Genotype Analysis and Phenotypic Manifestations of Children With Intermediate Sweat Chloride Test Results

Genotype Analysis and Phenotypic Manifestations of Children With Intermediate Sweat Chloride Test Results* Pascale Desmarquest, MD; Delphine Feldmann, PhD; Aline Tamalat, MD; Michele Boule, MD, PhD; Brigitte Fauroux, MD; Guy Tournier, MD; and Annick Clement, MD, PhD

Study objectives: Cystic fibrosis (CF) is one of the most common inherited diseases among whites. Since the cloning of the CF transmembrane conductance regulator (CFTR) gene, a number of studies have focused on associations between the genotype and phenotype in CF. This had led to the progressive identification of new groups of patients, including those who have mild lung disease and those who have normal sweat chloride values (< 60 mEq/L). The aim of the present work was to provide information on the genotype and the phenotypic characteristics of children with intermediate-range sweat chloride test results. Patients and results: We focused on children referred to the pulmonary department for various types of pulmonary disease and who had several sweat chloride test results with median values in the range of 40 to 60 mEq/L. Twenty-four patients over a 10-year period were enrolled (mean age, 4.8 years). Respiratory manifestations at initial evaluation included recurrent bronchitis, wheezing, chronic cough, and pneumonia. The duration of the follow-up ranged from 0.5 to 10.5 years. Sputum cultures revealed the presence of Haemophilus influenzae (10 children), Staphylococcus aureus (4 children), and Pseudomonas aeruginosa (3 children). Pancreatic insufficiency was found in two patients. Analysis of the entire coding sequence allowed identification of 16 known mutations in CFTR gene. Fifteen chromosomes (31.2%) carried a mutation in CFTR gene and one allele carried two mutations. Three patients were homozygous or double heterozygous (⌬F508/⌬F508, ⌬F508/3849 ⴙ 10 kb C3 T, S1235R/G551D). The 5-thymidine allele was identified in four children. Conclusion: These results indicate an higher frequency of CFTR gene mutations in patients with borderline sweat chloride test results, compared to data reported in the general population. They lead to the recommendations for complete pulmonary and GI investigations in this group of patients, as well as assiduous care and medical follow-up. (CHEST 2000; 118:1591–1597) Key words: children; cystic fibrosis; genotype; sweat chloride tests Abbreviations: cAMP ⫽ cyclic adenosine monophosphate; CF ⫽ cystic fibrosis; CFTR ⫽ CF transmembrane conductance regulator; Cldyn ⫽ dynamic lung compliance; 5T ⫽ 5-thymidine

fibrosis (CF) is one of the most common C ystic inherited diseases among whites. It is caused by 1

mutations in the CF transmembrane conductance regulator (CFTR) gene, which encodes a transmem*From the Departements de Pneumologie Pediatrique-INSERM U515 (Drs. Desmarquest, Tamalat, Boule, Fauroux, Tournier, and Clement), et de Biochimie (Dr. Feldmann), Hopital Trousseau AP-HP, Universite Paris VI, Paris, France. This work was supported by Association Franc¸aise de Lutte contre la Mucoviscidose. Manuscript received August 25, 1999; revision accepted June 20, 2000. Correspondence to: Annick Clement, MD, PhD, Departement de Pneumologie Pediatrique, Hopital Trousseau, 26 av Dr. Netter, 75012 Paris, France; e-mail: [email protected]

brane glycoprotein.2,3 The CF transmembrane conductance regulator has been shown to function as a cyclic adenosine monophosphate (cAMP)-regulated chloride channel at the apical membrane of epithelial cells.4,5 One of the main consequences of mutations in the CFTR gene is a dysfunction of ion channels resulting in elevated sweat chloride concentrations, pancreatic insufficiency, and progressive lung disease.6 –10 Since the discovery of the gene, a number of studies have been focused on associations between the genotype and phenotype in CF.11,12 At the present time, it is well recognized that there is a wide variability in clinical presentations, organ involvement, disease severity, and life expectancy.13 CHEST / 118 / 6 / DECEMBER, 2000

