www.elsevier.com/locate/jcf

Journal of Cystic Fibrosis

Cellular proteolysis and systemic inflammation during exacerbation

in cystic fibrosis

Alina A. Ionescu, Lisette S. Nixon, Dennis J. Shale*

Section of Respiratory and Communicable Diseases, Department of Medicine, University of Wales College of Medicine, UK

Academic Centre, Llandough Hospital, Cardiff and Vale NHS Hospital Trust, Penarth, South Glamorgan, CF64 2XX, UK

Received 3 March 2004; accepted 27 July 2004

Abstract

Background: Weight loss indicates a poor prognosis in cystic fibrosis (CF). We hypothesised that fat-free mass (FFM) depletion and

increased systemic inflammation would be associated with increased cellular proteolysis during an exacerbation of the respiratory symptoms.

Patients were studied prospectively from the beginning of treatment with antibiotics when admitted to the Adults CF Centre.

Methods: Twenty six patients with CF were studied at the start and end of antibiotic treatment and 2 weeks later. Mean (95% CI) FEV1 when

clinically stable was 54.1 (44.5, 62.6)% predicted. Urinary excretion of Pseudouridine (5-ribosyluracil, PSU) was determined as an indicator

of cellular protein breakdown. Body composition was assessed by dual energy X-ray absorptiometry (DXA).

Results: Patients had increased concentrations of PSU at all assessments ( pb0.01). Those with a low FFM had greater PSU (ratio to FFMI)

than those with a normal FFM at all assessments. At the start of treatment, PSU was related to FFM, C-reactive protein (CRP) ( pb0.05) and

tumour necrosis factor (TNF)a soluble receptors (sr) I and II ( pb0.01). Circulating inflammatory mediators were greater in patients than in

healthy subjects at all assessments.

Conclusion: Increased protein breakdown is associated with a low FFM and increased systemic inflammation and it may be a contributory

mechanism of poor weight preservation in CF.

D 2004 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved.

Keywords: Protein catabolism; Systemic inflammation; Cystic fibrosis; 5-Ribosyluracil; Protein breakdown; Fat-free mass

1. Introduction

Weight loss with depletion of fat-free mass (FFM) is

associated with increased morbidity and mortality in cystic

fibrosis (CF) [1,2]. The cause of such loss is likely to be

multifactorial with reduced nutrient intake contributing to a

negative energy balance. Other factors include an increased

energy and oxygen cost of breathing, which is added to by

systemic inflammatory and catabolic responses, also

1569-1993/$ - see front matter D 2004 European Cystic Fibrosis Society. Publish

doi:10.1016/j.jcf.2004.07.003

* Corresponding author. Section of Respiratory and Communicable

Diseases, University of Wales College of Medicine, Academic Centre,

Llandough Hospital, Penarth, South Glamorgan, CF64 2XX, UK. Tel.: +44

29 20716948; fax: +44 29 20716416.

E-mail address: shaledj@cardiff.ac.uk (D.J. Shale).

reported in chronic obstructive pulmonary disease (COPD)

[3–6].

We have reported increased excretion of 5-ribosyluracil

(pseudouridine, PSU), an indicator of RNA and cellular

protein breakdown, in clinically stable adults with CF

compared with matched healthy subjects [7]. Excretion of

PSU has been validated against amino-acid infusion and

used to assess cell turnover in inflammatory diseases,

leukaemia, lymphomas, and solid tumours [8–12]. PSU is

derived from RNA breakdown, is independent of diet, is

neither metabolised further nor re-incorporated, and hence

indicates cell turnover and cell-derived protein breakdown.

PSU was related to leucin turnover (Amol/kg/min, r=0.68,

pb0.01) and to ketoisocaproic acid turnover (r=0.96,

pb0.001) [8]. Excretion of PSU is influenced by age, but

with little reported variation from day to day [13]. In other

3 (2004) 253–258

ed by Elsevier B.V. All rights reserved.

