Fertility and pregnancy outcomes in men and women with cystic fibrosis in the United Kingdom

Jenny M. Boyd1, Anil Mehta2 and Deirdre J. Murphy2,3

1 Women's and Child Health, Ninewells Hospital and Medical School, Dundee and 2 Division of Maternal and Child Health Sciences, University of Dundee, Ninewells Hospital and Medical School, Dundee DD1 9SY, United Kingdom

3 To whom correspondence should be addressed at: Division of Maternal and Child Health Sciences, Ninewells Hospital and Medical School, Dundee DD1 9SY, UK. Tel: 01382 632147; Fax: 01382 633847; Email: d.j.murphy{at}dundee.ac.uk


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
BACKGROUND: The survival of patients with cystic fibrosis (CF) now occurs into the fourth decade of life. Our aim was to evaluate the fertility and pregnancy outcomes of men and women with cystic fibrosis within a large British cohort. METHODS: A population based cohort study of 4659 males and females registered with the UK Cystic Fibrosis Database in 2001. RESULTS: Poor respiratory health (forced expiratory volume in 1 s FEV1 <50%) and CF-related diabetes increased significantly post-puberty compared with childhood. Few individuals with CF sought fertility treatment (1% men, 0.5% women) or achieved pregnancies (1.3% of partners of men, 5.7% women). However, the majority of pregnancies had a good outcome (67% live term birth among male partners, 74% among women). Women who achieved a pregnancy were less likely to suffer from poor respiratory health (FEV1 <50%), age adjusted Odds Ratio 0.6 (95% CI 0.3–0.9), less likely to be homozygous for the DF508 genotype OR 0.5 (95% CI 0.3–0.9) and had a significantly earlier median age at diagnosis (0 years vs 2 years, P=0.001). CONCLUSIONS: Increasing numbers of males and females with cystic fibrosis are reaching reproductive age but currently very few have a child. Optimal adult health should improve the reproductive prognosis for both men and women.

Key words: cohort study/cystic fibrosis/fertility/pregnancy outcomes


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
Cystic fibrosis (CF) is the commonest life limiting autosomal recessive disorder arising from many different abnormalities in the cystic fibrosis transmembrane conductance regulator protein (Kerem et al., 1989Go). This leads to thickened and inspissated secretions throughout the respiratory and alimentary tracts. In adults its main manifestations are due to undernutrition that results from pancreatic insufficiency and malabsorption. Death occurs most often from respiratory failure resulting from chronic infection, bronchiectasis and progressive lung destruction (Edenborough, 2001Go). The median age at death from cystic fibrosis for those individuals born in the 1980s is currently estimated to be around 22 years of age (McCormick et al., 2002Go). Conversely, in recent years the survival of cystic fibrosis patients has improved markedly with survival into the fourth decade becoming more commonplace (Ratjen and Doring, 2003Go). With almost 50% of patients now in the adult population, the issues of fertility and pregnancy have been raised and warrant careful evaluation for both male and female CF sufferers.

It is known that menarchal age is directly related to health status. This includes height, weight and the percentage of body fat exceeding 18% as well as general health. Once established, most girls with CF will exhibit normal or near-normal menstrual cycles (Edenborough, 2001Go). The most striking feature among girls with menstrual problems was that those who were amenorrhoeic had forced expiratory volume in 1 s (FEV1) <55% predicted. Recent studies have shown that there is no difference in the numbers of CF females who are in relationships, age of coitarche or prevalence of intercourse compared to age matched non-CF controls. Sexual adaptation in CF sufferers is normal (Edenborough, 2001Go). In terms of fertility, older studies have quoted poor figures, with fertility at <20% among female CF sufferers (Rodgers et al., 2000Go). There appears to be little evidence from more recent data to suggest that healthy women with CF have fertility problems apart from differences in the quality of cervical mucus (Kopito et al., 1973Go). It is widely known that 98% of males with CF are infertile due to absent or atretic vas deferentia, and dilated or absent seminal vesicles. However, with the advent of assisted conception techniques such as microscopic epididymal sperm injection (MESA) and intracytoplasmic sperm injection (ICSI) many men are able to father children, and therefore factors such as general health must be taken into account (Edenborough, 2002Go).

