Making the diagnosis of Cystic Fibrosis


In most cases the diagnosis of cystic fibrosis (CF) is clear but paradoxically with increased understanding of the biochemistry of the disease it may be difficult, in a minority of cases, to say definitely if someone has CF or not. We now know that the amount of functional CFTR (chloride channel) on the cell surface is very variable and can range from virtually nothing (affected by CF) to normal (not a carrier) and to almost any level in between. The level in any individual depends on which and how many CF causing mutations are present and possibly on other factors not yet understood. In routine clinical practice it is not possible to directly quantify the amount and efficiency of the CFTR in respiratory epithelial cells. Investigations such as sweat tests, and electrophysiological measurements on nasal and bowel mucosa, semi quantify the efficiency of the CFTR in cells of the sweat gland, nose and bowel respectively. It is difficult to know where to draw a cut off above which someone is labeled as having CF and below which they are reassured that they do not have CF. On occasions different tests give conflicting results and even the same test when repeated may give variable results. Any such cut off is likely to be incorrect in a number of cases. The clinical outcome of CF can vary significantly even for the same genotype (genetic mutation). It is clear that outcome depends on a complex interaction between CF genotype, other putative modifier genes and the environment. Therefore CF should remain a clinical diagnosis (Rosenstein, 2002).

To allow for this heterogeneity (variability) and to avoid inappropriately labelling someone with a life shortening disease it has been suggested that there should be three diagnostic categories; CF unlikely, non classic CF, and classic CF. This acknowledges that it is very difficult to exclude CF but recognises that there are a group of patients who may need long term follow up and possible intervention depending on progression (De Boeck et al, 2006).

The vast majority of individuals with CF present as clear cut cases. Generally such patients have at least one undisputedly abnormal sweat test, two identified CF causing mutations and a phenotype (clinical presentation) compatible with CF (chronic sinopulmonary disease, nutritional abnormalities, salt loss syndromes or male infertility because of obstructive azoospermia).

Presentation of cystic fibrosis

Cystic fibrosis (CF) may present in several different ways and varies with patient age. In this section we provide a brief summary of some of the important clinical features which result in a diagnosis.

Presentation following screening

Over the next three years neonatal screening for CF will be introduced to all of the UK
In the future most patients will present as a result of a positive neonatal screening test. However, it is very important to continue to suspect CF and perform appropriate investigations if a child or adult presents with suggestive symptoms. It is estimated that neonatal screening will miss up to 5% of children born with CF. In addition, older children and adults, who have not had neonatal screening, will continue to present with symptoms for some time.

Presentation following meconium ileus

In 15% to 20% of newborn infants with CF the bowel is blocked by sticky secretions. The baby has signs of intestinal obstruction soon after birth with bilious vomiting, abdominal distension and delay in passing meconium. The obstruction can often be relieved by Gastrografin® enemas but some infants require an operation. The outlook for these infants is now good as a result of the impressive improvements in neonatal surgery, anaesthesia and nutritional support.

Presentation following intestinal malabsorption

Over 90% of individuals with CF have intestinal malabsorption (Morgan et al, 1999). This is usually evident in infancy but may be less apparent in breast fed newborns. The main cause of malabsorption is a severe deficiency of pancreatic enzymes and bicarbonate. There is also evidence that the transport of some other substances across the wall of the intestine is abnormal. Fortunately, treatment with acid-resistant pancreatic extracts will control malabsorption in most patients (Littlewood & Wolfe, 2000; Littlewood et al, 2006).

Presentation with repeated respiratory infections

Most untreated patients with CF have recurrent chest infections, usually from an early age (Armstrong et al, 2005). The viscid (sticky) mucus in the airways predisposes the lungs to bacterial infections that, once established, are difficult to eradicate. The extent and severity of the damage to the bronchial tubes and lungs from these infections is the main factor in determining the physical state and survival of the patient. However, with early intensive antibiotic treatment, particularly by intravenous and nebulised routes, most people with CF will remain free of significant chest problems throughout childhood. Some patients with respiratory symptoms and bronchiectasis attending general respiratory clinics remain undiagnosed and formal screening with sweat tests and genetics should be undertaken in any patient where CF is a differential diagnosis.

Other clinical presentations

Cystic Fibrosis may present in other ways including neonatal or chronic liver disease, nasal polyposis, asthma, rectal prolapse, bowel obstruction, chronic sinusitis, heat exhaustion from salt depletion, male infertility, growth failure or pancreatitis (Masaryk & Achkar, 1983; Conwayet al, 2002).