A.A. Ionescu et al. / Journal of Cystic Fibrosis 3 (2004) 253–258254

inflammatory states associated with weight loss and

cachexia, such as rheumatoid arthritis and HIV, PSU has

been found to relate to disease severity and outcome

[11,12]. Excretion of PSU was linked to the severity of

lung disease and to the loss of FFM [7]. A possible link

between inflammation and proteolysis is that IL-6 and

TNFa stimulate proteolysis in skeletal muscle probably by

activation of the ATP-ubiquitin-proteasome pathway [14].

The transgenic IL-6 mouse constitutively secretes IL-6 and

develops a severe proteolytic myopathy, which was reversed

by an IL-6 receptor blocking antibody [15]. Infusion of

TNFa into the forearm arteries of healthy subjects leads to a

rapid efflux of amino acids and in vitro acts via the NF-kB

pathway to inhibit the differentiation of myoblasts into

myocytes [16]. Previous studies in children with CF have

reported abnormalities of protein turnover related to poor

nutrition, but the relationship to systemic inflammation has

not been explored [17,18].

Based on such findings, we hypothesised that increased

cellular proteolysis occurs during an exacerbation of

respiratory symptoms, persists after the antibiotic treatment

and is linked to a reduced FFM and increased systemic

inflammation. We tested this hypothesis by investigating the

relationship between a marker of cellular proteolysis and

systemic inflammatory mediators in adults with CF during

and after the treatment of an exacerbation of respiratory

symptoms.

2. Materials and methods

2.1. Patients

Twenty six patients (12 male) with known CF (clinical

findings, sweat Na+ and Cl� N70 mmol/L and appropriate

genotype) mean (95% CI) age 23.8 (22.2, 25.4) years were

included at the beginning of a 14-day course of antibiotic

treatment for an exacerbation of their respiratory symptoms.

Mean (95% CI) FEV1 when clinically stable was 54.1 (45.5,

62.7)% predicted. An exacerbation was defined as increased

cough, sputum production and purulence, shortness of

breath, systemic symptoms such as fever or weight loss,

and a reduction in FEV1N10% compared with values when

clinically stable in the preceding year. All patients were

colonised with Pseudomonas aeruginosa, defined as more

than three isolations from sputum in the preceding 6

months. All patients had a DF508/DF508 genotype, all

received enzyme supplements with meals for pancreatic

insufficiency and had no clinical signs of malabsorption

when studied. Diabetes mellitus, liver cirrhosis, chronic

respiratory failure with cor pulmonale and oral cortico-

steroid treatment were exclusion criteria. The study had

Local Research Ethics Committee approval and all patients

gave written consent. The study was carried out over an 8-

month period, which allowed 26 consecutive patients to be

included.

2.2. Healthy subjects

Twenty five age matched healthy subjects (11 male) age

24.6 (23.3, 25.9) years had blood samples taken for

concentrations of inflammatory mediators and urine samples

for concentrations of pseudouridine and creatinine.

2.3. Methods

2.3.1. Body composition

Height and weight were measured and the body mass

index calculated (kg/m2). Fat mass and FFM were assessed

by dual energy X-ray absorptiometry (DXA, Hologic,

Waltham, Ma, USA, software denhanced body composition

v5.70T). Two groups of patients were identified: (1) low

FFM if less than the lower 5th percentile for healthy age

matched subjects who had previously been studied with

the same methodology [7]; (2) normal FFM if more than

the lower 5th percentile for healthy subjects. This cut-off

point was used in order to clearly identify those patients

with severe depletion of FFM [7]. DXA was assessed once

within the study period, when the patients were clinically

stable.

2.3.2. Food intake assessment

The nutritional intake was assessed over three consec-

utive days, at the beginning and end of antibiotic treatment

and 2 weeks later. All patients were reviewed by a dietician

on admission and all received nutritional supplement drinks.

Three patients received supplementary overnight feeding

through naso-gastric (two patients) or percutaneous gastro-

stomy tubes (one patient). The mean energy intake of the 3

days was reported, including nutritional supplements, as

calculated.

