Elsevier

The Journal of Pediatrics

Volume 131, Issue 6, December 1997, Pages 809-814
The Journal of Pediatrics

Longitudinal analysis of pulmonary function decline in patients with cystic fibrosis,☆☆,

From Division of Gastroenterology and Nutrition and Chest Division, The Hospital for Sick Children, Toronto, Ontario, Canada; Department of Preventive Medicine and Biostatistics and Department of Pediatrics, University of Toronto; and Department of Biostatistics and Department of Medicine, University of North Carolina at Chapel Hill.
https://doi.org/10.1016/S0022-3476(97)70025-8Get rights and content

Abstract

Background: Chronic progressive lung disease is the most prominent cause of morbidity and death in patients with cystic fibrosis (CF), but severity of lung disease and rate of lung function decline are widely variable. Accurate estimates of decline have been difficult to define and compare because the timing of measurements and duration of follow-up differ in various patient groups.

Patients: Three hundred sixty-six patients with CF, born from 1960 to 1974, were selected from a CF database birth cohort if they had two or more measurements of pulmonary function, at least one of which was performed before the age of 10 years.

Methods: Mixed model regression analysis provided estimates of the average rate of decline of spirometry measurements in subgroups on the basis of survival age, sex, pancreatic status, and genotype.

Results: Patients who died before the age of 15 years had significantly poorer pulmonary function when first tested and a more rapid decline in pulmonary function thereafter than patients who survived beyond the age of 15 years. In the latter, functional levels at the age of 5 years were normal, but average rates of decline were significantly related to survival age. Female patients had significantly steeper decline than male patients, and those with pancreatic insufficiency had much steeper decline than those with pancreatic sufficiency. In the subset of 197 who survived to 1990 and were subsequently genotyped, rate of decline was greater in those homozygous for the ΔF508 mutation, compared with those who were heterozygous for ΔF508 or those who had two other mutations.

Discussion: All but the most severely affected patients, who died before age 15, appear to have had normal pulmonary function when first tested in early childhood. Pancreatic sufficiency, male gender, and some non-ΔF508 mutations are associated with a slower rate of pulmonary function decline. Mixed model analysis is a valuable tool for describing and comparing pulmonary function decline in groups of patients with CF. (J Pediatr 1997;131:809-14)

Section snippets

Patients

The study cohort was retrospectively assembled from the CF database at the Hospital for Sick Children in Toronto.9 All patients who were born in the years 1960 to 1974, who had at least two recorded pulmonary function tests, and whose first test was performed before the age of 10 years were included. The HSC clinic began in 1960, and routine pulmonary function testing began in 1967. Nineteen seventy-four was selected as the cutoff birth year to allow at least 10 years of follow-up from the age

Results

There were 553 patients identified in the CF database who were born between 1960 and 1974, of whom 366 satisfied the study criteria. Forty-six patients died before the age of 10 years and without pulmonary function follow-up. Another 40 patients in the cohort were lacking pulmonary function records, either because they transferred out of the clinic before pulmonary function testing was initiated or because HSC was not the site of their regular follow-up. The remaining 101 patients did not have

Discussion

This analysis has quantified the differences in rate of pulmonary function decline in patients who died at a young age compared with those who died later. All but the most severely affected patients, the relatively small proportion who died before age 15, appeared to have normal pulmonary function when first tested in early childhood. This suggests that in future studies of young patients with normal pulmonary function, it is highly desirable to consider other genetic factors, such as

Acknowledgements

We thank Ken Oliver, MD, for his review and useful comments on the manuscript.

References (22)

  • NM Laird et al.

    Random-effects models for longitudinal data

    Biometrics

    (1982)
  • Cited by (314)

    View all citing articles on Scopus

    Supported by research grants from the Canadian Cystic Fibrosis Foundation and the Medical Research Council of Canada.

    ☆☆

    Reprint requests: Mary Corey, PhD, Division of Gastroenterology and Nutrition, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario, M5G 1X8 Canada.

    9/21/83366

    View full text