Thorax 2001;56:237 241 237
Paediatric origins of adult lung disease c 7
Series editors: P Sly, S Stick
Cystic fibrosis
P Robinson
Cystic fibrosis (CF) remains an incurable life same population in the period before the intro-
limiting condition which is the most common duction of screening.2 We have recently exam-
inherited lethal condition in most Western ined similar data from our clinic but have
countries. Advances in the care of this excluded those patients in the non-screened
condition, particularly advances in respiratory group who presented before the average age of
management, have seen the focus of this review diagnosis of the screened CF infants.3 We rea-
become a reality. In the 1940s adult conse- soned that those infants would not have
quences of CF were minimal as most patients benefited from the existence of population
died during infancy and early childhood. based newborn screening as their clinical
Today, the life expectancy of patients with CF symptoms led to their diagnosis earlier than
has advanced well into adult life and has would have occurred on the basis of screening.
prompted the development of specific adult CF When these early diagnosis infants with CF
care centres in many major hospitals as well as were excluded from the unscreened group, we
publications addressing adult issues relating to found no diVerence in lung function or
this disease.1 Future projections suggest that, nutritional parameters at the age of 9 years
within the next decade, most patients with CF between the unscreened (n=29) and the
in many countries will, in fact, be adults rather screened (n=31) groups. Further work is
than children. While debate continues between required to identify whether population based
proponents of newborn screening for CF and newborn screening will result in better lung
those against it, the drive of CF care for the function in later life. While the answer to this
paediatric age group is aimed at providing the question is still unclear when considered on the
patient with a high quality life for as long as basis of population based screening, there is no
possible. The underlying element to achieving question that limitation of lung disease,
this is minimisation of lung disease. While particularly in infancy and early childhood, is
nutritional factors together with endocrino- the major factor in determining the adult con-
logical factors, psychological factors, and sequences of CF lung disease.
sexual factors are vitally important in provid-
ing a good quality of life to the patient with CF,
Bacterial colonisation
the basic determinant of longevity and quality
CF lung disease is clinically characterised by
of life in most CF patients is the degree of lung
the production of thick tenacious mucus with
disease and its rate of progression with increas-
an increased propensity for chronic colonisa-
ing age.
tion with bacterial organisms. At birth and
before the onset of lower airway infection there
Screening is no destructive inflammatory process
Newborn screening for cystic fibrosis has been present.4 One of the first steps used to prevent
adopted in some centres as a way of identifying lower airway bacterial colonisation is the early
asymptomatic patients and initiating early detection of, and aggressive early treatment of,
treatment with the aim of minimising long any lower airway colonisation. In a situation
term changes. Proponents of this method similar to that of population based screening,
would suggest that this means a much more some centres prefer to treat all patients with
aggressive approach to management of lung prophylactic antibiotics in an attempt to
disease in the early years as opposed to patients prevent, or at least delay, early colonisation of
born in unscreened populations where clinical the airway with Staphylococcus aureus. While
evidence of respiratory disease is a frequent anti-staphylococcal antibiotic prophylaxis has
cause of presentation. The natural extension of been shown to reduce the amount of cough
this claim would be that early aggressive and symptoms as well as the number of antibiotic
often preventive treatment of lung disease will courses and hospital admissions in the first few
CF Services, Royal
slow the deterioration of lung function in the years of life, there is also some evidence that it
Children s Hospital,
long term. Waters and colleagues have studied may be associated with earlier acquisition of
Melbourne, Australia
long term lung function in a group of infants Pseudomonas aeruginosa.5 7
3052
born during a period of population based new- An alternative approach to assessment and
P Robinson
born screening and have shown that those with treatment of early bacterial colonisation of the
Correspondence to:
CF born during this period have better lung lower airways in infants with CF is the aggres-
Dr P Robinson
function and improved growth parameters sive use of bronchial lavage as a means of
philrob@
cryptic.rch.unimelb.edu.au compared with children with CF born in the obtaining direct evidence of lower airway colo-
www.thoraxjnl.com
238 Robinson
nisation. While this technique may permit early with a rise in specific anti-pseudomonal
detection and aggressive treatment of any S antibodies. Hoiby has further shown that
aureus colonisation, it does not directly achieve
aggressive treatment of P aeruginosa during
what is aimed for with S aureus prophylaxis
intermittent colonisation can achieve clearing
that is, the delay or prevention of S aureus colo-
of the organism from the airways. He compared
nisation. While S aureus colonisation may
48 patients treated with aggressive anti-
certainly produce symptoms when present in
pseudomonal therapy with 42 historic controls
the lower airways of infants with CF, it is colo-
and found that only 16% of the treated patients
nisation with P aeruginosa that is associated
developed chronic infection with P aeruginosa
with a significant increase in airway inflamma-
over a period of 3.5 years compared with 72%
tion and lung injury. Long term prognosis is
of the controls. For this reason continuous sur-
impaired in infants whose airways are infected
veillance for the first sign of airway colonisa-
with this organism when compared with
tion, either by regular bronchial lavage or
airways intermittently colonised by S aureus.