Health status is crucial. In terms of respiratory disease, sufferers of CF develop accelerated obstructive patterns of airways disease with poor peak flow, reduced FEV1 and decreased oxygen saturation. They have superimposed infection with organisms such as Pseudomonas aeruginosa, Staphylococcus aureus, the Burkholderia complex, and the fungus Aspergillus(Mehta et al., 2004Go). Insulin dependant diabetes and on-going prophylactic therapy with antibiotics and corticosteroids become increasingly prevalent. These health complications and therapeutic interventions have important implications for planning and achieving pregnancies with a good maternal and perinatal outcome. The aim of this study was to explore the fertility and pregnancy outcomes of men and women with cystic fibrosis within a large British cohort (McCormick et al., 2002Go; Mehta et al., 2004Go) and to relate their reproductive outcomes to post-pubertal health status.


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
A cross-sectional population based cohort study was performed reporting data on health status, fertility and pregnancy outcomes of 4659 males and females registered in the United Kingdom Cystic Fibrosis (UKCF) Database in 2001. The UKCF Database was established at the Tayside Institute of Child Health, University of Dundee, Scotland in 1995. A range of information is collected from 56 paediatric and adult specialist cystic fibrosis clinics around the UK producing patient clinical data, which are updated at least annually. Data extracted from this resource included basic demographic data, all indicators of fertility in the index case or their partner, all pregnancies and pregnancy outcomes. In addition, details about the social impact of CF, glucose tolerance, Shwachmann scores, marital status and employment status were collected. The Shwachmann score is made up of information from four sections; general activity, physical examination, nutrition and chest X-ray. Each section is awarded up to 25 points and the total potential score is 100. It offers a guide to the severity of CF: severe (0–40) moderate (41–55) mild (56–70) and excellent (86–100).

We examined all indicators of proven fertility, and general and specific markers of health and well-being. Information was not available about the use of contraception or the length of time trying for a pregnancy. Proven fertility was taken as being a positive pregnancy test. We considered an FEV1 of <50% predicted to represent poor respiratory health. We compared the characteristics and health status of pre- and post-pubertal males and females within the cohort. For the purpose of this analysis, pre-pubertal individuals were chosen as the reference category. The definition of puberty was the voice breaking in males and menarche in females. We also included data about fertility treatment, individual pregnancies and pregnancy outcomes in those of reproductive age. We then examined the health status of both men and women in relation to reproductive outcomes and compared our data with other recently published studies (Edenborough et al., 2000Go; Gilljam et al., 2000Go; Gillet et al., 2002Go; Odergaard et al., 2002aGo,bGo). Comparisons between groups were presented as relative risk ratios (RR) and 95% confidence intervals (CI). Age adjusted comparisons for post-pubertal individuals were estimated by logistic regression and presented as odds ratios (OR) and 95% CI. Statistical significance was set at a P-value <0.05. The data was analysed using SPSS (v 11.5) software.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
The cohort contained 2518 males (54%) and 2141 females (46%). The current reproductive population, i.e. the post-pubertal population, comprised 1355 males (54%) and 1148 females (46%) (Table I). The predominant genotype contained DF508 on at least 1 chromosome as would be expected from other studies (McCormick et al., 2002Go; McKone et al., 2003Go), and >90% of the study population had defined CF mutations. Poor respiratory health (FEV1 <50%) and CF-related diabetes were significantly more prevalent post-puberty; RR 8.58 (95% CI 6.45–11.4) and 10.2 (95% CI 6.43–16.20) respectively for men, and RR 6.60 (95% CI 5.11–8.52) and 8.49 (95% CI 5.52–13.10) for women. Height and weight centiles appeared to be determined in childhood, with CF sufferers demonstrating short stature and low body weight (<10th centile) at more than twice the expected rate both pre- and post-puberty. There was a significant increase in the prevalence of low bodyweight among post-pubertal males but this was not apparent among women. Body mass index was not a useful parameter, as CF individuals were frequently of low weight and height, giving normal values. Of post-pubertal females with CF, 24% had weight below the 10th centile and would therefore be at risk of infertility, pre-term birth and delivery complications. Relatively small numbers fell into the category of moderate to severe disease based on the Shwachmann score, and the difference between pre- and post-pubertal females was not significant.


