Reference percentiles for FEV1 and BMI in European children and adults with cystic fibrosis

Boëlle Pierre-Yves, Viviani Laura, Busson Pierre-Francois, Olesen, Ravilly Sophie, Stern Martin, Assael, Barreto Celeste, Drevinek Pavel, Thomas Muriel, Krivec Uros, Mei-Zahav Meir, Vibert Jean-François, Clement Annick, Mehta Anil, Corvol Harriet, On

Description

The clinical course of Cystic Fibrosis (CF) is usually measured using the percent predicted FEV 1 and BMI Z-score referenced against a healthy population, since achieving normality is the ultimate goal of CF care. Referencing against age and sex matched CF peers may provide valuable information for patients and for comparison between CF centers or populations. Here, we used a large database of European CF patients to compute CF specific reference equations for FEV 1 and BMI, derived CF-specific percentile charts and compared these European data to their nearest international equivalents. Methods 34859 FEV 1 and 40947 BMI observations were used to compute European CF specific percentiles. Quantile regression was applied to raw measurements as a function of sex, age and height. Results were compared with the North American equivalent for FEV 1 and with the WHO 2007 normative values for BMI. Results FEV 1 and BMI percentiles illustrated the large variability between CF patients receiving the best current care. The European CF specific percentiles for FEV 1 were significantly different from those in the USA from an earlier era, with higher lung function in Europe. The CF specific percentiles for BMI declined relative to the WHO standard in older children. Lung function and BMI were similar in the two largest contributing European Countries (France and Germany). Conclusion The CF specific percentile approach applied to FEV 1 and BMI allows referencing patients with respect to their peers. These data allow peer to peer and population comparisons in CF patients.

Sujets

• Cystic fibrosis
• Body mass index
• Registry

Informations

• Publié par: biomed
• Publié le: 01 janvier 2012
• Nombre de lectures: 17
• Langue: English