1591

For many years, elevated sweat chloride levels were the “gold standard” for diagnosis of CF.14 Abnormal values were assumed for chloride values ⬎ 60 mEq/L, the 60 mEq/L cutoff discriminating between the populations with CF and without CF.14,15 However, patients have been reported with characteristic manifestations of CF and chloride levels ⬍ 60 mEq/L.16,17 Moreover, in several cases, genetic analyses have documented two mutated alleles in the CFTR gene, and it is likely that such observations will increase in the next years with the identification of more mutations.18,19 Some of these mutations will possibly appear to correlate with a phenotype of mild diseases. Sweat chloride test values ⬍ 60 mEq/L can be separated into two groups: a group of values in the intermediate range of 40 to 60 mEq/L (borderline test results), and a group of values ⬍ 40 mEq/L.20 From the growing number of observations reported in the literature, there is some evidence that patients with borderline test results share more similarities with the CF patients (eg, patients with chloride values ⬎ 60 mEq/L) in terms of clinical presentations and disease progression than the group of patients with chloride values ⬍ 40 mEq/L. The crucial issue of definition of CF raised by the number of observations that do not fulfill the classical criteria explain the efforts that need to be made to provide more information on both the genotype and the phenotype of patients with chloride levels ⬍ 60 mEq/L.21 In the present work, we focused on children who were referred to our pulmonary pediatric department for various pulmonary diseases and whose sweat chloride tests were in the intermediate range of 40 to 60 mEq/L. The aim of the study was to analyze the genotype and the phenotypic characteristics of these patients.

Materials and Methods Patients All the children referred to the pulmonary pediatric department between 1988 and 1998 for various pulmonary diseases and who had sweat chloride tests on more than two occasions with median values in the range of 40 to 60 mEq/L were included in the study. Clinical Evaluation Personal and familial history, mode of presentation, and age at first diagnosis were recorded for each patient. Sweat chloride levels were measured by the standard protocol of Gibson and Cooke.14 The quantitative pilocarpine iontophoresis tests were performed with measurements of sweat weight and chloride concentrations in duplicate, with sweat specimens being collected concurrently from the right and left arm. A minimum 1592

sweat weight of 100 mg was required for analysis. At least four tests were performed for each patient. Height and weight were recorded at each visit, and the weight-to-height ratio representing the actual weight expressed as a percentage of the ideal weight for height was calculated to determine the nutritional status. Pulmonary disease was evaluated by physical examination, chest radiograph, blood gas analysis at rest, and pulmonary function tests. Before 6 years of age, pulmonary function tests included measurements of functional residual capacity, dynamic lung compliance (Cldyn), and total lung resistance. After 6 years of age, FVC and FEV1 were determined. Results of these tests were expressed as a percentage of the predicted values for height-matched children.22,23 Respiratory bacterial infections were evaluated using quantitative sputum cultures, or cultures of tracheal aspirates when sputum samples were not available. Cultures were considered positive when microorganisms were present at a concentration ⬎ 106 cfu/mL. Pancreatic status was determined by the fat content in stool samples collected over 3 days. Patients showing normal results (fecal fat ⬍ 5 g/d) and currently not treated by pancreatic enzyme replacement were defined as pancreatic sufficient, and the remaining were defined as pancreatic insufficient. The levels of blood vitamin A and E were also evaluated.24 DNA Analysis Peripheral blood samples were collected from the children and genomic DNA was extracted by standard methods. For each patient, all 27 exons of the CFTR gene were analyzed by denaturing gradient gel electrophoresis with polymerase chain reaction primers, as described previously.25 Polymerase chain reaction products that displayed an altered behavior in the gel were subsequently sequenced. Mutations were identified by direct DNA sequencing using an automatic DNA sequencer 373A (Applied Biosystems; Foster City, CA). The cryptic splice mutation 3849 ⫹ 10 kb C3 T was detected by restriction analysis according to the instruction of Highsmith et al.18 The 5-thymidine (5T) variant of the intron 8 polythymidine tract was also documented by direct DNA sequencing. Statistical Analysis Results of sweat chloride tests were reported as median (range). Differences between proportions were tested by the ␹2 statistic. All p values were two-tailed, and probabilities of ⬍ 0.05 were considered significant.