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics and health status of pre-pubertal and post-pubertal males and females with cystic fibrosis in the UK CF database

 
Complete data were available for 1347 post-pubertal males and 1143 females. Few individuals with CF sought fertility treatment (1% of men, 0.5% of women) or achieved pregnancies (1.3% of partners of men, 5.7% of women) (Table II). The CF status of partners was not recorded on the database. The total number of pregnancies recorded in the UKCF database in 2001 was 105. Some individuals had more than one pregnancy. One woman had a pregnancy through a surrogacy programme and another woman adopted a baby. There was no specific information about the use of donor sperm. The majority of pregnancies achieved had a good outcome; with a 67% live term birth rate among male partners, and 74% among women.


View this table:
[in this window]
[in a new window]
 
Table II. Fertility treatment, pregnancy and pregnancy outcomes among men (partners) and women of reproductive age with cystic fibrosis

 
Both men and women with pregnancies were older at the time of the study, which has important implications for rates of health complications (Table III). Women who had achieved a pregnancy were less likely to be homozygous for DF508, age adjusted OR 0.48 (95% CI 0.26–0.89) and had a significantly earlier median age at diagnosis, (0 years vs 2 years, P=0.001). Women who had achieved a pregnancy were less likely to have poor respiratory health (FEV1 <50%), age adjusted OR 0.56 (95% CI 0.32–0.94) and were less likely to work, OR 0.54 (95% CI 0.32–0.93). Otherwise, CF complications were similar between those who did and did not achieve a pregnancy. Health status in men was not significantly associated with parenting.


View this table:
[in this window]
[in a new window]
 
Table III. Health status of men and women with cystic fibrosis in relation to reproductive outcome

 
Pregnancy outcomes for women with CF were similar on comparing the UKCF database to reports from other developed countries (Table IV). The outcomes were better compared to previously published data from the UK (Edenborough et al., 2000Go). Data on reproductive outcomes for men with CF are limited, which restricts the potential for valid comparisons.


View this table:
[in this window]
[in a new window]
 
Table IV. Comparison of pregnancy outcomes among women with cystic fibrosis in different developed countries

 

    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
We found that very few men and women with CF achieved pregnancies or sought infertility treatment. Older men and women had children, at a time when respiratory and metabolic health was deteriorating. Women who achieved pregnancies had an earlier age at diagnosis, were less likely to be homozygous for DF508 and were less likely to have poor respiratory health (FEV1 <50%). Overall, pregnancy outcome was good. Physicians caring for adults with CF require high quality current data on fertility and reproductive outcomes in order to provide informed counselling.

Almost all the health parameters examined showed deterioration in adulthood compared with childhood, in keeping with the recognized pattern of disease progression in CF (Edenborough, 2001Go; Ratjen and Doring, 2003Go; Mehta et al., 2004Go). However, it is interesting to note that height and weight centiles appear to be determined in childhood with significant numbers of males and females recorded below the 10th centile. This may have important implications for fertility and for pregnancy outcomes. The prevalence of both impaired glucose tolerance and insulin dependant diabetes increased significantly in adulthood for both men and women. While historically, most pregnancies to CF individuals occurred in those with normal or near normal pancreatic function (Corkey et al., 1981Go; Canney et al., 1991Go), more recent studies suggest that many women do embark on pregnancy with diabetes or impaired glucose tolerance. Planning of pregnancies and careful management during pregnancy with specialist multidisciplinary input appears to result in favourable outcomes (Gilljam et al., 1999Go; Goss et al., 2003Go). The use of non-steroidal anti-inflammatory medication and choice of prophylactic antibiotics pose additional challenges when balancing maternal health with optimal foetal development. Obstetric cholestasis is associated with poor pregnancy outcomes (Milkiewicz et al., 2002Go) and may pose further concern when the high rate of disturbed liver metabolism among CF adults is considered.