2.3.3. Urinary pseudouridine

Pseudouridine was measured by high-performance

liquid chromatography in the second void dspotT morning

urine sample and reported as a ratio to urinary creatinine

(Amol/mmol) [19,20]. Repeated measurements from our

laboratory of PSU in patients with CF during clinical

stability and in healthy adults have found a day to day

variability of less than 10%. Patients with CF had PSU

measured at the start and end of antibiotic treatment and 2

weeks later. The healthy subjects had PSU measured on

one occasion. Excretion of PSU was also reported as ratio

to FFM index (FFMI, (Amol/mmol/kg/m2) to control for

the differences in body composition and height between

individuals (a greater protein turnover would be expected

for a larger FFM).

2.3.4. Spirometry

FEV1 was measured by dry wedge spirometry and

expressed as % predicted. The number of exacerbations of

respiratory symptoms treated with i.v. antibiotics in the year

previous to the study was recorded from the patients files.

Fig. 1. PSU/FFMI in patients at the start (Day 1) and end (Day 14) of

antibiotic treatment for an exacerbation, 2 weeks later (Day 28) and during

A.A. Ionescu et al. / Journal of Cystic Fibrosis 3 (2004) 253–258 255

2.3.5. Circulating inflammatory mediators

Blood was obtained at the same time as the urine

collection for PSU measurement. Serum C-reactive protein

(CRP), Interleukin 6 (IL-6), tumour necrosis factor a

(TNFa) and their soluble receptors (sr) I and II were

assayed by ELISA as previously described [21].

2.3.6. Statistical analysis

Non-normally distributed data was log10 transformed and

expressed as geometric means. Student’s T-test and ANOVA

were used for comparison between groups. Multiple

stepwise regression was used to assess the impact of

inflammatory, lung function and body composition param-

eters on PSU excretion.

clinical stability (Stable state). Data presented as means and 95%

confidence intervals. Patients with a low FFM (grey columns) had greater

PSU/FFMI than healthy subjects (dotted column) at all assessments

( pb0.001). Those with a normal FFM (white columns) had greater PSU/

FFMI at the start and end of antibiotic treatment for an exacerbation, 2

weeks later, but not during clinical stability.

3. Results

3.1. Body composition

The mean (95% CI) BMI of the patients was 20.5 (19.4,

21.7) kg/m2 measured 2 weeks after the end of treatment,

when clinically stable. Of the 26 patients, 12 had a low FFM

(5 males), 34.5 (29.9, 39.1) kg compared with 46.8 (42.4,

51.3) kg for those with a normal FFM, respectively, and

51.2 (43.9, 58.4) kg for the healthy subjects.

3.2. Food intake

Sixteen patients had lower energy intake than recom-

mended for CF. There was no difference in the energy intake

or the proportion energy from protein, fat or carbohydrate

between the three assessments or between patients with a

low or a normal FFM. The total energy intake for the whole

group of patients was 2145.0 (1838.0, 2453.4) at the

beginning and 2178.1 (1839.8, 2619.3) (kcal/day) at the

end of treatment; the percent energy derived from protein

was 14.4 (12.5, 16.7) and 13.6 (11.7, 15.2)%, respectively.

3.3. Excretion of PSU

At all three assessments patients had a greater absolute

PSU concentration than the healthy subjects, with no

differences between PSU measured at the three time

Table 1

Urinary pseudouridine (PSU) concentration ratio to FFMI in patients with a

low (n=12) or normal (n=14)

Pseudouridine

(Amol/mmol creatinine/kg/m2)

Low FFM Normal FFM

Start treatment 3.53 (2.48, 4.58) 2.21 (1.83, 2.58)*

End treatment 2.96 (2.16, 3.76) 2.14 (1.64, 2.64)**

2 weeks later 3.10 (2.22, 3.98) 2.17 (1.55, 2.79)***

* pb0.01.

** p=0.016.

*** p=0.025.

points for the patients; mean (95% CI) PSU at the start

49.7 (36.2, 63.2), and end of treatment 40.5 (33.9, 47.0),

and 2 weeks later 40.6 (33.2, 48.0) and for healthy

subjects 23.0 (18.2, 27.8) Amol/mmol creatinine (all

pb0.01). At the start of treatment patients with a low

FFM (n=12) had greater levels of PSU compared to those

with a normal FFM; 63.2 (37.4, 89.3) and 36.2 (30.8,

41.5), respectively ( p=0.03), but there were no differ-

ences after completion of the treatment or 2 weeks later.

However, PSU corrected for the FFMI was greater in

patients with a low FFM than those with a normal FFM

at all assessments (Table 1). PSU/FFMI for the healthy

subjects was 1.33 (1.01, 1.66) Amol/mmol/kg/m2. Patients

with either a normal or low FFM had a greater PSU/

FFMI than the healthy subjects at all assessments

( pb0.001 for all, Fig. 1). These findings were compared

with 40 patients in a stable clinical state [7], (Fig. 1). In

this larger number of patients, PSU/FFMI was similar

between those with a low or normal FFM ( p=0.09), while

only the patients with a low FFM had a PSU/FFMI

greater than the healthy subjects ( pb0.05).

3.4. Circulating inflammatory mediators

The circulating concentrations of inflammatory media-

tors were greater at the start compared to the end of

antibiotic treatment and 2 weeks after completing the

treatment (Table 2). At all assessments, the concentrations

of inflammatory mediators were greater in patients than the

healthy subjects ( pb0.01 for all). Two weeks after the end

of treatment, the circulating concentrations of CRP and IL-6

were greater than at the end of treatment ( pb0.01), while no

difference was found for the other inflammatory mediators.

CRP ( pb0.05), IL-6 ( pb0.05) and TNFa soluble receptors I

and II ( pb0.01 for both) were greater in the patients with a

Table 2

Inflammatory mediators in patients at the start and end of antibiotic treatment and 2 weeks later (n=26); geometric means (95% CI)

Start End 2 weeks later

CRP (Ag/ml) 27.5 (15.1, 51.2)* 4.6 (2.4, 9.5) 12.5 (3.9, 25.1)*

IL-6 (pg/ml) 5.9 (3.9, 8.7)* 2.9 (2.1, 3.9) 6.1 (3.3, 12.5)*

IL-6 sr (ng/ml) 42.6 (37.1, 46.7)* 43.6 (38.9, 48.9) 39.8 (31.6, 50.1

TNFa (pg/ml) 3.7 (3.1, 4.6)* 2.9 (2.5, 3.6) 2.8 (2.1, 3.6)

TNFa sr I (ng/ml) 1.3 (1.1, 1.6)* 1.0 (0.9, 1.2) 1.0 (0.8, 1.3)

TNFa sr II (ng/ml) 3.5 (3.1, 3.9)* 2.8 (2.6, 3.3) 2.5 (2.0, 3.9)

* pb0.01 compared to the end of treatment; 2 weeks after the end of treatment CRP and IL-6 were greater than at the end of treatment, but not the othe

inflammatory mediators.

A.A. Ionescu et al. / Journal of Cystic Fibrosis 3 (2004) 253–258256

low FFM compared to those with a normal FFM at the start

of treatment, but not at the end of treatment or 2 weeks later.

Patients with a low FFM had a greater number of

exacerbations in the year previous to the study compared

with those with a normal FFM; mean (95% CI) was 7.4 (4.6,

10.2) and 3.4 (2.4, 4.3), respectively ( pb0.01).

Analysis of regression showed that the following factors

were related to the concentration of PSU when commencing

treatment with antibiotics; FFM (r2=0.2, p=0.04), CRP (r2=

0.21, p=0.02), IL-6 (r2=0.37, p=0.02), (TNFa sr I (r2=0.3,

pb0.01), TNFa sr II (r2 0.25, p=0.01), and the number of

exacerbations (r2=0.3, pb0.01).