examination of upper airway washings, is now
Hudson and colleagues showed that, in infants
increasingly employed.12
aged under 2 years whose oropharyngeal flora
A third organism which directly influences
included P aeruginosa, there was a decrease in
adult consequences of paediatric CF lung dis-
clinical score at the age of 5 years and
ease is Burkholderia cepacia. Recognised as a
decreased pulmonary function at 7 years of
CF pathogen initially as an organism which
age.8 In those infants colonised with both S
produced aggressive lung disease and usually
aureus and P aeruginosa the same results were
death within a short period, later work
found; however, this group had a 10 year
identified two alternative clinical courses.13 14
survival estimate of 57% compared with
While some patients exhibited a slower but still
92 100% for infants not colonised or colonised
progressive decline in pulmonary function cul-
with P aeruginosa alone. Kerem and colleagues
minating in death, a third group of patients
studied 895 patients with CF and found that
colonised with B cepacia, many of whom were
infants colonised with P aeruginosa in the first
children, were shown to carry the organism for
year of life had similar 10 year survival
extended periods of time without apparently
estimates to those who became colonised at
any adverse outcome.1 Recent work has shown
1 7 years of age and also after 7 years. Early
that the diVerences in clinical severity of lung
colonisation was, however, associated with a
significant reduction in lung function at 7 years disease seen following colonisation with B
of age and this reduction continued through cepacia result from diVerent genotypes of the
into adult life.9 organism.15
Henry and colleagues showed that colonisa- While 40 80% of adult patients will be colo-
tion with mucoid strains of P aeruginosa was
nised with P aeruginosa, Hoiby has also shown
associated with an increased level of morbidity
that regular aggressive treatment with anti-
and mortality.10 They compared the survival of
pseudomonal agents in P aeruginosa positive
50 children who had mucoid P aeruginosa iso-
patients is associated with an improvement in
lated from their sputum cultures with that of 19
lung function over a period of 12 months.16
children whose sputum showed non-mucoid P
Given the drive to prevent or delay the onset
aeruginosa and 12 children who had no P aeru-
of lower airway colonisation with P aeruginosa,
ginosa isolated from their sputum. Over an 8
preventive public health measures have been
year follow up period 42% of mucoid positive P
shown to be important in limiting the spread of
aeruginosa patients died compared with 11% of
the organism between patients. Historically,
those with non-mucoid P aeruginosa and 8% of
camps for children with CF were considered a
those with no P aeruginosa colonisation.