Results Clinical Characteristics of the Patients at Initial Evaluation Twenty-four children (13 boys) could be included in the study. Familial history revealed that patient 6 had a brother followed up for classical CF, including positive sweat chloride test values (65 mEq/L), severe lung disease with chronic Pseudomonas aeruginosa colonization, and pancreatic insufficiency. There was no history of meconial ileus in our 24 children. Characteristics of the children when referred to our pulmonary department are given in Table 1. The mean age of the patients was 4.8 years Clinical Investigations

Table 1—Characteristics of Children at Initial Evaluation* Patient No.

Sex

Age at the Initial Evaluation, yr

Sweat Chloride Test, mEq/L median (range)

Respiratory Symptoms and Findings

Pao2, mm Hg

Wt/Ht,† %

Pancreatic Status

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

F M F F F F M M F F M F F M F M M M M M F M M M

9.3 13.5 10 9 4.1 7 0.6 10 4.9 2.4 4.2 0.5 2.2 5 3 5 6.5 2 2.3 4 5 1.2 4 0.5

53 (22–76) 43 (37–52) 40 (33–52) 49 (44–52) 46 (28–69) 49 (37–58) 56 (45–67) 55 (37–53) 52 (32–67) 58 (37–68) 46 (36–60) 59 (56–63) 53 (43–60) 40 (36–54) 40 (31–49) 55 (37–42) 40 (37–43) 44 (20–70) 41 (40–49) 53 (45–60) 47 (46–52) 43 (40–45) 50 (42–58) 50 (48–51)

Wheezing Wheezing Wheezing Wheezing Chronic cough Chronic cough Wheezing Recurrent bronchitis Pneumonia Recurrent bronchitis Chronic cough Recurrent bronchitis Recurrent bronchitis Recurrent pneumonia Pneumonia Chronic cough Wheezing Wheezing Wheezing Pneumonia Recurrent bronchitis Recurrent bronchitis Recurrent bronchitis Chronic cough

98 111 101 112 103 93 97 97 90 89 107 105 101 97 65 90 97 87 ND 104 99 100 ND 100

99 120 85 93 106 96 118 95 100 119 102 79 113 85 86 84 103 97 88 103 95 113 92 97

PS PS PS PS PS PI PS PS PS PS PS PI PS PS PS PS PS PS PS PS PS PS PS PS

*M ⫽ male; F ⫽ female; ND ⫽ not determined; PS ⫽ pancreatic sufficiency; PI ⫽ pancreatic insufficiency. †Weight (Wt) as a percentage of ideal weight for height (Ht).

(range, 6 months to 10 years). The mean sweat chloride level was 48.4 mEq/L (range, 40 to 59 mEq/L). Respiratory symptoms and findings at first evaluation indicated mainly wheezing (eight children), chronic cough (five children) and recurrent bronchitis (seven children). The time elapsed between onset of respiratory symptoms and initial evaluation in the department ranged from 1 month to 5 years (median, 18 months). Results of blood gas analyses performed in 22 children were normal, except for 1 child who had a Pao2 of 65 mm Hg (patient 15). In this patient, hypoxemia progressively lessened and oxygen was discontinued after 12 days of treatment with IV antibiotics, corticosteroids, and inhaled ␤2-agonist. Growth and nutritional status indicated a low relative body weight index in six children, with values between 79% and 89% (values ⬍ 90%). Pancreatic insufficiency was documented in two patients (patients 6 and 12). Clinical Characteristics of the Patients During the Follow-up The duration of follow-up ranged from 0.5 to 10.5 years for 23 children (1 patient was lost to follow-up). Results are summarized in Tables 2, 3. Only three children (patients 9, 20, and 21) recovered from their

Table 2—Characteristics of the Patients During the Follow-up* Patient No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Follow-up, yr

Wt/Ht, %

Pancreatic Status

Sputum Culture

1 1 2 1.5 6 8 4 10.5 2 7.5 1 2.5 4.5 3 1 5 2 2.5 0.5 1 0.5 2 ND 0.5

99 118 96 93 116 100 105 94 109 98 100 76 108 92 93 86 102 94 88 102 95 111 ND 98

PS PS PS PS PS PI PS PS PS PS PS PI PS PS PS PS PS PS PS PS PS PS ND PS

— ND — SA SA, HI HI, SA, PA HI — — HI HI HI, SA, PA HI HI HI, PA — — ND ND HI — — — —

*SA ⫽ S aureus; HI ⫽ H influenzae; PA ⫽ P aeruginosa; see Table 1 for other abbreviations. CHEST / 118 / 6 / DECEMBER, 2000

1593

Table 3—Pulmonary Function Tests*

Molecular Analysis of the CFTR Gene and CF Diagnosis

Patient No.

