used. The final drug product should be terminally sterilized by an appropriate method (such as filtration using a sterile 0.2 micron filter). The investigators should develop appropriate specifications and test methods to ensure the purity and reproducibility of the drug product. Bronchoprovocation agents would often be made and used immediately in a research study. However, if the investigator plans to store the drug product for later use, additional specifications parameters should be developed to ensure the purity and stability of the drug product over the duration of storage, and to check for impurities and degradation products that can accumulate over time. In addition the investigator should provide complete information on the container system to ensure that the drug product is compatible with the container and closure system. For instance, a latex stopper in a storage container could lead to the presence of latex antigens in the drug solution. This could have untoward affects in latex allergic individuals.
For biologic products some additional considerations apply. Biologic products often contain unique contaminants and impurities depending on the source material for manufacturing. Allergenic extracts sometimes also have a high content of endotoxin. These contaminants and impurities will need to be characterized and controlled as appropriate. Also for a biologic product, some form of bioassay is recommended to ensure that the final product to be used in a bronchoprovocation study is biologically active.
Biologic products and products in a suspension are often difficult to sterilize. The investigator should develop methods to terminally sterilize the drug product. If that is absolutely not possible, then all the components of the drug product should be manufactured under sterile conditions and the components should be mixed under aseptic conditions.
The goals of the preclinical safety evaluation include the identification of safe starting doses for humans, the characterization of toxic effects with respect to target organs, and the reversibility of these toxic effects. In addition, the mutagenic potential of the drug needs to be documented (3-6). The need for animal data to support a research IND for a bronchoprovocation study would be dependent on the circumstances of use and the particulars of the provocative agent proposed. Wholly new drugs that have never been approved by any route and for which there is little or no human data in the literature would require the most preclinical data. On the other hand, drugs already approved by another route of administration with extensively documented use in the literature, particularly with good documentation of human safety data, may require no additional animal testing.
When required, preclinical toxicology studies should be conducted in at least two animal species, of which one must be a non-rodent species. A range of doses should be studied in both animal species. This will allow identification of a so-called “no observed adverse effect level” that will guide selection of the starting human dose, and also will identify toxic effects that can be monitored in human studies. The duration of exposure for the animals should be equal to or longer than that proposed in humans. The number of groups and the size of the groups should be sufficient to allow meaningful scientific interpretation of the data. Appropriate control groups should be included in the experimental design. The route of drug administration should be the inhalation route to match the human route, and the formulation should be representative of the drug product that will be administered to human subjects.
Prior to any human administration of a drug product, genotoxicity data (e.g., Ames mutagenicity assays and chromosomal aberration assays) for the evaluation of mutations and chromosomal damage would be needed. This may have been previously done (i.e., in approved drug products) or might need to be supplied by the investigator in the IND. If the genotoxicity results are positive, then the investigator must justify why the drug product should be allowed to be used in humans, and must disclose the potential carcinogenicity risk in the informed consent document and in the investigator brochure. Many bronchoprovocation studies utilize drugs approved for other routes of administration and/or other uses (e.g., allergenic extracts). The preclinical requirement for such approved agents would generally be less than for new molecular entities because the systemic toxicity profile of the drug product would have already been studied. If preclinical data were needed, these data should include a so-called “bridging study” to link to the existing preclinical safety data on the drug product to that obtained by inhalation exposure, with the safety study limited to the examination of the airway.
All IND applications must contain an adequately detailed clinical protocol. The focus of the protocol for an investigational IND would be on patient safety. The patient population that will be studied should be specified clearly in the protocol. Generally, subjects enrolled in bronchoprovocation studies would be limited to those with mild or moderate asthma or other airway disease of a severity that will not jeopardize patient safety in the study. Under certain circumstances, subjects with more severe airway disease may be studied, but the use of these subjects would need to be scientifically and ethically justified. In addition, the study design would need to incorporate additional safety measures to protect this more severe population.