good way of empowering young children with
Nixon and colleagues have recently shown
this condition to help manage their lives. How-
that, in a group of 56 infants diagnosed with
ever, recognition that cross infection occurs
CF on newborn screening, 43% were colonised
between P aeruginosa positive negative patients,
with P aeruginosa by the age of 7 years. This
as well as between B cepacia positive and nega-
group had higher hospital admission rates and
tive patients, has now led to most major CF
lower National Institutes of Health scores at 7
organisations recommending that such camps
years than culture negative children. Four
should not be held. Furthermore, in the hospi-
infants from this group died before 6 years of
tal setting many clinics are now using strategies
age, all of whom had mucoid P aeruginosa cul-
to limit the possible cross infection between
tured from the lower airway prior to death.
patients in areas such as physiotherapy, wards,
While these studies all confirm that early
and outpatient clinics. A much harder chal-
colonisation with P aeruginosa, with or without
lenge is to set suggested guidelines for limiting
S aureus, in the paediatric setting has significant
cross infection in the community such as at CF
implications for CF lung disease in adults,
group meetings, travelling together in cars, and
there is evidence that aggressive treatment of
working in closed spaces. Some CF organisa-
early P aeruginosa colonisation is associated
tions have drawn up guidelines in an attempt to
with a transient clearing of this organism from
address this issue.16 In addition to the early
the airway.11 Hoiby diVerentiates between
recognition of the first isolation of P aeruginosa
intermittent colonisation and chronic infection
from the lower airway, limitation of the
with P aeruginosa and describes intermittent
progressive nature of CF lung disease has been
colonisation as preceding chronic infection by
an average of 12 months. The change from shown to be achieved by regular aggressive
intermittent to chronic infection is associated therapy of anti-pseudomonal chest infections.17
www.thoraxjnl.com
Cystic fibrosis 239
Physiotherapy adolescent patients with CF who had signifi-
While antibiotic therapy is accepted as the cant lung disease requiring frequent admission
backbone of treatment for pseudomonal colo- to hospital and who were symptomatic of reflux
nisation in CF, the underlying abnormality in during postural drainage physiotherapy. When
mucociliary clearance produced by pseudomo- their physiotherapy was changed to PEP
nal infection of the airway has been addressed therapy all patients reported a reduction in
by the institution of regular chest physio- symptoms of reflux but, in addition, there was
therapy, both in acute deteriorations in lung a significant improvement in lung function in
function and also in the maintenance of long these patients over the first 6 months of the
term stability of lung function. Most trials of PEP therapy. This improvement was sustained
physiotherapy techniques have investigated the for a further 18 months and was associated
eVects of diVering types of treatment on short with a reduction in the number of hospital
term improvements in lung function or other inpatient days per year.22 In infants with CF,
parameters of lung disease such as sputum however, the presence of gastro-oesophageal
production.18 19 At least two trials have exam- reflux during head down physiotherapy re-
ined the benefits of various physiotherapy mains controversial.23
techniques on long term lung function and
thus can be considered in a discussion of adult Tobramycin
consequences of paediatric lung disease as Ramsey and colleagues have recently shown
institution of such treatment in childhood may that intermittent regular administration of high
limit adult lung diseases in CF. In 1988 dose (300 mg bd) nebulised tobramycin to 520
Reisman and colleagues compared the eVect of patients with CF of mean age 21 years was
a combination of conventional physiotherapy associated with improved lung function, de-
(postural drainage and chest percussion) and creased density of P aeruginosa in the sputum,
the forced expiratory technique (FET) with the and decreased risk of admission to hospital
eVects of FET alone over a three year period in over a 24 week study period.24 However, no
a group of patients with CF.20 A significantly significant benefit was found in patients aged
greater decline in lung function (as assessed by 6 12 years, so the role of high dose tobramycin
FEF25 75) was seen in patients using FET alone as a therapeutic tool for paediatric patients in
than in those who also performed conventional limiting the adult consequences of their lung
chest physiotherapy. They concluded that the disease remains uncertain. As a significant
long term course of pulmonary function is improvement in lung function was evident in
adversely aVected when conventional physio- patients aged 13 17 years, there may be a role
therapy is not used. While this finding justified for tobramycin in influencing the adult conse-
the continued use of the time consuming con- quences of paediatric lung disease.