FRC

FVC

FEV1

Rl

Cldyn

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

91 105 99 96 102 101 97 92 107 90 99 100 115 94 145

93 92 75 89 96 69 — — — — — — — — —

97 71 68 94 98 69 — — — — — — — — —

— — — — — — 158 193 84 108 217 200 99 200 189

— — — — — — 70 36 61 46 42 55 62 55 50

*Data are expressed as percentage of the predicted values. FRC ⫽ functional residual capacity; Rl ⫽ total lung resistance.

initial respiratory manifestations. Eight patients who presented with wheezing at initial evaluation had persistent asthma-like symptoms and received corticosteroids and ␤2-agonist treatment. The follow-up of the other patients indicated repeated episodes of cough and/or bronchitis. Microbiological studies revealed the presence of Haemophilus influenzae in 10 children, Staphylococcus aureus in 4 children, and P aeruginosa (nonmucoid strains) in 3 children. Treatment strategies included chest physiotherapy and oral or IV antibiotics adjusted to the results of bacteriological studies during acute exacerbations of lung disease. Persistent chest radiograph abnormalities (eg, bronchiectasis) were observed in patient 6. The follow-up of growth and nutritional status showed improvement of the relative body weight index in three patients (patients 3, 14, and 15). In three children (patients 12, 16, and 19), the relative body weight remained low. Pancreatic status was not modified during the follow-up. Three children (patients 4, 13, and 14) had low levels of vitamin A and E without pancreatic insufficiency and were supplemented with oral vitamins. The two children with pancreatic insufficiency had a severe form of pulmonary disease with chronic P aeruginosa colonization (patients 6 and 12). Pulmonary function tests could be performed in 15 children after 1 year of followup. Results are listed in Table 3. In the older patients, analysis of FEV1 results showed a moderate obstruction of the airflow for two patients, with values of 71% and 68% (patients 2 and 3, respectively). In the group of young children (⬍ 7 years old), low values of Cldyn (⬍ 75% of predicted values) were observed in all patients. 1594

Of the 24 patients, 12 patients carried one or two mutations in the CFTR gene (Table 4). One patient carried homozygous ⌬F508, and two patients carried compound heterozygous (⌬F508/3849 ⫹ 10 kb C3T, S1235R/G551D). Patient 3 was heterozygous for the mutations R75X and D1270H; however, the familial analysis revealed that the mutations R75X and D1270H were both carried by the paternal allele. Eight patients were heterozygous for a CFTR mutations previously described in CF patients. It should be pointed out that patient 6, who was heterozygous ⌬F508, had the same genotype identified in his brother, and familial analysis confirmed that the two children carried the same alleles. As 12 of the 24 patients (50%) with intermediate sweat chloride test results carried at least one CFTR mutation, the frequency of CFTR mutations in the studied population was significantly higher than in the general population (4%), with t ⫽ 37.5 (p ⬍ 0.001).8 The 5T allele variant of intron 8 was identified in four patients. Of the four patients, two patients carried a CFTR mutation on one chromosome and the 5T on the other chromosome (patients 5 and 22). Patient 9 carried the 5T allele and the mild mutation S1235R on the same chromosome. The frequency of the 5T allele in the population of patients with intermediate sweat chloride test results was not statistically different from the frequency

Table 4 —Genotypic Characteristics Patient No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Genotype

Poly T

⫺/⫺ R117C/⫺ R75X-D1270H/⫺ ⫺/⫺ G91R/⫺ ⌬F508/⫺ ⫺/⫺ ⫺/⫺ S1235R/G551D ⌬F508/⫺

7T/7T 7T/7T 7T/7T 7T/7T 7T/5T 7T/9T 7T/7T 7T/7T 5T/7T 9T/9T 7T/7T 9T/9T 7T/9T 7T/7T 7T/9T 7T/5T 7T/7T 7T/7T 7T/9T 7T/9T 7T/7T 7T/5T 7T/7T 7T/7T