Although this study included a large number of individuals, the overall pregnancy rates were very low, as were the number of individuals seeking fertility treatment. This may relate to social and personal factors as well as to health status. A survey in the United Kingdom reported that 70% of adults with cystic fibrosis were living full and productive lives, however, only 26% of men were married or co-habiting (Walters et al., 1993Go). It is known that 98% of men with CF have azoospermic infertility, and will require some form of assisted conception technique or donor sperm for conception. It may be that many individuals are not in the social, domestic or economic circumstances to embark on MESA, ICSI or donor sperm insemination. In addition, the ethical issues of assisted conception in the face of chronic ill-health and altered life expectancy are difficult. This may be another reason why so few men embark upon infertility treatment. Parenting and fertility issues are important for men and women with CF but many patients report deficiencies in the discussions with their health care professionals. Men in particular are reluctant to initiate discussions (Fair et al., 2000Go; Sawyer et al., 2001Go).

The number of pregnancies in this study was low but the outcomes were generally good with a high percentage of term live born infants. In the general population, the pre-term birth rare is ~6% compared with 18% for this study (Chamberlain and Morgan, 2002Go). The spontaneous miscarriage rate was 8% compared to a typical rate of 10–15% quoted for the general population (Regan et al., 1989Go). There were no therapeutic terminations in this study group, which either reflects a population who are in optimal health when planning their pregnancies, or, more likely, under-reporting. Our findings compare favourably with pregnancy outcomes for women with CF in other developed countries (Gilljam et al., 2000Go; Gillet et al., 2002Go; Odergaard et al., 2002aGo,bGo). A previous UK study reported a high rate of terminations at 19%, a lower full-term live-born pregnancy rate and a higher rate of pre-term births (Edenborough et al., 2000Go). There have been four further studies published since 2000 looking at pregnancy outcomes among women with cystic fibrosis in developed countries (Gilljam et al., 2000Go; Gillet et al., 2002Go; Odergaard et al., 2002aGo,bGo). The numbers of pregnancies in all the groups are similar but some of the studies have recorded data over a large number of years. The findings are remarkably consistent. Our study has the advantage of placing pregnancies achieved and their outcomes within the context of an entire cohort of CF women. It was also possible for us to establish reproductive outcomes for men, which has been an under-researched area to date (Sawyer, 1996Go; Sawyer et al., 1998Go).

Confusion can arise when comparing different study populations, as women in the UK database have genetic characteristics and possibly phenotypic features that are different from the North American and European CF Registries (Gillet et al., 2002Go; McCormick et al., 2002Go). The pattern of DF508 expression in the study population reflects what would be expected in a predominantly Caucasian population in the UK (McCormick et al., 2002Go). Women who achieved a pregnancy were significantly less likely to be homozygous for DF508. This is in keeping with the findings of a US study that demonstrated milder clinical manifestations of CF and lower mortality with distinct genetic subgroups heterozygous for DF508 compared with those homozygous for DF508 (McKone et al., 2003Go). The implication is that it may be possible to predict fertility and reproductive outcomes in terms of the individual's genotype. This may be particularly relevant to those patients diagnosed by neonatal screening (Farrell et al., 2001Go; Merelle et al., 2001Go; Mehta et al., 2004Go).

There are no data available on the CF status of partners, which may have some bearing on whether a couple would embark on a pregnancy. Attitudes to this have been explored by Sawyer et al. (1998)Go. Looking more closely at the health status of individuals with CF in relation to reproductive outcome, there appears to be little relationship between health status and parenting among men. However, having a pregnancy was associated with a lower prevalence of poor respiratory health (FEV1 <50%) among women. The median age at diagnosis of CF was also lower among women who had a pregnancy than among those who had not. It is presently controversial whether early diagnosis of CF or advances in antimicrobial therapy and nutritional support have led to the dramatic improvement in health, and therefore, life expectancy. Our findings suggest that there may be additional advantages of early diagnosis in terms of reproductive health outcomes. It remains to be seen whether neonatal screening for CF will further improve these outcomes. The numbers of men and women in full or part-time work was chosen as a marker of health, however whilst more men who have parented are in employment, fewer women who have had a pregnancy are working. This probably reflects a preference among women with CF to stay at home caring for their children.

