Pacing the Marathon: Rate of Decline of Pulmonary Function in Cystic Fibrosis
Article Outline
Abbreviations: CF, Cystic fibrosis, ESCF, Epidemiologic Study of CF, FDA, Food and Drug Administration
The course of lung disease in cystic fibrosis (CF) has much in common with running a marathon, where a steady pace is punctuated by occasional sprints. In CF, the sprints are the pulmonary exacerbations, whose number and severity influence the ultimate outcome by quickening the overall race. However, the day-to-day success or failure of holding the disease in check is similar to the steady pace of the marathon runner, and, for most patients, determines their ultimate fate. A child born today with CF can expect to live nearly four decades,1 and the duration and quality of that life depends heavily on the rate of decline of pulmonary function.2 In this issue of The Journal, Konstan et al3 define the influences on the rate of decline of pulmonary function in children with CF, using a very large encounter-based data set. Understanding of these influences is critical if we are to alter this inexorable decline.
See related article, p 134
The data set is critical to the importance of the results. Although many of the influences on the rate of decline identified in this article have been observed before, previous studies have utilized populations that are smaller, or followed at a single center, or accumulated over a period of years, or limited in scope. The use of the Epidemiologic Study of CF (ESCF) data set, which encompasses nearly all the patients in nearly all the CF Centers in the United States and Canada, makes these results generalizable, and the topic makes them important. This study was conducted over a limited time period, and careful consideration was given to including only patients with interpretable data (eg, sufficient number of pulmonary function tests, baseline parameters determined at least 1 year before the observation period to avoid regression to the mean). Like others, Konstan et al3 found that pancreatic enzyme use, infection with Pseudomonas, nutritional status, baseline pulmonary function, and age strongly and independently influence the rate of decline of pulmonary function. To these observations, they add the presence of sputum production and crackles as poor prognostic signs and the presence of wheezing (though not necessarily asthma) as a positive factor. Surprisingly, given the near-universal reports of negative influence of female sex on survival in CF, rate of decline of FEV1 was less in boys only in the 6- to 8-year-olds, but boys had greater rate of decline among the 9- to 12-year-olds and the 13- to 17-year-olds. This is unlikely to result from censoring of the data due to deaths among the girls because the studies were performed during a time period in which death at less than 8 years of age is rare in CF. Early in life, in other airway diseases, boys tend to do worse, presumably because their airway size relative to lung or body size is reduced relative to girls.4, 5, 6 The CF female disadvantage has always been a mystery because it occurs at such a young age, well before substantial differences in sex hormones should affect the outcome. It is possible that the statistical analysis conducted in this paper allowed separation of the influence of sex on other variables with independent influence on outcome. For example, girls tend to become infected with pseudomonas at an earlier age than boys.7 It is possible that this influence of sex was subsumed into the influence of pseudomonas infection on rate of decline. Elevated liver tests have modest negative impact on rate of decline from age 9 to 17 years; this finding should encourage more investigation into the causes and possible interventions for liver disease in CF.
The finding that the number of exacerbations treated with intravenous antibiotics is associated with increased rate of decline is also an important addition to the CF literature. Pulmonary exacerbation has become an important outcome measure for approval of drugs for CF therapy by the Food and Drug Administration (FDA) because it meets the definition for an efficacy measure that is “clinically meaningful.” Linking pulmonary exacerbations to FEV1 decline may take this outcome measure a step further, with the suggestion that perhaps the number or frequency of exacerbations may be a surrogate measure for survival (given that FEV1 decline is a recognized surrogate for survival). Studies to investigate this hypothesis directly are probably warranted, since it is important to use the most meaningful outcome measures that are feasible in clinical investigations.
Konstan et al3 provide an algorithm for predicting rate of decline for an individual patient, which should be considered a tool for identifying patients at high risk rather than a definitive prognostication. Patients at high risk may need to be seen more frequently, treated more aggressively and with less hesitation, and their pulmonary function followed more assiduously than those at lower risk. In children, clinicians often recognize those with crackles, poor nutritional status, and pseudomonas infection as being at risk (so this paper brings old news on those accounts) but may be lulled into a false sense of security by a robust patient with excellent pulmonary function. Because an important goal of CF therapy is to retain a high level of function for as long as possible, such patients may be particularly critical to follow closely. In this context, it would have been useful to know whether other measures of pulmonary function said to be more sensitive to the presence of early disease, such as the ratio of residual volume to total lung capacity, or the instantaneous flow rates at low lung volumes, added to the predictive value of the algorithm, but the report is silent on this account.
The research implications should not be overlooked. The factors identified here as contributing to the rate of decline should be considered in the design of clinical trials and the evaluation of the equivalence of comparison groups, even if the primary outcome measure is not the rate of decline. Despite the obvious importance of the rate of decline of pulmonary function to the ultimate outcome in patients with CF, few clinical trials target this parameter as an outcome measure. This is most likely because, to determine whether the slow rate of decline in CF has actually been altered by intervention, either inordinate numbers of patients or lengthy studies are required.8 Such studies sap time, patient adherence, patient availability for other studies, and (if the drug is novel or expensive) drug supplies. For some uses, however, this may be the most important outcome measure. Some drugs might not be expected to confer short term benefits but to have their greatest effect long term. Anti-inflammatory drugs, probably the most underutilized category of drugs in CF today, fall into this category.9, 10 The results in this report will help to design the proper studies to evaluate such drugs and minimize patient numbers and study duration.
Even for drugs that obtain FDA approval or labeling for CF based on their short-term benefits, it is probably important to study their effect on rate of decline. So-called Phase IV studies should be considered for drugs that achieve approval based on short-term improvements in FEV1 or the reduction in exacerbations. Such drugs, of proven efficacy in other measures, might not improve, and might actually worsen, the rate of decline. In this case, strategies of administration might be devised that take advantage of their favorable properties while avoiding those that are unfavorable. This report will assist in such evaluations.
Konstan et al also illustrate the potential importance of clinical databases for research. As we come into the era of electronic medical records, data warehouses that permit clinicians full access to long-term patient laboratory records, and sophisticated computer programs that allow us to relate clinical outcomes to interventions, we should be mindful of the power of such data. We should ensure that we are able both to vouch for its quality and to use it with patient and parent knowledge and assent. Konstan et al, others from the ESCF data base, and still others from the CF Foundation’s Patient Registry are publications from databases populated with explicit patient consent and with the explicit intent of clinical and epidemiologic research. They illustrate for one disease with a modest number of patients nationwide the power of national and North American data sets in understanding the disease and its determinants. They also provide the foundation for design of clinical trials. The broader implications of these studies for less common diseases are obvious, and all but demand the creation of similar data sets if clinical progress is to be gauged. It should also be clear that collecting information on diseases that are more common will have substantial benefits as well. Universal data are powerful, and the lack of selection bias is to be valued greatly. Care must be taken to ensure a high quality of data, clear standards for clinical diagnosis, validated phenotype data, and rigorous attention to detail in the collection and analysis of data. The ESCF data collection and the analysis of these data by Konstan et al3 have set a high benchmark.
References
- Cystic Fibrosis Foundation Patient Registry. 2005 Annual Data Report. Bethesda, Md.
- . Jointly modeling the relationship between survival and pulmonary function in cystic fibrosis patients. Stat Med. 2002;21:1271–1287
- Risk factors for rate of decline in FEV1 in children and adolescents with cystic fibrosis. J Pediatr. 2007;151:134–139
- . Changes in lung, airway, and chest wall function in boys and girls between 8 and 12 yr. J Appl Physiol. 1984;57:304–308
- . Longitudinal analysis of lung function growth in healthy children and adolescents. J Appl Physiol. 1991;70:770–777
- . The relationship between airway size and lung size. Br J Dis Chest. 1985;79:183–188
- . Gender differences in cystic fibrosis: Pseudomonas aeruginosa infection. J Clin Epidemiol. 1995;48:1041–1049
- . Identifying treatments that halt progression of pulmonary disease in cystic fibrosis. Pediatr Res. 1997;41:161–165
- . Effect of high-dose ibuprofen in patients with cystic fibrosis. N Engl J Med. 1995;332:848–854
- . Pharmacological approach for the discovery and development of new anti-inflammatory agents for the treatment of cystic fibrosis. Adv Drug Deliv Rev. 2002;54:1409–1423
PII: S0022-3476(07)00327-7
doi:10.1016/j.jpeds.2007.03.060
© 2007 Mosby, Inc. All rights reserved.
Refers to article:
- Risk Factors For Rate of Decline in Forced Expiratory Volume in One Second in Children and Adolescents with Cystic Fibrosis , 26 June 2007