⌬F508/⌬F508 ⌬F508/⫺ ⫺/⫺ ⌬F508/⫺ ⫺/⫺ ⫺/⫺ ⫺/⫺ ⫺/⫺ ⌬F508/⫺ ⫺/⫺ W1282X/⫺ ⫺/⫺ ⌬F508/3849 ⫹ 10 kb C 3 T

Clinical Investigations

reported in the general population (frequency of the 5T allele in the general population, 5.2%).26 Based on the results of DNA analysis and according to the consensus statement on the diagnosis of CF, three patients (patients 9, 12, and 24) met the criteria of both respiratory manifestations and identification of two CF mutations.21 For patient 6, there was a diagnostic dilemma. Although only one mutation has been identified so far and the sweat chloride concentration was 47 mEq/L, a number of characteristic phenotypic features were present, including history of CF in a sibling, chronic pulmonary disease with infection with typical CF pathogens, and pancreatic insufficiency. After many medical discussions, this child was registered in our CF center.

Discussion In the present study, we focused on the phenotypic manifestations and genotypic analysis of children with intermediate sweat chloride test results. Among the patients referred to our pulmonary pediatric department for various respiratory manifestations over a 10-year period, we were able to identify 24 patients who had sweat chloride tests on more than two occasions with median values in the range of 40 to 60 mEq/L. Analysis of the genotype included study of the 27 exons of the CFTR gene, as well as the surrounding intronic sequence 3849 ⫹ 10 kb C3 T and the 5T allele. Results indicated that 15 (of 48) chromosomes had a known mutation in CFTR gene, with 1 chromosome bearing two mutations (R75X and D1270H). It is interesting to point out that two patients were compound heterozygous (⌬F508/ 3849 ⫹ 10 kb C3 T, S1235R/G551D), and one was homozygous ⌬F508. This genotype is usually observed in patients with high values of sweat chloride concentrations. Our results documenting the presence of various CFTR mutations extend the conclusions of several reports in the literature that have analyzed the levels of sweat chloride tests in different groups of CF patients. The hypothesis that sweat chloride concentrations may be directly reflective of CF transmembrane conductance regulator activity and that different functional classes of CFTR mutations would display differences in epithelial chloride conductance and in sweat chloride values is no longer sustained. In a study reviewing 455 patients who had two identified CF mutations, Wilschanski et al27 analyzed the concentrations of sweat chloride levels in CF in relation with the different classes of mutations. They found no differences between patients bearing class I mutations (85 patients), class II

mutations (294 patients), and class III mutations (48 patients), all of these patients except 1 in each group having a phenotype of pancreatic insufficiency. They also found similar sweat chloride levels in the group of class V mutations (11 patients), with all the patients included in this group having a pancreatic sufficient status. The only statistical difference they could document was in the group of patients bearing class IV mutations. The mean sweat chloride level in this group was 95 mmol/L, compared to 104 mmol/L in the group of ⌬F508/⌬F508 patients. This difference should be interpreted with caution, because of the smaller number of patients in the group of class IV mutations (17 patients). Indeed, one would have expected class IV and class V mutations to give similar sweat chloride concentrations: these classes of mutations produce proteins that reach the apical membrane and generate cAMP-regulated apical membrane chloride current, with the only observed difference being a reduction in the amount of current for class IV mutations and a reduced amount of normal protein for class V mutations.28 The 60 mEq/L value of sweat chloride concentrations has been used for a long time to discriminate between the populations of patients with CF and without CF. As for the relation between sweat chloride concentration and CF transmembrane conductance regulator activity, the 60 mEq/L cutoff is questionable. Indeed, from several reports in the literature, it is admitted that normal sweat chloride values do not exclude the diagnosis of CF.18,19 In the present study, one patient was ⌬F508/⌬F508. Similar observations have already been reported by other investigators for various mutations. Stewart et al17 described two patients from two families with the compound heterozygotic CF mutations ⌬F508/ 3849 ⫹ 10 kb C3 T. Phenotypic description indicated that these patients had a mild expression of the disease with sweat chloride concentrations of 28 mEq/L and 46 mEq/L, respectively. Interestingly, investigation of the siblings of one patient revealed sweat test levels of 49 mEq/L and 45 mEq/L, and sputum culture positive for P aeruginosa. In the population included in the present study, a patient also carried the genotype ⌬F508/3849 ⫹ 10 kb C3 T, with a sweat chloride concentration of 50 mEq/L. The 3849 ⫹ 10 kb C3 T mutation was identified by Highsmith et al18 and corresponds to a point mutation in intron 19. This mutation seemed to be associated with milder disease.29 A report by Augarten et al30 of sweat chloride concentrations in patients with this mutation indicated various results less or more than 60 mEq/L. Another interesting finding was the result of the 5T. In our study population, the 5T allele was present in four patients. The 5T allele corresponds to a DNA variant of the CHEST / 118 / 6 / DECEMBER, 2000