The UKCF database is a growing resource allowing longitudinal follow-up of over 5000 individuals with cystic fibrosis. Regular updates on current health status and annual reviews allow analysis of disease progression as well as accuracy checks of baseline data. The analyses described here represent a snapshot view of the current profile of CF sufferers in the UK. The current generation of CF children are the next potential generation of sexually active adults who will have reproductive health care needs. It has been possible to evaluate the reproductive health and pregnancy outcomes of the current adults within a population-based setting and to relate this to CF features and health status. However, it was not possible to compare health before and after pregnancies or to address the detailed aspects of obstetric and postnatal care. There may also be under-reporting of sensitive information relating to sexual activity, infertility and termination of pregnancy (Henneman et al., 2001Go; Nixon et al., 2003Go). Future studies with prospective follow-up of adolescent cohorts should allow evaluation of contraceptive practices, pre-pregnancy counselling and multi-disciplinary approaches to planning and achieving safe pregnancy outcomes.


    Conclusions
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
Identifying phenotypic and genetic factors that are associated with good fertility outcomes will facilitate the provision of accurate information to patients. It is clear that many adults with cystic fibrosis fail to have a child at the present time and that advances in health care need to be matched by frank discussions about sexual health issues throughout adolescence and into adulthood.


    Acknowledgements
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
The authors are grateful for the support of the UKCF database and committee who reviewed and approved the study protocol. We thank M.Fraser and S.Krawczyk for their expert data validation, Gita Mehta, and in the large CF clinics, all the data entry staff and the directors for their support. D.J.M. and A.M. had the initial idea for the study. All three authors designed the study and sought approval from the UKCF database. J.M. and D.J.M. performed the analyses and drafted the paper. A.M. contributed to manuscript preparation. All authors have seen and approved the final version of the paper. We declare no conflicts of interest.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 Acknowledgements
 References
 
Canney GJ, Corey M, Livingstone RA, Carpenter S, Green L and Levison H (1991) Pregnancy and cystic fibrosis. Obstet Gynaecol 77, 850–853.[Abstract]

Chamberlain G and Morgan M (2002) ABC of Antenatal Care. London: BMJ Publishing.

Corkey CW, Newth CJL, Corey M and Levison H (1981) Pregnancy in cystic fibrosis: A better prognosis in patients with pancreatic function. Am J Obstet Gynecol 140, 737–742.[ISI][Medline]

Edenborough FP, Mackenzie WE and Stableforth DE (2000) The outcome of 72 pregnancies in 55 women with cystic fibrosis in the United Kingdom 1977–1996. Br J Obstet Gynaecol 107, 254–261.[ISI]

Edenborough FP (2001) Women with cystic fibrosis and their potential for reproduction. Thorax 56, 649–655.[Free Full Text]

Edenborough FP (2002) Pregnancy in women with cystic fibrosis. Acta Obstet Gynecol Scand 81, 689–692.[CrossRef][ISI][Medline]

Fair A, Griffiths K and Osman LM (2000) Attitudes to fertility issues among adults with cystic fibrosis in Scotland. Thorax 55, 672–677.[Abstract/Free Full Text]

Farrell PM, Kosorok MR, Rock MJ, Laxova A, Zeng L, Lai HC, Hoffman G, Laessig RH and Splaingard ML (2001) Early diagnosis of cystic fibrosis through neonatal screening prevents severe malnutrition and improves long-term growth. Pediatrics 107, 1–13.[Abstract/Free Full Text]

Gillet D, de Braekeleer M, Bellis G and Durieu I (2002) French Cystic Fibrosis Registry: Cystic fibrosis and Pregnancy. Report from French data (1980-1999). Br J Obstet Gynaecol 109, 912–918.