Safety parameters, and the intervals at which they will be determined, should be clearly delineated in the protocol. Safety monitoring should include, as a minimum, physical examination, clinical chemistry, and serial spirometry. All study subjects should be monitored for excessive bronchospasm, including late-phase reactions (particularly for whole lung allergen challenge), and the protocol should have appropriate safety measures in place to handle excessive bronchospasm. The protocol should also indicate the outcome measures of clinical interest and state how that data will be gathered.
Conflict of Interest Statement
Neither of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
Ethical Issues Related to Bronchoprovocation and Bronchoscopy Research
Robert J. Levine, M.D. (1), Robert M. Nelson, M.D. PhD (2), Alan R. Fleischman, M.D. (3), Jeremy Sugarman, M.D., M.P.H., M.A (4)
Research involving bronchoprovocation and bronchoscopy is of substantial scientific
importance. The ethical obligation to conduct research involving human subjects was articulated
well by the National Commission for the Protection of Human Subjects of Biomedical and
Behavioral Research in its Belmont Report:
Nevertheless, the conduct of research involving human subjects, including research protocols in which bronchoprovocation and bronchoscopy are used, is governed by policies requiring that all such research conform to scientific and ethical standards. Of special relevance to such research is that the risks and discomforts to which research subjects are exposed must be justified by either the importance of the knowledge to be gained or the potential for direct benefit to the research subjects themselves. Such a determination is prerequisite to obtaining informed consent for participation in a particular research protocol, managing conflicts of interests and obligations, determining the appropriate use of financial incentives and selection of subjects as well as other ethical requirements for research with human subjects. Indeed, the two reported unfortunate deaths of volunteers in this type of research seem related primarily to issues regarding the assessment and minimization of risk.
N. W. died in 1996 from a fatal overdose of lidocaine one to two hours after the completion of a bronchoscopy done for research purposes. Subsequent investigation revealed the lack of guidelines for the maximum dose of lidocaine, inadequate monitoring during the procedure, and lack of adequate IRB oversight (2). E. R. died in 2001 after inhaling the chemical hexamethonium as part of a study designed to test the physiology of airway narrowing in asthma. Although the chemical was not licensed for human use, the investigator did not obtain approval from the Food and Drug Administration (FDA). In a warning letter, the FDA cited the investigator for (among other things): (a) failing to seek FDA approval for the investigational use of the chemical; (b) not submitting animal toxicity data to support safety; and (c) not obtaining adequate informed consent (3).
Limits to justifiable risk
The level of risk is an important attribute of research that determines the substantive and procedural protections that are necessary for the ethical justification of either a category of research or of an individual research protocol. Consideration must be directed not only at the actual risk but also to the perception of risk. For research involving adults, the risks of an intervention that does not offer the prospect of direct benefit must be justified by the importance of the knowledge that reasonably can be expected to result from the research. When children are involved (as discussed below), the allowable risks of such interventions are restricted regardless of the importance of the anticipated knowledge.
The risks of procedures performed only for research may be over- or under-estimated because the data used to estimate such risks are usually derived from clinical experience. Adverse event data from routine clinical procedures may not be systematically and prospectively collected, leading to inaccuracies in estimates of their probabilities. In estimating the risks of performing procedures for research purposes, it is necessary to account for the full range of risks that accompany a procedure, such as those related to the sedation and/or anesthesia that might be used for bronchoscopy. Standards for performance of research procedures such as bronchoscopy and BAL should be developed and these should be based on prospective data collection. The obligation to minimize research risks entails that these standards meet or exceed those that are, or ought to be, in place for the clinical performance of these same procedures. In addition, the risks of performing procedures vary inversely with the experience of the individual who performs them (4, 5). There should also be a mechanism for credentialing individual physician-researchers as competent to perform bronchoscopy, BAL and related procedures for research purposes. Such a mechanism should include the training of fellows in research bronchoscopy only after they have obtained competency and certification with clinical bronchoscopy.