1595

intron 8 polypyrimidine tract at the branch acceptor site of exon 9. This variant gives rise to a normal transcript and to a transcript with an in-frame deletion of exon 9, leading to a protein without cAMPactivated chloride conductance activity.26,31–33 Kerem et al26,32 showed that the 5T allele can be associated with various clinical presentation and sweat chloride levels ranging from normal to elevated. Several studies have attempted to correlate genotype and phenotype in CFTR mutations.11–13,34,35 Although it is well admitted that some mutations, such as ⌬F508, are associated with a more severe clinical presentation, expression of pulmonary disease among the patients carrying these mutations varies considerably. The same observation applies to the pancreatic status as well as to the levels of sweat chloride. This extreme variability supports the concept that disease expression may be the result of CFTR mutations in association with additional genetic and/or environmental factors.6,36 Hull and Thomson36 have addressed the question of the contribution of genes other than CFTR to disease severity in a group of CF patients. Based on the current concept of the role of inflammation in CF pathophysiology, they speculated that the proinflammatory cytokine tumor necrosis factor-␣ and the detoxifying enzyme glutathione S-transferase M1 could influence disease severity. Interestingly, they provided data suggesting some relation between alteration in pulmonary function and tumor necrosis factor-␣-308 promoter polymorphism, as well as homozygosity for the null allele of glutathione Stransferase M1. The report of the consensus conference initiated by the CF Foundation in the United States stated that the criteria for the diagnosis of CF should include the following: (1) one or more characteristic phenotypic features, or a history of CF in a sibling, or positive newborn screening test results; and (2) an elevated sweat chloride concentration by pilocarpine iontophoresis (⬎ 60 mmol/L) on two or more occasions, or identification of two CF mutations, or demonstration of abnormal nasal epithelial ion transport.21 According to these criteria, two of our patients fulfilled the two groups of criteria. For one patient, there was a diagnostic dilemma. To follow the recommendations of the authors of the consensus report, it is clear that there is a need to define more precisely the spectrum of CF phenotypic features, as well as to redefine the guidelines for sweat chloride test interpretation. This is sustained by studies indicating a higher incidence of CFTR gene mutation in patients with diffuse bronchiectasis.37 In addition, the growing number of described mutations and the complexity of exploring CFTR 1596

gene expression indicates that the genetic analysis of the CFTR gene for a given patient may never be complete.38 Stewart et al17 discussed the value of nasal transepithelial voltage measurements; however, such measurements are not validated in young children. This leads to the key question: How can we progress to confirm the diagnosis of CF? Considering the difficulties in answering this question, several authors39 suggest introducing the terms of CFTRassociated disease or CFTR-related disease to cover the various expressions of the disease. With similar concern, Farrell and Koscik20 discussed the levels of sweat chloride concentrations that should be considered as normal. They concluded that any sweat chloride value ⬎ 40 mEq/L had a low probability of being a true-normal and, therefore, that a diagnosis of CF is likely for levels in the range of 40 to 60 mEq/L. Data reported in the present work provide support to the conclusion of Farrell and Koscik.20 From the current understanding of the role of CFTR in airway disease, assiduous care and medical follow-up of the patients with intermediate sweat chloride test results should be recommended. These patients should undergo a treatment program according to protocols designed for CF patients, with an aggressive management of pulmonary exacerbations.40 – 44 ACKNOWLEDGMENT: The authors thank Catherine Magnier, Corinne Chauve, and France Laroze for technical assistance. They also thank the Association Franc¸aise de Lutte contre la Muscoviscidose.