Presentation of CF in adult life

Cystic Fibrosis is diagnosed in adults with increasing frequency. In Toronto, before and after 1990, 3% and 18% of new diagnoses respectively have been in adults (Gilljam et al, 2004). The spectrum of presentation is different to that seen in children; pancreatitis 4%, genetic screening 9%, pulmonary and gastrointestinal symptoms 22%, infertility 26% and pulmonary symptoms 39% (Gilljam et al, 2004). It is important to recognise that a late diagnosis in an adult patient does not always equate with mild disease (Peckham et al, 2006).

The USA Cystic Fibrosis Foundation consensus criteria for the diagnosis of CF

These criteria assist in the diagnosis of “classic CF” and are applicable to most straightforward cases
One or more of the characteristic phenotypic features

Or a history of CF in a sibling

Or a positive neonatal screening test result

An increased sweat chloride concentration (> 60 mmol/l)

Or identification of two CF mutations

Or demonstration of abnormal nasal epithelial ion transport

The USA Cystic Fibrosis Foundation consensus panel criteria for establishing the diagnosis of CF (Rosenstein & Cutting, 1998).

Difficult diagnostic scenarios

There are a minority of patients who may present in a variety of ways, in whom it proves difficult to confirm, or importantly, to exclude CF. Patients may present with a borderline sweat chloride between 30-60 mmol/l and any combination of the following:

1) Elevated initial IRT on screening
2) Phenotype suggestive of CF
3) One identified CF causing mutation
4) Two mutations of the CFTR gene when there is uncertainty regarding disease causing effect of at least one of them
5) Inconclusive nasal potential difference (NPD) measurements or other electrophysiology
6) CF in a close relative

“Non classic” CF

“Non classic” CF describes patients with a CF phenotype in at least one organ system (table 2) and borderline sweat test results with insufficient evidence from genotype or electrophysiology to support the diagnosis. These patients may have a very different prognosis to patients with “classical CF” but some may develop progressive lung disease as a result of chronic airway infection in adult life (Conway et al, 2002; Peckham et al, 2006; O’Sullivan et al, 2006). Therefore, these patients should have follow-up. Preventative and acute treatment may be necessary depending upon individual circumstances and evidence of the development of lung disease. Care may involve full CF follow-up or only annual review.

&Isolated obstructive azoospermia
Chronic pancreatitis
Allergic bronchopulmonary aspergillosis
Disseminated bronchiectasis
Diffuse panbronchiolitis
Sclerosing cholangitis
Neonatal hypertrypsinogenaemia

WHO List of single organ disease phenotypes associated with CFTR mutations (Joint Working Group of WHO/ICF (M)A/ECFS/ECFTN, 2001)

The sweat test

he sweat test remains central to the diagnosis of CF and it is essential that it is performed by experienced personnel in accordance with national guidelines (National Committee for Clinical Laboratory Standards, 2000; Association for Clinical Biochemistry, 2002). There is controversy as to what should be considered a borderline sweat chloride. Original guidelines suggested a sweat chloride of 40–60 mmol/l, 40 mmol/l representing the mean +2 standard deviations in carriers. Recent evidence suggests that a proportion of patients with “classical CF” have chloride concentrations of 30-60 mmol/l. Sweat chloride concentrations of 30-60 mmol/l are seen in about 4% of sweat tests and 23% of these will subsequently be found to have two CF causing mutations. CF affected patients occur with equal frequency in the 30-40 mmol/l as they do in the 40-60 mmol/l ranges (Lebecque et al, 2002). The sweat test remains the gold standard for confirming the diagnosis of CF and must conform to agreed standards (Littlewood, 1986; Green et al, 2007).
Experienced laboratory personnel should perform the test

It should be repeated at least three times if genotyping does not fully support the diagnosis

A weight of at least 100mg of sweat is required (70mg is accepted in the St James’ laboratory). We consider a sodium and chloride value of >60 mmol/l positive, between 30 and 60 mmol/l equivocal and less than 30 mmol/l negative

If the sweat test is performed in early infancy it has been suggested that the upper limit of sweat chloride should be 40 mmol/L (Farrell et al, 1996; Massie et al, 2000)

Tests with a chloride less than the sodium or a discrepancy between the two of > 20 mmol should be regarded with suspicion

Required standards for sweat tests

The following conditions have been associated with a false positive sweat test result (Rosenstein & Cutting, 1998)
Adrenal insufficiency
Anorexia nervosa
Atopic dermatitis
Autonomic dysfunction
Coeliac disease
Ectodermal dysplasia
Familial cholestasis (Byler’s disease)
G6PD deficiency
Glycogen storage disease type 1
Klinefelter’s syndrome
Mucopolysaccharidosis type 1
Nephrogenic diabetes insipidus
Psychosocial problems