For research protocols that employ bronchoprovocation, there may be a tendency for IRBs to regard “natural” provocations (such as cold air) as less dangerous than the use of drugs (such as methacholine). However, the use of a drug that permits the administration of a specific dose under controlled conditions may be justified as the effects are generally more predictable.
The requirements for informed consent and its documentation when the prospective subjects are legally competent and have the capacity to give informed consent are widely understood and generally non-controversial with regard to protocols involving bronchoprovocation and bronchoscopy (45 CFR 46.116 & 117) (6,7). Much more controversial are the circumstances, if any, in which prospective subjects, having limited capacities to consent, such as children, may authorize their own involvement in research (8, 9). This issue is discussed in further detail below.
Conflicts of interest and obligation
Many investigators have a conflict of interest (COI) or conflict of obligation. A financial COI is easiest to detect; psychological and professional COI and conflicts of obligation can be subtler and therefore more difficult to detect and manage. Nevertheless, managing these conflicts is critical since they may be detrimental to the integrity of the research and may undermine the informed consent process. Institutions, sponsors, and IRBs are responsible for having specific procedures in place to identify conflicts of interest and assure that subjects of research are not compromised by these conflicts. The researcher may consciously or unconsciously engage in behaviors that encourage the subject to acquiesce to the researcher’s requests, and discourage the exercise of the right to withdraw without prejudice. A clinician-researcher’s conflict of obligation may also reinforce the ‘therapeutic misconception’ as subjects have a tendency to believe falsely that all interventions and procedures performed by a clinician-researcher are, at least in part, designed to contribute directly to the health needs of the subject.
In the United States, it is widely agreed that it is ethically appropriate to reimburse subjects for their ‘out-of-pocket’ expenses and to provide modest stipends for their service as research subjects. The amount of the stipend should not be so much that it would overpower the prospective subject’s capability to exercise ‘free power of choice.’ Excessive stipends, as well as the provision without charge to subjects of otherwise costly goods and services can be ethically problematic and may be ‘undue inducements’ to research participation.
One particularly controversial aspect of financial incentives is whether it is appropriate to increase the level of incentive according to the amount of risk. Ideally, all research subjects should receive modest stipends and reimbursement for out-of-pocket expenses. This would work well if there were in place a comprehensive plan to provide no-fault compensation for researchinduced injury. Although such compensation plans have been strongly recommended by multiple national advisory commissions, federal regulations now require only that prospective subjects be informed as to whether there is a plan for compensation for injury and provision of medical treatment for research-induced injury. There is no requirement that either be provided.
Selection of subjects
The selection of subjects for participation in research should be responsive to the demands of scientific merit, subject safety and equitability in the distribution of the burdens and benefits of research. Inclusion and exclusion criteria should be established to assure that the subjects will have the biological, behavioral and social attributes necessary to accurately test the hypothesis of the research and to exclude those individuals especially likely to be harmed by the research. Historically, the concern for equitability was focused primarily on protecting vulnerable persons from bearing more than their fair share of the burdens of research participation. More recently, another primary concern has developed, that there will be no unjust exclusion of groups from the benefits of research participation (10, 11).
Additional Protections for Research Involving Children
Although research involving children as subjects was not the main focus of discussion at this conference, it is evident that bronchoprovocation and bronchosopy are increasingly being employed or considered for younger subjects. This raises several concerns that are mentioned below. This topic will require further discussion by other groups in the future. Meanwhile we offer a general and necessarily limited overview of some of the major issues. Federal regulations for the protection of children as research subjects contain criteria for ethical justification which vary according to the degree of risk and by whether or not the intervention or procedure that presents the risk “holds out the prospect of direct benefit for the individual subject [beneficial procedures]….” Research involving children as subjects requires generally the child’s assent and the permission of the parent(s) or guardian. Research that presents no more than minimal risk may be carried out without additional substantive or procedural protections. Research in which more than minimal risk is presented by a beneficial intervention or procedure is permitted if the risks are minimized and are justified by the anticipated benefit and the relationship of anticipated benefit to the risk is at least as favorable as that presented by available alternatives. Research in which more than minimal risk is presented by a non-beneficial intervention or procedure is permitted only if the level of risk is a minor increase over minimal, the intervention or procedure presents experiences to subjects that are commensurate with actual or expected medical, dental, psychological, social, or educational situations and the research is likely to yield generalizable information of vital importance about the subjects’ disorder or condition. Research that is not otherwise approvable under the first three categories but presents an opportunity to understand, prevent, or alleviate a serious problem affecting the health or welfare of children, may be permitted by the Secretary of Health and Human Services after expert consultation and opportunity for public review. This regulatory framework imposes a significant limit on the discretion of investigators and parents to permit the participation of children in research that entails more than minimal risk but at the same time permits much research of vital importance to the health and well being of children now and in the future.