4. Discussion

Increased breakdown of cellular protein as indicated by

urinary PSU concentrations was present in adult patients at

the beginning and end of antibiotic treatment for an

exacerbation of respiratory symptoms and 2 weeks later.

The concentration of PSU was approximately twice that of

the matched healthy subjects. Patients with a low FFM had a

greater output of PSU compared with those with a normal

FFM at all assessments. To allow for differences in total

FFM and height, which was most variable in the patients

and could reflect either impaired growth as a child or current

loss of FFM, we expressed PSU concentration in terms of

the FFMI. Using this ratio, patients with a low FFM had

consistently greater PSU concentrations than those with a

normal FFM, and both had markedly greater concentrations

than the healthy subjects at all assessments. In both patient

subgroups, PSU concentrations were unaffected by anti-

biotic treatment, indicating that the underlying protein

catabolic state was not modified by such treatment.

However, the mean PSU decreased by nearly 20% after

treatment, but the concentration was not statistically differ-

ent from the start of treatment, possibly due to the wide CI

of the mean. We have previously demonstrated that

antibiotic treatment was associated with a reduction in

resting energy expenditure, correction of the cortisol–IGF1

ratio, a reduction in lipolysis and circulating catecholamines

and modest weight gain [3,7], and it might be expected that

such changes would be associated with a reduction in

proteolysis. However, our finding of no change in PSU is in

keeping with those of N-telopeptides of Collagen 1 (NTx) in

)

r

CF where antibiotic treatment led to a reduction in

excretion, but not to levels similar to those in healthy

subjects [22]. These data taken together with the PSU

findings reported in clinically stable adults with CF suggest

that protein catabolism in adults with CF and chronic

pulmonary bacterial infection is a chronic or near contin-

uous process, linked to the persistent systemic inflammatory

response, and is likely to contribute to the loss of skeletal

muscle and bone mineral density, and their associated

complications [22,23]. The inappropriately high protein

breakdown in patients with a low FFM suggests a non-

adapting catabolic response, which was related to a greater

inflammatory response in this group.

Our findings support the view that loss of skeletal muscle

is not due purely to an inadequate energy intake, as this did

not change after the treatment, while it was likely that at the

beginning of treatment, the intake may have been inad-

equate for the increased energy requirements. In the patients

with CF, there is circumstantial evidence of a possible effect

of the increased concentration of systemic inflammatory

mediators and increased proteolysis [7]. While correlation

does not imply causation, the analysis of regression

suggested that at the beginning of an exacerbation, the

circulating concentrations of inflammatory mediators have

an effect on PSU excretion. The increased systemic

inflammation could be one of several mechanistic factors

in the breakdown of cellular protein. The precise source of

increased PSU excretion in our patients is unknown, but

skeletal muscle is likely to be involved, because it

constitutes a large proportion of tissue-related protein. In

other chronic inflammatory lung disease, such as COPD,

altered amino-acid metabolism occurred in both the

circulation and in lower limb muscles at rest and after

exercise [24]. Such changes were related to the acute phase

inflammatory response and to metabolic derangements

[4,5]. It is unclear if changes in amino-acid metabolism

relate to loss of muscle mass, though they could indicate

muscle protein breakdown, particularly during exercise. In

addition to potential inflammatory factors and altered

muscle metabolism, the effect of a negative energy balance

secondary to increased energy costs and an insufficient

energy intake for these needs may be another factor. This

would represent an adaptation in intermediary metabolism

to allow storage and structural tissues, such as skeletal

muscle, to be utilised for energy. Lung disease itself may

A.A. Ionescu et al. / Journal of Cystic Fibrosis 3 (2004) 253–258 257

contribute to extra energy costs as approximately 50% of the

excess in resting energy expenditure was explained by

increased oxygen cost of breathing and 30% by the systemic

inflammatory response [3,25,26].