Extrait

Boëlle et al. Orphanet Journal of Rare Diseases 2012, 7:64
http://www.ojrd.com/content/7/1/64
RESEARCH Open Access
Reference percentiles for FEV and BMI in1
European children and adults with cystic fibrosis
1,2,3 4 2 5 6 7Pierre-Yves Boëlle , Laura Viviani , Pierre-Francois Busson , Hanne V Olesen , Sophie Ravilly , Martin Stern ,
8 9 10 11 12 13Baroukh M Assael , Celeste Barreto , Pavel Drevinek , Muriel Thomas , Uros Krivec , Meir Mei-Zahav ,
1,2,3 1,2,3 14 1,2,3*Jean-François Vibert , Annick Clement , Anil Mehta and Harriet Corvol on behalf of the French CF Modi-
fier Gene Study Investigators and the European CF Registry Working Group
Abstract
Background: The clinical course of Cystic Fibrosis (CF) is usually measured using the percent predicted FEV and1
BMI Z-score referenced against a healthy population, since achieving normality is the ultimate goal of CF care.
Referencing against age and sex matched CF peers may provide valuable information for patients and for
comparison between CF centers or populations. Here, we used a large database of European CF patients to
compute CF specific reference equations for FEV and BMI, derived CF-specific percentile charts and compared1
these European data to their nearest international equivalents.
Methods: 34859 FEV and 40947 BMI observations were used to compute European CF specific percentiles.1
Quantile regression was applied to raw measurements as a function of sex, age and height. Results were compared
with the North American equivalent for FEV and with the WHO 2007 normative values for BMI.1
Results: FEV and BMI percentiles illustrated the large variability between CF patients receiving the best current1
care. The European CF specific percentiles for FEV were significantly different from those in the USA from an earlier1
era, with higher lung function in Europe. The CF specific percentiles for BMI declined relative to the WHO standard
in older children. Lung function and BMI were similar in the two largest contributing European Countries (France
and Germany).
Conclusion: The CF specific percentile approach applied to FEV and BMI allows referencing patients with respect1
to their peers. These data allow peer to peer and population comparisons in CF patients.
Keywords: Cystic fibrosis, Forced expiratory volume in one second, Body mass index, Registry
Background percent predicted FEV referenced against a healthy1
The outcome of cystic fibrosis (CF) patients has population [4,5]. This particular choice reflects a
improved in recent decades, with mortality less than 5 widely held opinion that achieving normality remains
percent during the first 10 years of life in patients trea- the ultimate goal of CF care. Irrespective of whether or
ted with current multidisciplinary care [1]. However, not this aspiration is achievable, a limitation of the
disease severity remains variable among children, ado- current ‘reference against the normal range’ approach
lescents and adults [2,3]. In CF, severity is principally is that it does not provide a ranking of an individual
assessed by the decline in lung function because lung patient’s status relative to age and sex-matched CF
disease still remains the most common cause of death. peers. Similar considerations apply to nutrition, which
Lung function in CF is almost universally calculated as must be adequate to maintain lung function [6], or sur-
rogate markers of disease severity calculated from
semi-quantitative screening scales such as the Chrispin
* Correspondence: [e-mail]
1 Norman Score that measures lung damage from chestAP-HP, Hôpital Trousseau - Pediatric Pulmonary Department, Hopital St
Antoine – Public Health Department, Paris, France radiography [7].
2
INSERM, U938, INSERM U707, Paris, France
Full list of author information is available at the end of the article
© 2012 Boelle et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.Boëlle et al. Orphanet Journal of Rare Diseases 2012, 7:64 Page 2 of 8
http://www.ojrd.com/content/7/1/64
There have been attempts to overcome such limitations. The MUCONAT project collects data on prevalent
Kulich and coworkers converted the absolute FEV into and incident CF cases. It was approved by the French1
percentiles calculated from a registry of lung function ethical committee (CPP n°2004/15) and the information
values in North American CF patients [8]. In a related collection was approved by the CNIL (n°04.404). As for
manner, McCormick and colleagues converted a chest x- the ECFSPR, the MUCONAT data were collected on a
ray severity score in childhood into population based per- strictly anonymous basis. The project is based on the
centiles[9]. Such self-referencing approaches, despitetheir participation of 38 out of the 49 French CF centers. Pro-
calculation from cross sectional data, provide a reference spective data collection started in 2007 for all prevalent
base that allows the longitudinal tracking of CF disease and incident cases.
outcome and informs on the relative position of a given The following information was extracted from both
patientagainst his or her peers. databases: FEV measurements (in L), country of resi-1
As of now, the CF FEV specific percentiles have of dence, CFTR genotype, sex, height, weight and age, BMI.1
necessity been obtained from US CF patients. It remains The ECFSPR covered the period 2004 to 2007, with one
to be established whether these data are appropriate for observation per year and per patient. Data for years
European patients with CF. Indeed, there are many dif- 2008 to 2010 were obtained for French patients from the
ferences not only in how CF care is organized (for ex- MUCONAT database, so that there was no overlap with
ample, coverage of neonatal screening, timely referral to the ECFSPR source. The same patient may have contrib-
reference centers, variation in standards of care) but also uted data over several years; however the current
in the environment between EU nations and across the ECFSPR did not allow reliable data linkage throughout
Atlantic. Furthermore Kulich analyzed data from 15 years the years for all the participating countries, so longitu-
ago (1994 to 2001) that would not account for recent dinal aspects were not taken into account. Measurements
progress in CF care. To mitigate against such issues, we in patients after lung transplantation were removed for all
combined data from the European Cystic Fibrosis Soci- the analyses.
ety Patient Registry (ECFSPR) and data from the current
French CF Modifier Gene Study, to obtain CF reference Statistical analysis
percentile equations for lung function and body mass The FEV percent predicted (FEV pp) were esti-1 1
index that would apply to current European CF patients mated according to the Knudson equations [5], and
and would additionally provide the ranking of an indi- the BMI z-scores were computed using the WHO
vidual patient’s status relative to age and sex-matched 2007 standards [13].
CF peers. As a second objective, we examined whether In data contributed to the ECFSPR, some countries
FEV and BMI in CF patients differed between European reported the “best” annual FEV measurement, while1 1
countries and across two continents, Europe and North others, including the 2 largest contributing countries
America (USA). (France and Germany), reported an unselected meas-
urement. Using the best measurement for computation
leads to overestimation in the CF specific percentile
Methods curves and limit its use to assess patients from unse-
Patients lected measurement. To limit such bias, we transformed
We used the multinational ECFSPR as primary source of “best” FEV values before calculation (see Additional file1
data and additional observations from the French CF 1). In short, the correction was computed as follows:
Modifier Gene Study (MUCONAT). The former was set using the French data, where a systematic longitudinal
up to “measure, survey and compare CF aspects and collection of all FEV values is carried out, we deter-1
treatments” in European countries [3,10]. Fourteen mined, by sex and age, the average difference between
countries contributed data to the present study (Austria, the “best” annual FEV and an unselected value of the1
Belgium, Bulgaria, Czech Republic, Denmark, France, same year. A corrected FEV measurement was then1
Germany, Greece, Israel, Italy, the Netherlands, Portugal, obtained by subtracting this value from the reported
Slovenia, Sweden). The coverage of the study relative to FEV value in countries reporting best values. As a sen-1
the overall CF population in a given country was esti- sitivity analysis, we also analyzed the data without cor-
mated using genetic prevalence estimates reported by rection. This correction was not required for BMI, as
Farrell et al. [11], by Efrati et al. [12] for Israel, and by the reported data was not selected.
the French CF national registry for France. Patients’con- Quantile regression was used to estimate CF-specific
sent was obtained from every participating country and reference equations for FEV and BMI. The q-quantile1
all the registry protocols