Long-Term Nonpulmonary Complications in Adults with Cystic Fibrosis

Cystic fibrosis (CF), the most common lethal inherited disorder in North America, Europe, and Australia, is a multisystem disorder affecting epithelial tissues, predominantly the airways, liver, pancreas, and gastrointestinal system. Dramatic improvement in outcomes has resulted from quality multidisciplinary care. Over the last 50 years, the median survival of people with cystic fibrosis (pwCF) has progressively increased, demonstrating an increasing average/median age together with increasing numbers of adults in the CF registries (1–4).

After the development of modulators of the cystic fibrosis transmembrane conductance regulator (CFTR) (5, 6), survival has increased further, with projections now very close to those of the general population. But with the increased survival comes increasing recognition of both long-term complications of CF and long-term side effects from the different therapies. Finally, as pwCF age, the development of age-related pathologies will gradually increase in importance.

Thus, it is important to examine CF registries to determine which comorbidities and complications are becoming evident, allowing planning for the future. In this issue of AnnalsATS, Arya and colleagues (pp. 1874–1880) describe a Canadian CF registry study of 1,643 pwCF compared with Ontario Health Insurance Plan data for those over the age of 19 years (n = 16,435,300) (7). Of the pwCF, 1,435 were not transplant recipients, and 208 were post–lung transplant.

In the first analysis, the pwCF nontransplant and post-transplant cohorts were compared. As expected, the post-transplant pwCF were older, had a lower median forced expiratory volume in 1 second percent predicted, and had a higher rate of cystic fibrosis–related diabetes (CFRD) with an increased incidence of chronic kidney disease (CKD), cardiovascular disease (CVD), and cancer. Many of these changes reflect the effect of the transplant process together with prolonged immunosuppression. This updates previous studies with similar findings.

Comparison of the pwCF (nontransplant) cohort with the general population in Ontario showed that the median age at study entry in the CF cohort was 19 years (interquartile range, 19–28) compared with 34 years in the non-CF population. Clearly, the CF group is not normally distributed; indeed, it is “skewed” if the cohort includes only those over 19, with a median age of entry of 19 years. Even if the median age is actually 22 or 23 years, as outlined in Table E4C for the different subgroup analysis, it is still skewed to the younger ages compared with the general population of Ontario. So, the development of any CVD in a young population becomes more significant.

After adjustment for age and sex, pwCF had almost three times the risk of kidney stones, two times the risk of CKD, and almost two times the risk of cancer compared with the general population, as shown in Table 2 (7). This occurred despite a large difference in median age (Table E5). Importantly, 57 (4%) of the 1,435 of the pwCF (nontransplant group) had a diagnosis of cancer during the study.

But the important new data in this paper is the increased risk of CVD. The reader should note that although the initial definition of CVD was listed as “any diagnosis of myocardial infarction, stroke, or congestive heart failure, any emergency department (ED) visit with a diagnosis of cardiac or vascular disease, or cardiac procedure with catheterization or coronary artery bypass graft,” the Ontario Health Insurance Plan also includes “hypertensive diseases, valvular diseases, pulmonary vascular disease, disease of the venous system and capillary system, arterial disease, endocarditis, pericarditis, myocarditis, and cardiac arrest.”

Using the broader definition, 15% (216 of 1,435) of the CF cohort was listed as having a CVD outcome, giving a relative risk in pwCF compared with the non-CF population of 2.94 (95% confidence interval, 2.57–3.35) (7). The three most common CVD outcomes in pwCF were venous thrombosis (27%), congestive heart failure (23%), and cardiac arrhythmias (12%). Hypertensive disease accounted for 8% of CVD outcomes, and ischemic heart disease contributed to 3% of CVD outcomes in the pwCF (nontransplant) cohort. In contrast, in the general population (non-CF nontransplant), the CVD outcomes were hypertensive diseases (33%) followed by ischemic heart diseases (21%) followed by cardiac arrhythmias (16%) (7).

Using the initial narrower definition of CVD, there was still a relative risk of 2.36 (95% confidence interval, 1.97–2.83) comparing CF with non-CF. This finding confirms a previous study from the United Kingdom in which researchers found increased risks of major adverse cardiac events in the CF cohort (8). Recent studies have also started to show some subtle abnormalities on echocardiograms in pwCF compared with the general population (9). Because the CFTR modulators are associated with the development of hypertension (10) and weight gain (5), care will be needed to examine for new CVD.

The strengths of this study include the relatively large groups and the comparison with a health insurance program to provide good data concerning the general population. The Canadian CF registry and the health administrative database were successfully linked with a 94% linkage rate.

Future studies will hopefully separate the relative impact of CFRD on the outcomes listed above. It is well recognized that CFRD is a risk factor for CVD and CKD (8), so it will be interesting to know what role CFRD plays in these outcomes. Over the next decade, as the population ages, pwCF will need to watch carefully for the development of nonpulmonary complications.