With regression analysis, the number of exacerbations in

the year prior to the study was shown to have an effect on

protein catabolism. Patients experiencing more frequent

exacerbations may enter a dvicious spiralT where they have

increased proteolysis and weight loss during exacerbations

and are unable to recover the loss of weight and FFM after

each exacerbation because of a high frequency of exacer-

bations. This interpretation would explain our findings for

both PSU and NTx in this and other studies of increased

protein catabolism in adults with CF [7,22]. Such a spiral

associated with severe impairment of lung function and loss

of FFM suggests that it is important to identify such patients

who are more likely to have difficulties in recouping the

weight loss between exacerbations due to sustained protein

catabolism.

4.1. Limitations of this study

This study could be criticised for only determining tissue

protein breakdown by an indirect method, rather than

protein turnover. Generally methods to determine protein

turnover are not applicable to routine clinical use or large

group investigations, because they are laborious and costly

[8,13]. The urinary excretion of methylated nucleosides has

been previously used as a non-invasive method for the

assessment of protein metabolism [9].

This study found, with a non-invasive method, persis-

tently increased cellular proteolysis in adults with CF and

chronic pulmonary bacterial infection. This was associated

with FFM status, exacerbation rates and persistent systemic

inflammation. Future studies will be needed to clarify the

relationship between the increased proteolysis and protein

synthesis in CF.

Acknowledgements

This work was supported by Glaxo Smith Kline, the

Cystic Fibrosis Trust (UK), the Astra Foundation (UK), the

British Lung Foundation and the British Thoracic Society.

The authors would like to thank Dr. Derek Buss and Mr.

Paras Mavani for their assistance in setting up the method

used for the measurement of pseudouridine and Professor

PA Routledge for permitting access to some of the

laboratory equipment.

References

[1] Elborn JS, Shale DJ, Britton JR. Cystic fibrosis: current survival and

population estimates to the year 2000. Thorax 1991;46:881–5.

[2] Davis PB, Drumm M, Konstan MW. Cystic fibrosis. Am J Respir Crit

Care Med 1996;154:1229–56.

[3] Bell SC, Bowerman AM, Nixon LS, Macdonald IA, Elborn JS,

Shale DJ. Metabolic and inflammatory responses to pulmonary

exacerbations in adults with cystic fibrosis. Eur J Clin Invest 2000;

30:553–9.

[4] Schols AMWJ, Buurman WA, Brekel AJS, Dentener MA, Wouters

EFM. Evidence for a relation between metabolic derangements and

increased levels of inflammatory mediators in a subgroup of patients

with chronic obstructive pulmonary disease. Thorax 1996;51:819–24.

[5] Eid A, Ionescu AA, Nixon LS, Lewis-Jenkins V, Matthews S, Grifiths

TL, et al. Inflammatory response and body composition in COPD. Am

J Respir Crit Care Med 2001;164:1414–8.

[6] Ionescu AA, Nixon LS, Routledge PA, Shale DJ. Protein catabolism

in Cystic Fibrosis and COPD. Am J Respir Crit Care Med

2003;167(7):A918 [Abstract presented (poster) at the International

Conference of the ATS, Seattle 2003.].

[7] Ionescu AA, Nixon LS, Luzio S, Lewis-Jenkins V, Evans WD, Stone

MD, et al. Pulmonary function, body composition, and protein

catabolism in adults with cystic fibrosis. Am J Respir Crit Care

Med 2002;165:495–500.

[8] Bohles H, Brendel L, Forster H, Trager K, Vogt J, Georgieff M. The

effect of human growth hormone therapy on l-(methyl-2H3)-leucine

turnover and urinary pseudouridine concentration in patients with

Ullrich–Turner syndrome. Eur J Pediatr 1996;155:275–80.

[9] Sander G, Hulsemann J, Topp H. Protein and RNA turnover in

preterm infants and adults: a comparison based on urinary excretion of

3-methylhistidine and of modified one-way RNA catabolites. Ann

Nutr Metab 1986;30:137–42.

[10] Lu JY, Lai RS, Liang LL, Wang HC, Lin TI. Evaluation of urinary

pseudouridine as a tumour marker in lung cancer. J Formos Med

Assoc 1994;93:25–9.