The key to understanding these categories is the definition of ‘minimal risk’ set forth in the regulations: “Minimal risk means that the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical examinations or tests.” 45 CFR 46.102(i). However, the definition of minimal risk is difficult to apply because it is not clear; there is wide disagreement on the probability and magnitude of risk ordinarily encountered. Consistent with The National Commission’s original intent, a recent Institute of Medicine report recommends that minimal risk be interpreted in relation to “the normal experiences of average, healthy, normal children”(12,13). A ‘minor increase over minimal risk’ is then a “slight increase” over this level of risk (13).
Bronchoscopy presents children with more than a ‘minor increase over minimal risk’ and thus could only be approved by a local IRB for research purposes when the procedure also offered the prospect of direct benefit. The addition of a research objective to an otherwise clinically indicated bronchoscopy may present no more than a ‘minor increase over minimal risk’ depending on the necessary changes in the procedure, such as duration, anesthetic approach and so forth. In this context, the incremental risk to the child by adding the research objective to the clinical procedure becomes the determining factor in assessing the research risks of the bronchoscopy. In some but not all circumstances bronchoprovocation may present only a ‘minor increase over minimal risk’. Factors that must be considered would include subject population (i.e., risk for severe bronchospasm), agent (including dose and duration), ease of reversibility, the setting of the research including monitoring equipment, professional availability, competence to respond rapidly to an untoward event, and so forth. Researchers must have adequate pediatricspecific expertise and experience to perform the research in order to minimize risks (13).
Recently a federal panel reviewed a request to perform bronchoscopy for research purposes on infants with cystic fibrosis. Of the six panelists, four assessed the risks of bronchoscopy as more than a ‘minor increase over minimal risk.’ All six panelists recommended to the Office of Human Research Protection that the research be approved by the Secretary of HHS, pending certain modifications (14). The Secretary of HHS has not yet made a final determination. The process of federal review of research involving children that cannot be approved by a local IRB has come under scrutiny (15). In principle, there should be broad public discussion whenever research involving children moves beyond the standard categories of risks and potential benefits that IRBs are permitted to approve.
Public Discussion of the Ethics of Research
The foregoing discussion of the ethics of research involving bronchoscopy, bronchoprovocation and related procedures is offered as an overview and interpretation of contemporary standards. These standards and their interpretation will require further elaboration and revision particularly as new information becomes available about such matters as the risks of the procedures when they are performed for research purposes by suitably qualified experts and as we develop better consensus about the meaning of such terms as ‘minor increase over minimal risk’. We recommend that in the future, such discussions should include as full members, representatives of other groups such as patients, patient advocates and community members with health care interests. Further, an effort should be made to make such proceedings suitably transparent. Such an approach promises to enhance the protection of the rights and interests of participants in important research endeavors.
Conflict of Interest Statement
R.J.L. is a member of the Bioethics Committee of Eli Lilly Corporation, for this he received $5,000 in 2004, $0 in 2003, $8,194 in 2002 and $4,475 in 2001. R.M.N. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. A.R.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. J.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
Co-leaders: Busse, William W., M.D., University of Wisconsin-Madison, Madison, Wisconsin and Wanner, Adam, M.D., University of Miami School of Medicine, Miami, Florida
Other NIH Attendees: