Cystic fibrosis and survival in patients with advanced lung diseaseBMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d726 (Published 07 March 2011) Cite this as: BMJ 2011;342:d726
- Elliott C Dasenbrook, assistant professor of medicine and pediatrics
- 1LeRoy W Matthews Cystic Fibrosis Center, University Hospitals Case Medical Center, Case Western Reserve University School of Medicine, Cleveland, OH 44106, USA
When Dorothy Andersen provided the first comprehensive description of cystic fibrosis in 1938,1 survival was often measured in days and months. However, the introduction of penicillin resulted in children recovering from previously fatal infective respiratory exacerbations.2 Subsequently, antibiotics given in combination with aggressive treatment of malnutrition resulted in some people with the condition living into adulthood.3 Today, a myriad therapeutic strategies are directed at the infectious, inflammatory, and mucociliary defects in cystic fibrosis.4 Combined with other non-pharmacological advances, children born today with cystic fibrosis are expected to live into their 50s⇓.5
In the linked study (doi:10.1136/bmj.d1008), George and colleagues report on survival in a subgroup of adult patients with cystic fibrosis and advanced lung disease at the Royal Brompton Hospital, London, over 17 years (1990-2007).6 Median survival improved by four years in those entering the cohort between 2002 and 2003 compared with those who entered between 1990 and 1991. Their data suggest that the widespread introduction of recombinant human deoxyribonuclease (rhDNase) at the centre in 1994 may be responsible for some of this improvement. If true, it would be the first treatment in the modern management era to be associated with an improvement in survival.
The authors found a 41% reduction (95% confidence interval 21% to 56%) in the adjusted hazard of death in patients with cystic fibrosis and advanced lung disease (defined as forced expiratory volume in one second (FEV1) <30% predicted) prescribed rhDNase compared with those who were not prescribed rhDNase. Furthermore, increased use of rhDNase in 1994-5 was associated with a corresponding drop in the hazard ratio of death. However, can a strong causal relation between the use of rhDNase and survival be inferred from this observational study?
An association between rhDNase and a decreased hazard of death is biologically plausible. Treatment with rhDNase in patients with cystic fibrosis is associated with decreased pulmonary exacerbations,7 and a recent systematic review found that rhDNase was well tolerated and improved lung function.8 Pulmonary exacerbations and lung function are important predictors of survival.9 Because patients with advanced lung disease have multiple exacerbations a year, long term administration of rhDNase could improve survival in patients with cystic fibrosis and advanced lung disease.
It is important to assess the study methods to determine whether the evidence is strong enough to alter current rhDNase practice patterns. The authors discuss many of the pertinent strengths and limitations in the conclusions, but several points deserve special consideration. Secondary outcomes examining factors associated with survival were decided a priori. This distinction is often not reported, but it is important because it decreases the probability that the association is spurious or overestimated.10 Missing data and loss to follow-up are common in observational research, but this is not a problem in this study, so the risk of bias from informative censoring is low.10 These methodological strengths lend credence to the conclusions.
However, one factor with the main Cox regression analysis is that, unlike another cystic fibrosis survival analysis,11 the researchers did not adjust for year of entry into the cohort. Survival times would vary according to the year of entry because of differences in therapeutic options and general cystic fibrosis care. Therefore, the year of entry should be included in any survival analysis of a chronic disease that spans decades. For instance, when the authors divide the cohort into two time periods (1990-1995 v 1996-2003) and add this variable to the model in a sensitivity analysis, the association between rhDNase and survival is significantly attenuated. This suggests that at least part of the beneficial effect of rhDNase might be the result of other unmeasured aspects of cystic fibrosis care that changed or became available during the course of the study.
A four year improvement in survival among patients with cystic fibrosis and advanced lung disease is an important finding for clinicians, patients, and care givers. The rhDNase results should be compared with those from other multicentre patient registries, and if they are replicated this will strengthen the argument for causality. In the meantime, the current study adds to the evidence supporting the use of rhDNase in patients with cystic fibrosis and advanced disease and is strongly recommended in treatment guidelines for cystic fibrosis.12 Finally, the encouraging survival rates can be used as a source of motivation for patients to increase their adherence with the entire chronic treatment regimen as they await the next advance in cystic fibrosis care.
Although survival in patients with cystic fibrosis has improved greatly, we need to do better. At best, rhDNase slows the progression of lung disease. For the next dramatic leap in survival, interventions must stop the decline in lung function associated with the disease. Small molecule treatments or gene therapy (or both) that target the underlying defect show promise. The ultimate goal is for cystic fibrosis to be transformed into a lifelong condition that patients must live with but do not die from.
Cite this as: BMJ 2011;342:d726
Competing interests: All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Commissioned; not externally peer reviewed.