ventional postural drainage and chest percus-
sion technique, further advances in chest Anti-inflammatory therapy
physiotherapy have included the introduction While the aggressive and regular assessment
of devices such as the flutter valve and positive and treatment of bacterial colonisation of the
expiratory pressure (PEP) mask which allow lower airways is the benchmark for limitation of
application of positive end expiratory pressure progressive CF lung disease, increasing interest
during physiotherapy.21 In a recent study Mc- is being shown in the adjunctive use of
Ilwaine and colleagues compared the long term anti-inflammatory agents. Earlier trials using
eVect of PEP mask physiotherapy with conven- corticosteroids were unsuccessful because of
tional postural drainage and percussion the increased incidence of side eVects in the
therapy in a group of 40 patients with CF over steroid treated groups of patients.25 More
a 12 month period and found a significantly recently, work has focused on less toxic oral
greater improvement in lung function, assessed agents such as ibuprofen and inhaled agents
by measurements of forced vital capacity with anti-inflammatory actions such as inhaled
(FVC), forced expiratory volume in one second steroids. Konstan et al26 showed, in a small
(FEV1) and forced expiratory flow (FEF25 75), group of patients treated with high doses of
in the patients who used PEP mask physio- ibuprofen over 4 years, that there was less dete-
therapy compared with those who used con- rioration in lung function in the ibuprofen
ventional postural drainage and percussion treated group (n=27) than in a placebo treated
therapy. This finding, taken in association with group (n=30). The ibuprofen treated group
the earlier findings of Reisman, indicate that had a significantly lower reduction in FEV1
physiotherapy has a major influence in limiting than the placebo treated group (1.48% v
the adult consequences of CF. The study by 3.57% per year); however, the rate of decline in
McIlwaine et al illustrates the importance of the control group in this study is significantly
assessing newer techniques against proven higher than that recorded in other larger stud-
established older techniques before such tech- ies. Ina4year study of oral steroids in children
niques are widely incorporated into recom- with CF the placebo treated group had an
mended therapeutic regimes. annual rate of decline in FEV1 of only 1.5%.25
One area of chest physiotherapy that has Recent data on 2100 patients released from the
recently received considerable attention, and European Epidemiologic Registry of Cystic
has direct implications for the adult conse- Fibrosis showed an annual rate of decline in
quences of lung disease in CF, is the relation- FEV1 of 1.1% for the group as a whole. It
ship between chest physiotherapy, gastro- remains to be confirmed whether long term
oesophageal reflux, and lung disease in young treatment with high doses of ibuprofen pro-
infants with CF. Button et al22 examined six duces a clinical diVerence in reducing the rate
www.thoraxjnl.com
240 Robinson
of decline in FEV1 as opposed to a statistical Nutritional factors
diVerence. A wide range of other anti- Early studies of the relationship between
inflammatory agents has been postulated as malnutrition and lung disease in children with
CF found a direct relationship between the two
possibly being eVective in influencing the long
parameters and an equal relationship between
term progression of lung disease in CF. At least
both factors and survival rates. What was
three trials of inhaled steroids have been
unclear, however, was whether this was a direct
reported but none has produced convincing
causal relationship and, further, whether im-
long term benefits.27 29 In addition, other
proving the nutritional aspects would improve
inflammatory agents under review include
pulmonary disease. In a comparative study
pentoxifylline, fish oil, tyloxapol, antiproteases
between the Boston and Toronto CF clinics in
(including secretory leukoprotease inhibitor,
the early 1980s a significantly diVerent median
-antiprotease inhibitor, cell penetrant mono-
1
age of survival between the two clinics was evi-
cyclic lactam inhibitor, and monocyte/
dent with Toronto, a clinic which placed a
neutrophil elastase inhibitor), and antioxidants
heavy emphasis on nutritional therapy, being
such as glutathione and -carotene.30 Whether
associated with a much better survival rate.