Gilljam M, Corey M and Tullis D (1999) Diagnosis of cystic fibrosis: epidemiological data from Toronto 1992-1997. Am J Respir Crit Care Med 159, A684.[ISI]

Gilljam M, Antoniou M, Shin J, Dupuis A, Corey M and Tullis DE (2000) Pregnancy in Cystic Fibrosis. Chest 118, 85–91.[Abstract/Free Full Text]

Goss C, Rubenfield GD, Otto K and Aitken M (2003) The effect of pregnancy on survival in women with cystic fibrosis. Chest 124, 1460–1468.[Abstract/Free Full Text]

Henneman L, Bramsen I, Van Os TA, Reuling IEW, Heyerman HGM, van der Laag J, van der Ploeg HM and ten Kate LP (2001) Attitudes towards reproductive issues and carrier testing among adult patients and parents of children with cystic fibrosis (CF). Prenat Diagn 21, 1–9.[CrossRef][ISI][Medline]

Kerem E, Corey M, Kerem B, Durie P, Tsui LC and Levison H (1989) Clinical and genetic comparisons of patients with cystic fibrosis, with or without meconium ileus. J Pediatr 114, 767–773.[ISI][Medline]

Kopito LE, Kosasky HJ and Schwachmann H (1973) Water and electrolytes in cervical mucus from patients with cystic fibrosis. Fertil Steril 24, 512–516.[ISI][Medline]

McCormick J, Green MW, Mehta G, Culross F and Mehta A (2002) Demographics of the UK cystic fibrosis population: implications for neonatal screening. Eur J Hum Genet 10, 583–590.[CrossRef][ISI][Medline]

McKone EF, Emerson SS, Edwards K and Aitken M (2003) Effect of genotype on phenotype and mortality in cystic fibrosis: a retrospective study. Lancet 361, 1671–1676.[CrossRef][ISI][Medline]

Mehta G, Sims EJ, Culross F, McCormick J and Mehta A (2004) Potential benefits of the UK Cystic Fibrosis Database. J R Soc Med 97 (Suppl. 44), 60–71.[ISI][Medline]

Merelle ME, Nagelkerke NF, Lees CM and Dezateux C (2001) Newborn screening for cystic fibrosis. Cochrane Database Syst Rev 3, CD001402.[Medline]

Milkiewicz P, Elias E, Williamson C and Weaver J (2002) Obstetric Cholestasis. BMJ 324, 123–124.[Free Full Text]

Nixon GM, Glazener JA, Martin JM and Sawyer SM (2003) Female sexual health care in cystic fibrosis. Arch Dis Child 88, 265–266.[Abstract/Free Full Text]

Odergaard I, Stray-Pedersen B, Hallberg K, Haanaes OC, Storrosten OT and Johannesson M (2002a) Prevalence and outcome of pregnancies in Norwegian and Swedish women with cystic fibrosis. Acta Obstet Gynaecol Scand 81, 693–697.[CrossRef][ISI][Medline]

Odergaard I, Stray-Pedersen B, Hallberg K, Haanaes OC, Storrosten OT and Johannesson M (2002b) Maternal and fetal morbidity in pregnancies of Norwegian and Swedish women with cystic fibrosis. Acta Obstet Gynaecol Scand 81, 698–705.[CrossRef][ISI][Medline]

Ratjen F and Doring G (2003) Seminar–Cystic fibrosis. Lancet 361, 681–688.[CrossRef][ISI][Medline]

Regan L, Braude PR and Trembath PL (1989) Influence of past reproductive performance on risk of spontaneous abortion. BMJ 299, 541–545.[ISI][Medline]

Rodgers HC, Knox AJ, Toplis PJ and Thornton S (2000) Successful pregnancy and birth after IVF in a woman with cystic fibrosis. Hum Reprod 15, 2152–2153.[Abstract/Free Full Text]

Sawyer SM (1996) Reproductive and sexual health in adolescents with cystic fibrosis. BMJ 313, 1095–1096.[Free Full Text]

Sawyer SM, Tully MA, Dovey ME and Colin AA (1998) Reproductive health in males with cystic fibrosis: knowledge, attitudes, and experiences of patients and parents. Pediatr Pulmonol 25, 226–230.[CrossRef][ISI][Medline]

Sawyer SM, Tully MA and Colin AA (2001) Reproductive and sexual health in males with cystic fibrosis: a case for health professional education and training. J Adolesc Health 28, 36–40.[CrossRef][ISI][Medline]

Walters S, Britton J and Hodgson M (1993) Demographic and social characteristics of adults with cystic fibrosis in the United Kingdom. BMJ 306, 549–552.[ISI][Medline]

Submitted on April 13, 2004; accepted on June 15, 2004.