References 1 Wood R, Boat T, Doershuk C. Cystic fibrosis. Am Rev Respir Dis 1976; 113:833– 878 2 Quinton P. Physiological basis of cystic fibrosis: a historical perspective. Physiol Rev 1999; 79(1 Suppl):S3–S22 3 Davidson D, Porteous D. The genetics of cystic fibrosis lung disease. Thorax 1998; 53:389 –397 4 Sheppard D, Welsh M. Structure and function of the CFTR chloride channel. Physiol Rev 1999; 79(1 Suppl):S23–S45 5 Schwiebert E, Benos D, Egan M, et al. CFTR is a conductance regulator as well as chloride channel. Physiol Rev 1999; 79(1 Suppl):S145–S166 6 Pilewski J, Frizzell R. Role of CFTR in airway disease. Physiol Rev 1999; 79(1 Suppl):S215–S255 7 Welsh M, Smith A. Molecular mechanisms of CFTR chloride channel dysfunction in cystic fibrosis. Cell 1993; 73:1251– 1254 8 Welsh M, Tsui L-C, Boat T, et al. Cystic fibrosis. In: Scriver CL, Beaudet AL, Sly WS, et al, eds. The metabolic basis of inherited disease. 7th ed. New York, NY: McGraw-Hill, 1995; 3799 –3876 9 Bals R, Weiner D, Wilson J. The innate immune system in cystic fibrosis lung disease. J Clin Invest 1999; 103:303–307 10 Wine J. The genesis of cystic fibrosis lung disease. J Clin Invest 1999; 103:309 –312 11 Kerem E, Corey M, Kerem B, et al. The relation between genotype and phenotype in cystic fibrosis: analysis of the most Clinical Investigations

12 13 14 15 16

17 18

19 20 21 22 23 24

25

26

27

common mutation (⌬F508). N Engl J Med 1990; 323:1517– 1522 Kerem E, Kerem B. Genotype-phenotype correlations in cystic fibrosis. Pediatr Pulmonol 1996; 22:387–395 The Cystic Fibrosis Genotype-Phenotype Consortium. Correlation between genotype and phenotype in patients with cystic fibrosis. N Engl J Med 1993; 329:1308 –1313 Gibson L, Cooke R. A test for concentration of electrolytes in sweat in cystic fibrosis of the pancreas utilizing pilocarpine by iontophoresis. Pediatrics 1959; 23:545–549 LeGrys V. Sweat testing for the diagnosis of cystic fibrosis: practical considerations. J Pediatr 1996; 129:892– 897 Stern R, Bast T, Abramowsky C, et al. Intermediate-range sweat chloride concentration and Pseudomonas bronchitis: a cystic fibrosis variant with preservation of exocrine pancreatic function. JAMA 1974; 239:2676 –2680 Stewart B, Zabner J, Shuber A, et al. Normal sweat chloride values do not exclude the diagnosis of cystic fibrosis. Am J Respir Crit Care Med 1995; 151:899 –903 Highsmith W, Burch L, Zhaoqing Zhou M, et al. A novel mutation in cystic fibrosis gene in patients with pulmonary disease but normal sweat chloride concentrations. N Engl J Med 1994; 331:974 –980 Strong T, Smit L, Turpin S, et al. Cystic fibrosis gene mutation in two sisters with mild disease and normal sweat electrolyte levels. N Engl J Med 1991; 325:1630 –1634 Farrell P, Koscik R. Sweat chloride concentrations in infants homozygous or heterozygous for F508 cystic fibrosis. Pediatrics 1996; 97:524 –528 Rosenstein B, Cutting G. The diagnosis of cystic fibrosis: a consensus statement. J Pediatr 1998; 132:589 –595 Bellon G, So S, Adeleine P, et al. Flow-volume curve in children in health and disease. Bull Eur Physiopathol Respir 1982; 18:705–715 Boule M, Gaultier C, Tournier G, et al. Lung function in children with recurrent bronchitis. Respiration 1979; 38:127– 134 Osika E, Cavaillon J, Chadelat K, et al. Distinct cytokine profiles in sputum from children with cystic fibrosis (CF) and from non CF children with chronic inflammatory airway disease. Eur Respir J 1999; 14:339 –346 Plouvier E, Cougoureux E, Sardet A, et al. CFTR gene analysis in cystic fibrosis patients: detection of 91% of molecular defects and identification of the novel mutation D979V. Ann Genet 1997; 40:185–188 Kerem E, Harel-Rave N, Augarten A, et al. Clinical presentation of patients carrying the 5T thymidine tract in intron 8-from healthy individuals to typical CF [letter]. Pediatr Pulmonol 1995; S12:203 Wilschanski M, Zielenski J, Markiewicz D, et al. Correlation of sweat chloride concentration with classes of the cystic fibrosis transmembrane conductance regulator gene muta-