When the delta F508 gene was identified in 1989, clinicians hoped that genetic testing would provide a sensitive and specific diagnostic test. Unfortunately this has not proved to be the case. If a patient has two known CF mutations the diagnosis is confirmed. There are, however, more than 1500 different CFTR mutations and in routine practice it is only possible to check for approximately 100 of these (for an up-to-date register consult the World Wide Web site at It is therefore possible to confirm the diagnosis of CF by means of the genotype but not possible to exclude it. There are some mutations that may not be enough alone to confirm the diagnosis. R117H may fall into this category and further genetic tests may be required (Chmiel et al, 1999) although there is some debate about this. Most DNA diagnostic laboratories will screen for the most common mutations, although there will be significant variation according to the population ethnicity. The frequency of different mutations in the Yorkshire region is shown in table 4.

Prevalence (%)
delta F508
394 395delTT
v rare
v rare
v rare
v rare
3905 3906insT
v rare

Incidence of cystic fibrosis mutations in the Yorkshire population (UK)

Nasal potential difference (NPD) measurements
These tests assess a voltage across the membrane in the nose that correlates with the transport of sodium across cell membranes. The characteristic ion transport observed in the respiratory epithelium of patients with CF differs from the pattern of NPD found in individuals with normal healthy epithelia (Knowles et al, 1995; Delmarco et al, 1997; Wilson et al, 1998). To assist in the evaluation of difficult cases a complete bioelectrical profile should be carried out. This includes basal potential difference, response to perfusion with amiloride and to a chloride free solution in conjunction with isoproterenol and ATP. Similar electrophysiological tests can be performed across small bowel mucosa biopsies (Veeze et al, 1991).

The importance of the national neonatal screening programme

With the widespread introduction of neonatal screening, the majority of cases of CF will present with a positive neonatal screening test and will be found to have two disease causing mutations. We advise that all patients should continue to have a confirmatory sweat test. In those patients in whom it proves difficult to confirm or exclude the diagnosis a category such as “non classical CF” will prove useful. Each of these patients should be considered individually and evidence for respiratory disease and other chronic problems should be sought. Depending on symptoms, examination findings and investigations these patients may start on a CF like management regimen which can be reduced, withdrawn or intensified as determined by the individual’s progress. Various classifications and diagnostic algorithms exist and can be helpful in difficult cases (Bush & Wallis, 2000; De Boeck et al, 2006).

It is difficult to say with absolute confidence that someone with a phenotype compatible with CF does not have CF, so a diagnostic label such as “CF unlikely” is useful. In such patients alternative diagnoses such as Primary Ciliary Dyskinesia, Shwachman-Diamond syndrome, gastro-oesophageal reflux and immunodeficiency should be pursued.

Evaluation of atypical cystic fibrosis

Sweat test and genetics as described above.

Microbiology sputum/cough swab culture/bronchoscopy.

Chest x-ray, CT scan for bronchiectasis, Sinus CT scan.

Pancreatic function, fat soluble vitamin levels, faecal elastase (Cade et al, 2000), faecal fat assessment (Walters et al, 1990), quantitative faecal fat estimation (if available)
CT or MRI examination of pancreas.

Clinical examination of male genitalia, semen analysis (when appropriate), ultrasound examination of testis.

Exclusion of other diagnoses such as ciliary structure and function, immunological deficiency,
g enetics for Shwachman Diamond syndrome, IgE, Aspergillus serology, skin prick tests, Mantoux test, video fluoroscopy, pH probe, barium swallow

Key points

• The diagnosis of CF is straightforward in the majority of patients

• Sweat tests remain central to the diagnosis and need to be performed by experienced laboratory personnel

• The diagnosis “non classical CF” is useful but it is important to recognise that it does not equate with mild disease and long term monitoring is advised

• If there is any doubt it is important to review the diagnosis regularly

• It is essential to keep the patient and family fully informed if any uncertainty exists


Armstrong DS, Grimwood K, Carzino R et al. Lower respiratory tract infection and inflammation in infants with newly diagnosed cystic fibrosis. BMJ 1995; 310: 1571-1572. [PubMed]

Bush A, Wallis C. Time to Think Again: Cystic Fibrosis is Not an “All or None” Disease. Pediatr Pulmonol 2000; 30: 139-144. [PubMed]

Cade A, Walters MP, McGinley N, et al. Evaluation of fecal pancreatic elastase-1 as a measure of pancreatic exocrine function in children with cystic fibrosis. Pediatr Pulmonol 2000; 29: 172-176. [PubMed]

Chmiel JF, Drumm ML, Konstan MW, et al. Pitfall in the use of genotype analysis as the sole diagnostic criterion for cystic fibrosis. Pediatrics 1999; 103: 823-826. [PubMed]