any of these agents will influence the adult
Interestingly, despite the diVerence in survival
consequences of paediatric CF lung disease is
rates between the two clinics, the level of
yet to determined.
pulmonary function was the same. With the
Recognition of the group of compounds
recognition of the CF gene in the late 1980s,
called inbiotics such as tegrins has led to
attention was drawn to whether there were
current trials of agents such as protegrin, a
genotype-phenotype correlations which could
polypetide of porcine origin which has been
predict the severity of lung disease. While pan-
shown to have antibacterial properties. Suc-
creatic suYcient patients are known to have
cessful use of these agents in the paediatric set-
better pulmonary function at any given age and
ting may limit the adult consequences of CF
longer survival than those with pancreatic
lung disease. In addition, successful identifica-
insuYciency, it is as yet unclear whether this
tion of agents which can promote the function
milder lung disease is secondary to a milder CF
of the CFTR molecule or alter the airway envi-
genetic lesion or whether it is directly associ-
ronment (such as amiloride) may, in the long
ated with better nutrition (pancreatic suYcient
term, be able significantly to influence the
patients having normal levels of fat soluble
adult consequences of paediatric CF lung dis-
vitamins and normal fatty acid composition of
ease by limiting the rate of progression of this
membrane lipids32). While this discussion
condition.
suggests that nutrition may be an influencing
Pulmozyme or recombinant DNase has been
factor in limiting the adult consequences of
shown to produce an improvement in lung
paediatric CF lung disease, it does highlight a
function in some patients with CF and is now
larger area of influence namely, the variation
widely used in many countries to help patients
in the clinical course of CF lung disease associ-
expectorate thick mucus. A large multinational
ated with diVerent CF genotypes.
placebo controlled study (the Pulmozyme
Early Intervention Trial) is currently being
Conclusions
conducted which aims to investigate whether
Although this review has focused on therapeu-
regular administration of Pulmozyme in chil-
tic interventions which may limit the adult
dren with mild lung disease (starting FEV1 consequences of paediatric CF lung disease, it
>80% predicted) will limit the deterioration in
is important to recognise the challenge of long
FEV1 over the 2 year period of the study.
term adherence to prescribed therapy in
patients with CF from the perspective of the
patient, the family, and the healthcare person-
nel. A disappointing result in this present era is
Gene therapy
to witness the natural course of CF in a child as
The ultimate way to limit the adult conse-
a result of either misdiagnosis or failure to
quences of CF is to correct the underlying bio-
accept or use conventional therapy. While con-
chemical eVect in the respiratory system soon
siderable advances are yet to be made to permit
after birth. While trials of gene therapy in CF
accurate genotype-phenotype correlations, for
have been occurring for over 5 years, there are
most genotypes there is no question that, if
still fundamental problems with the therapy
untreated, the adult consequences of paediatric
both in terms of safety (viral vector gene
CF lung disease are generally very poor. For
therapy) and eYciency (liposomal gene
this reason measures aimed at improving and
therapy). At present trials of gene therapy for
encouraging adherence to therapy are vital
CF are being conducted in the adult popula- tools in minimising these adult consequences.
tion; however, gene therapy introduced early in
As mentioned earlier, changes in paediatric
life after the diagnosis of CF has been made
treatment for CF have meant that concerns
would have the potential to produce a major
regarding the adult consequences of paediatric
alteration in the spectrum of CF lung disease in
CF lung disease have become a reality. While
the adult. Despite the intense desire of many the preceding discussion has concentrated on
parents whose children have CF for this some of the specific steps that have been iden-
therapy to be available and eVective in the near tified as minimising the adult consequences of
future, the reality is that it is still many years paediatric lung disease, it is implicit that care of
away from being a regular and eVective patients with CF is best conducted in centres
therapy.31 with experience in CF care and the ability to
www.thoraxjnl.com
Cystic fibrosis 241
14 Lewin LO, Byard PJ, Davis PB. EVect of Pseudomonas cepa-
provide the wide spectrum of expertise neces-
cia colonization on survival and pulmonary function of
sary for total care not only medical staV of
cystic fibrosis patients. J Clin Epidemiol 1990;43:125 31.
various disciplines including respiratory physi- 15 Sun L, Jiang R-Z, Steinbach S, et al. The emergence of a
highly transmissible lineage of cbl+ Burkholderia cepacia
cians, gastroenterologists, endocrinologists,
causing cystic fibrosis centre epidemics in North America
and interventional radiologists but also dieti- and Britain. Nature Med 1995;1:661 6.