tions. J Pediatr 1995; 127:705–710 28 Bradbury N. Intracellular CFTR: localization and function. Physiol Rev 1999; 79(1 Suppl):S175–S192 29 Stern R, Doershuk C, Drumm M. 3849⫹10kb C T mutation and disease severity in cystic fibrosis. Lancet 1995; 349:274 – 276 30 Augarten A, Kerem B, Yahav Y, et al. Mild cystic fibrosis and normal or borderline sweat test in patients with the 3849⫹10 kb C T mutation. Lancet 1993; 342:25–26 31 Castellani C, Bonizzato A, Mastella G. CFTR mutations and IVS8 –5T variant in newborns with hypertrypsinaemia and normal sweat test. J Med Genet 1997; 34:297–301 32 Kerem B, Rave-Harel N, Nissim-Rafina M, et al. Variable levels of aberrantly spliced CFTR mRNA transcribed from the 5T allele: the cause for variable disease severity among individuals and between organs of the same individual [letter]. Am J Hum Genet 1995; S57:1412 33 Kerem E, Rave-Harel N, Augarten A, et al. A cystic fibrosis transmembrane conductance regulator splice variant with partial penetrance associated with variable cystic fibrosis presentation. Am J Respir Crit Care Med 1997; 155:1914 – 1920 34 Corey M, Edwards H, Levison H, et al. Longitudinal analysis of pulmonary function decline in patients with cystic fibrosis. J Pediatr 1997; 131:800 – 814 35 Hamosh A, Corey M. Cystic fibrosis genotype-phenotype consortium: correlation between genotype and phenotype in the patients with cystic fibrosis. N Engl J Med 1993; 329: 1308 –1313 36 Hull J, Thomson A. Contribution of genetic factors other than CFTR to disease severity in cystic fibrosis. Thorax 1998; 53:1018 –1021 37 Pignatti P, Bombieri C, Marigo C, et al. Increased incidence of cystic fibrosis gene mutations in adults with disseminated bronchiectasis. Hum Mol Genet 1995; 4:635– 639 38 Kopito R. Biosynthesis and degradation of CFTR. Physiol Rev 1999; 79(1 Suppl):S167–S174 39 Romano L, Padoan R, Romano C. Disseminated bronchiectasis and cystic fibrosis gene mutations. Eur Respir J 1998; 12:998 –999 40 Davis P, Drumm M, Konstan M. State of the art: cystic fibrosis. Am J Respir Crit Care Med 1996; 154:1229 –1256 41 Konstan MW. Treatment of airway inflammation in cystic fibrosis. Curr Opin Pulm Med 1996; 2:452– 456 42 Resnikoff JR, Conrad DJ. Recent advances in the understanding and treatment of cystic fibrosis. Curr Opin Pulm Med 1998; 4:130 –134 43 Weller PH. Implications of early inflammation and infection in cystic fibrosis: a review of new and potential interventions. Pediatr Pulmonol 1997; 24:143–145; 159 –161 44 Zeitlin P. Novel pharmacological therapies for cystic fibrosis. J Clin Invest 1999; 103:447– 452

CHEST / 118 / 6 / DECEMBER, 2000

1597