Conway SP, Peckham DG, Chu CE, et al. Cystic fibrosis presenting as acute pancreatitis and obstructive azoospermia in a young adult male with a novel mutation in the CFTR gene. Pediatr Pulmonol 2002; 34: 491-495. [PubMed]

De Boeck K, Wilshanski M, Castellani C, et al. Cystic Fibrosis: terminology and diagnostic algorithms. Thorax 2006; 61: 627-635. [PubMed]

Delmarco A, Pradal U, Cabrini G, et al. Nasal potential difference in cystic fibrosis patients presenting borderline sweat test. Eur Respir J 1997; 10: 1145-1149. [PubMed]

Farrell PM, Koscik RE. Sweat chloride concentrations in infants homozygous or heterozygous for DF508 cystic fibrosis. Pediatrics 1996; 97: 524-528. [PubMed]

Gilljam M, Ellis L, Corey M, et al. Clinical manifestations of cystic fibrosis among patients with diagnosis in adulthood. Chest 2004; 126: 1215-1224. [PubMed]

Green A, Kirk J; Guidelines Development Group. Guidelines for the performance of the sweat test for the diagnosis of cystic fibrosis. Ann Clin Biochem 2007; 44: 25-34. [PubMed]

Joint Working Group of WHO/ICF(M)A/ECFS/ECFTN: Classification of Cystic Fibrosis and related Disorders. Geneva, World health Organisation, 2001.

Knowles MR, Paradiso AM, Boucher RC. In vivo nasal potential difference: techniques and protocols for assessing efficacy of gene transfer in cystic fibrosis. Hum Gene Ther 1995; 6: 445-455. [PubMed]

Lebecque P, Leal T, De Boeck K, et al. Mutations of the cystic fibrosis gene and intermediate sweat chloride levels in children. Am J Respir Crit Care Med 2002; 165: 757-763. [PubMed]

Littlewood JM. The sweat test. Arch Dis Child 1986; 61: 1041-1043. [PubMed]

Littlewood JM, Wolfe SP. Control of malabsorption in cystic fibrosis. Paediatr Drugs 2000; 2; 205-222. [PubMed]

Littlewood JM, Wolfe SP, Conway SP. Diagnosis and treatment of intestinal malabsorption in cystic fibrosis. Pediatr Pulmonol 2006; 41: 35-49. [PubMed]

Masaryk TJ, Achkar E. Pancreatitis as initial presentation of cystic fibrosis in young adults. A report of 2 cases. Dig Dis Sci 1983; 28: 874-878. [PubMed]

Massie J, Gaskin K, Van Asperen P, et al. Sweat testing following newborn screening for cystic fibrosis. Pediatr Pulmonol 2000; 29: 452-456. [PubMed]

Morgan WJ, Butler SM, Johnson CA, et al. Epidemiologic study of cystic fibrosis: design and implementation of a prospective, multicentre, observational study of patients with cystic fibrosis in the U.S. and Canada. Pediatr Pulmonol 1999; 28: 231-241. [PubMed]

National Committee for Clinical Laboratory Standards (NCCLS). Sweat testing: sample collection and quantitative analysis, Approved guidelines C34-A2. Wayne, Pennsylvania: NCCLS, 2000.

O’Sullivan BP, Zwerdling RG, Dorkin HL, et al. Early pulmonary manifestations of cystic fibrosis in children with the DeltaF508/R117H-7T genotype. Pediatrics 2006; 118: 1260-1265. [PubMed]

Peckham D, Conway SP, Morton A, et al. Delayed diagnosis of cystic fibrosis associated with R117H on a background of 7T polythymidine tract at intron 8. J Cystic Fibros 2006; 5: 63-65. [PubMed]

Rosenstein BJ, Cutting GR for the Cystic Fibrosis Consensus Panel. The diagnosis of cystic fibrosis: A consensus statement. J Pediatr 1998; 132: 589-595. [PubMed]

Rosenstein BJ. Cystic fibrosis diagnosis: new dilemmas for an older disorder. Pediatr Pulmonol 2002; 33: 83-84. [PubMed]

Veeze HJ, Sinaasappel M, Bijman J, et al. Ion transport abnormalities in rectal suction biopsies from children with cystic fibrosis. Gastroenterol 1991; 101: 398-403. [PubMed]

Walters MP, Kelleher J, Gilbert J, et al. Clinical monitoring of steatorrhoea in cystic fibrosis. Arch Dis Child 1990; 63: 99-102. [PubMed]

Wilson DC, Ellis L, Zielenski J, et al. Uncertainty in the diagnosis of cystic fibrosis: possible role of in vivo nasal potential difference measurements. J Pediatr 1998; 132: 596-599. [PubMed]