16 Frederiksen B, Koch C, Hoiby N. Changing epidemiology
tians, physiotherapists, and psychologists.
of Pseudomonas aeruginosa infection in Danish cystic fibro-
Early reports suggesting that care of CF
sis patients (1974 1995). Pediatr Pulmonol 1999;28:159
66.
patients in tertiary centres provided a better
17 Fridkin SK, Weinstein RA. Resistant gram-negative infec-
long term prognosis were sometimes criti-
tions, including cystic fibrosis. In: Abrutyn E, Goldman D,
cised.33 34 However, in a recent study by
Scheckler W, eds. Saunders Infection Control Reference Serv-
ice. New York: WB Saunders, 1998: 662 4.
Mahadeva et al, in which the long term
18 Braggion C, Cappelletti LM, Cornacchia M, et al. Short
outcome of patients treated in a tertiary care
term eVects of three chest physiotherapy regimens in
patients hospitalized for pulmonary exacerbations of cystic
centre specialising in CF was compared with
fibrosis: a cross over randomized study. Pediatr Pulmonol
that in patients treated in paediatric centres not
1995;19:16 22.
specialising in CF care, the improved nutri- 19 Desmond KJ, Schwenk WF, Thomas E, et al. Immediate
and long term eVects of chest physiotherapy in patients
tional and pulmonary status of patients treated
with cystic fibrosis. J Pediatr 1983;103:538 42.
at a specialist centre is good evidence that
20 Reisman J, Rivington-Law B, Corey M, et al. Role of
conventional physiotherapy in cystic fibrosis. J Pediatr
regular treatment and assessment in centres
1988;113:632 6.
specialising in CF care is a further way of mini-
21 McIlwaine M, Wong LT, Peacock D, et al. Long-term com-
mising the adult consequences of paediatric parative trial of conventional postural drainage and percus-
sion versus positive expiratory pressure physiotherapy in
CF lung disease.34
the treatment of cystic fibrosis. J Pediatr 1997;131:570 4.
22 Button BM, Heine RG, Catto-Smith AG, et al. Postural
drainage in cystic fibrosis: is there a link with gastro-
1 Yankaskas JR, Knowles MR. Cystic fibrosis in adults.
oesophageal reflux? J Pediatr Child Health 1998;34:330 4.
Philadelphia: Lippincott-Raven, 1999.
23 Phillips GE, Pike SE, Rosenthal M, et al. Holding the baby:
2 Waters DL, Wilcken B, Irwig L, et al. Clinical outcomes of
newborn screening for cystic fibrosis. Arch Dis Child 1999; head downwards positioning for physiotherapy does not
80:F1 7. cause gastro-oesophageal reflux. Eur Respir J 1998;12:954
3 Robinson PJ, Newman Y, Vyas J. Does neonatal screening
7.
influence spirometry at 9 years of age? Pediatr Pulmonol
24 Ramsey BW, Pepe MS, Quan JM, et al. Intermittent admin-
1999;Suppl 19:S337.
istration of inhaled tobramycin in patients with cystic fibro-
4 Armstrong DS, Grimwood K, Carzino R, et al. Lower respi-
sis. N Engl J Med 1999;340:23 30.
ratory infection and inflammation in infants with newly
25 Eigen H, Rosenstein BJ, FitzSimmons S, et al. A multicenter
diagnosed cystic fibrosis. BMJ 1995;310:1571 2.