[11] Tebib JG, Reynaud C, Cedoz JP, Letroublon MC, Niveleau A.

Relationship between urinary excretion of modified nucleosides and

rheumatoid arthritis process. Br J Rheumatol 1997;36:990–5.

[12] Intrieri M, Calcagno G, Oriani G, Pane F, Zarilli F, Cataldo PT, et al.

Pseudouridine and 1-ribosylpyridin-4-one-3-carboxamide serum con-

centrations in human immunodeficiency virus type 1-infected patients

are independent predictors for AIDS progression. J Infect Dis

1996;174:199–203.

[13] Itoh K, Aida S, Ishiwata S, Sasaki S, Ishida N, Mizugaki M. Urinary

excretion patterns of modified nucleosides, pseudouridine and 1-

methyladenosine in healthy individuals. Clin Chim Acta 1993;

217:221–3.

[14] Mitch WE. Insights into the abnormalities of chronic renal disease

attributed to malnutrition. J Am Soc Nephrol 2002;13:S22–7.

[15] Tsujinaka T, Fujita J, Ebisui C, Yano M, Kominami E, Suzuki K,

Tanaka K, Katsume A, Ohsugi Y, Shiozaki H, Monden M. Interleukin

6 receptor antibody inhibits muscle atrophy and modulates proteolytic

systems in interleukin 6 transgenic mice. J Clin Invest 1996;197:

244–9.

[16] Starnes Jr HF, Warren RS, Jeevanandam M, Gabrilove JL, Larchian

W, Oettgen HF, et al. Tumour necrosis factor and the acute

metabolic response to tissue injury in man. J Clin Invest 1988;82:

1321–5.

[17] Kien CL, Zipf WB, Horswill CA, Denne SC, McCoy KS, O’Dorisio

TM. Effects of feeding on protein turnover in healthy children and in

children with cystic fibrosis. Am J Clin Nutr 1996;64:608–14.

[18] Shepherd RW, Holt TL, Johnson LP, Quirk P, Thomas BJ. Leucine

metabolism and body cell mass in cystic fibrosis. Nutrition 1995;

11:138–41.

[19] Li Y, Wang S, Zhong N. Simultaneous determination of Pseudouridine

and creatinine in urine of normal children and patients with leukaemia

by HPLC. Chromatography 1992;6:191–3.

[20] Gehrke CW, Kuo KC, Waalkes TP, Borek E. Patterns of urinary

excretion of modified nucleosides. Cancer Res 1979;39:1150–3.

[21] Nixon LS, Yung B, Bell SC, Elborn JS, Shale DJ. Circulating

immunoreactive Interleukin-6 in cystic fibrosis. Am J Respir Crit Care

Med 1998;157:1764–9.

A.A. Ionescu et al. / Journal of Cystic Fibrosis 3 (2004) 253–258258

[22] Ionescu AA, Nixon LS, Evans WD, Stone MD, Lewis-Jenkins V,

Chatham K, Shale DJ. Bone density, body composition and inflamma-

tory status in Cystic Fibrosis. Am J Respir Crit Care Med 2000;

162:789–94.

[23] Aris RM, Renner JB, Winders AD, Buell HE, Riggs DB, Lester GE,

Ontjes DA. Increased rate of fractures and severe kyphosis; sequelae of

living into adulthood with cystic fibrosis. Ann Intern Med

1998;128:186–93.

[24] Pouw EM, Schols AM, Deutz NE, Wouters EF. Plasma and muscle

amino acid levels in relation to REE and inflammation in stable

COPD. Am J Respir Crit Care Med 1998;158:797–801.

[25] Bell SC, Saunders MJ, Elborn JS, Shale DJ. REE and oxygen

cost of breathing in patients with cystic fibrosis. Thorax 1996;

51:126–31.

[26] Elborn JS, Cordon SM, Western P, Macdonald IA, Shale DJ. TNFa,

resting energy expenditure and cachexia in cystic fibrosis. Clin Sci

1993;85:563–8.