study of alternate-day prednisolone therapy in patients
5 Weaver LT, Green MR, Nicholson K, et al. Prognosis in
with cystic fibrosis. J Pediatr 1995;126:515 23.
cystic fibrosis treated with continuous flucloxacillin from
26 Konstan MW, Byard PJ, Hoppel CL, et al. EVect of
the neonatal period. Arch Dis Child 1994;70:84 9.
high-dose ibuprofen in patients with cystic fibrosis. N Engl
6 Wright GLT, Harper J. Fusidic acid and lincomycin therapy
J Med 1995;332:848 54.
in staphylococcal infections in cystic fibrosis. Lancet 1970;
27 Nikolaizik WH, Schoni MH. Pilot study to assess the eVect
i:9 14.
of inhaled corticosteroids on lung function in patients with
7 Harrison CJ, Marks MI, Welch DF, et al. A multicentric
cystic fibrosis. J Pediatr 1996;128:271 4.
comparison of related pharmacologic features of ce-
28 Van Haren EHJ, Lamers J-WJ, Festen J, et al. The eVects of
phalexin and dicloxacillin given for two months to young
children with cystic fibrosis. Pediatr Pharmacol 1985;5:7 the inhaled corticosteroid budesonide in lung function and
16. bronchial hyperresponsiveness in adult patients with cystic
8 Hudson VL, Wielinski CL, Regelmann WE. Prognostic fibrosis. Respir Med 1995;89:209 14.
implications of initial oropharyngeal bacterial flora in 29 Balfour-Lynn IM, Klein NJ, Dinwiddie R. Randomized
patients with cystic fibrosis diagnosed before the age of two
controlled trial of inhaled corticosteroids (fluticasone pro-
years. J Pediatr 1993;122:854 60.
pionate) in cystic fibrosis. Arch Dis Child 1997;77:124 30.
9 Kerem E, Corey M, Gold R, et al. Pulmonary function and
30 Oerman C, Sockrider MM, Konstan MW. The use of anti-
clinical course in patients with cystic fibrosis after
inflammatory medications in cystic fibrosis. Chest 1999;
pulmonary colonisation with Pseudomonas aeruginosa. J
115:1053 8.
Pediatr 1990;116:714 9.
31 Geddes DM, Alton EWFW. The CF gene: 10 years on. Tho-
10 Henry RL, Mellis CM, Petrovic L. Mucoid Pseudomonas
rax 1999;54:1052 3.
aeruginosa is a marker of poor survival in cystic fibrosis.
32 Gaskin K, Gurwitz D, Durie P, et al. Improved respiratory
Pediatr Pulmonol 1992;12:158 61.
prognosis in patients with cystic fibrosis with normal fat
11 Valerius NH, Koch C, Hoiby N. Prevention of chronic Pseu-
absorption. J Pediatr 1982;100:857 62.
domonas aeruginosa colonisation in cystic fibrosis by early
33 Phelan P, Hey E. Cystic fibrosis mortality in England and
treatment. Lancet 1991;338:3725 6.
Wales and in Victoria, Australia. Arch Dis Child 1984;59:
12 Frederiksen B, Koch C, Hoiby N. Antibiotic treatment of
71 3.
initial colonization with Pseudomonas aeruginosa postpones
34 Lewis PA. Inferences for health provision from survival data
chronic infection and prevents deterioration of pulmonary
in cystic fibrosis Arch Dis Child 1998;79:297 9.
function in cystic fibrosis. Pediatr Pulmonol 1997;23:330 5.
35 Mahadeva R, Webb K, Westerbeek RC, et al. Clinical
13 Tablan OC, Chorba TL, Schidlow DV, et al. Pseudomonas
cepacia colonisation in patients with cystic fibrosis: risk fac- outcome in relation to care in centers specializing in cystic
tors and clinical outcome. J Pediatr 1985;107:382 7. fibrosis: cross sectional study. BMJ 1998;316:1771 5.
www.thoraxjnl.com