The diagnosis of cystic fibrosis (CF) has evolved over the past decade as newborn screening has become universal in the United States and elsewhere. The heterogeneity of phenotypes associated with CF transmembrane conductance regulator (CFTR) dysfunction and mutations in the CFTR gene has become clearer, ranging from classic pancreatic-insufficient CF to manifestations in only 1 organ system to indeterminate diagnoses identified by newborn screening. The tools available for diagnosis have also expanded. This article reviews the newest diagnostic criteria for CF, newborn screening, prenatal screening and diagnosis, and indeterminate diagnoses in newborn-screened infants and symptomatic adults.
Key points
- •
Most new diagnoses of cystic fibrosis (CF) are now identified by newborn screening, which provides the opportunity to improve outcomes by initiating monitoring and treatments in the presymptomatic period.
- •
The sweat chloride measurement is still the cornerstone of CF diagnosis. It should be performed according to national guidelines. Sweat conductivity and osmolality are not acceptable substitutes.
- •
In the United States, newborn screening algorithms vary by state, although all involve measurement of immunoreactive trypsinogen in the dried blood spot and most involve genetic testing with a panel of CF transmembrane conductance regulator (CFTR) mutations.
- •
CF newborn screening inevitably identifies infants with indeterminate diagnoses (referred to interchangeably as CF-related metabolic syndrome in the United States and CF screen–positive, inconclusive diagnosis in Europe). Although most of these infants do not develop signs or symptoms of CF, they should be monitored regularly at least in the first few years of life, because some will develop evidence of CFTR dysfunction.
Introduction
The landscape of the diagnosis of cystic fibrosis (CF) has changed dramatically over the past decade, as universal screening for CF has become a reality in all 50 states in the United States since 2010. Most countries with a high prevalence of CF around the world have also implemented universal CF newborn screening. Now, instead of being diagnosed based on symptoms, typically after having endured a long, difficult, and expensive diagnostic odyssey, most individuals are diagnosed after a positive newborn screen. In 2013, 62% of new diagnoses in the United States were detected by newborn screening. Early diagnosis affords the opportunity to improve long-term outcomes through close monitoring and appropriate interventions beginning before severe nutritional deficits or irreversible airway damage have occurred. As CF transmembrane conductance regulator (CFTR) modulator therapies that treat the basic defect in CF become available across a broad range of ages and CFTR genotypes, initiation of these therapies in infancy holds the promise of being disease modifying.
However, even in the current era of universal newborn screening, some individuals are diagnosed symptomatically, either because they were born before implementation of newborn screening or in regions in which newborn screening was not offered, or because of a false-negative newborn screen. Thus, clinicians must always maintain an index of suspicion for CF in individuals with 1 or more signs or symptoms of CF, regardless of age or newborn screening results.
Since identification of the CFTR gene on the long arm of chromosome 7 in 1989, it has become increasingly clear that CFTR dysfunction encompasses a wide range of phenotypes, from classic pancreatic-insufficient CF, to single-organ-system manifestations often diagnosed in adulthood, to indeterminate diagnoses in infants identified by newborn screening. Thus, although the diagnosis of CF is straightforward in most cases, with a sweat chloride level greater than or equal to 60 mmol/L and/or 2 CF-causing mutations identified, establishing the diagnosis in the minority of patients in whom these diagnostic conditions are not met can be challenging and time consuming. Newer diagnostic modalities, such as nasal potential difference and intestinal current measurements, can also be used to assess CFTR dysfunction and aid in diagnosis, but are currently only conducted at specialized centers.
Introduction
The landscape of the diagnosis of cystic fibrosis (CF) has changed dramatically over the past decade, as universal screening for CF has become a reality in all 50 states in the United States since 2010. Most countries with a high prevalence of CF around the world have also implemented universal CF newborn screening. Now, instead of being diagnosed based on symptoms, typically after having endured a long, difficult, and expensive diagnostic odyssey, most individuals are diagnosed after a positive newborn screen. In 2013, 62% of new diagnoses in the United States were detected by newborn screening. Early diagnosis affords the opportunity to improve long-term outcomes through close monitoring and appropriate interventions beginning before severe nutritional deficits or irreversible airway damage have occurred. As CF transmembrane conductance regulator (CFTR) modulator therapies that treat the basic defect in CF become available across a broad range of ages and CFTR genotypes, initiation of these therapies in infancy holds the promise of being disease modifying.
However, even in the current era of universal newborn screening, some individuals are diagnosed symptomatically, either because they were born before implementation of newborn screening or in regions in which newborn screening was not offered, or because of a false-negative newborn screen. Thus, clinicians must always maintain an index of suspicion for CF in individuals with 1 or more signs or symptoms of CF, regardless of age or newborn screening results.
Since identification of the CFTR gene on the long arm of chromosome 7 in 1989, it has become increasingly clear that CFTR dysfunction encompasses a wide range of phenotypes, from classic pancreatic-insufficient CF, to single-organ-system manifestations often diagnosed in adulthood, to indeterminate diagnoses in infants identified by newborn screening. Thus, although the diagnosis of CF is straightforward in most cases, with a sweat chloride level greater than or equal to 60 mmol/L and/or 2 CF-causing mutations identified, establishing the diagnosis in the minority of patients in whom these diagnostic conditions are not met can be challenging and time consuming. Newer diagnostic modalities, such as nasal potential difference and intestinal current measurements, can also be used to assess CFTR dysfunction and aid in diagnosis, but are currently only conducted at specialized centers.
Establishing the diagnosis of cystic fibrosis
General Principles
The US CF Foundation has convened 3 panels of experts to establish and then refine the diagnostic criteria for CF ( Box 1 ), first in 1996, then in 2007, and most recently in 2015. As of the writing of this article, the most recent guidelines are still in draft form. In addition, a European consensus conference established a similar diagnostic algorithm. Although understanding of the heterogeneity of disease presentation and of the complexity of CFTR mutations has greatly increased, many of the basic tenets of establishing the diagnosis have remained virtually unchanged. The sweat chloride test remains the cornerstone of diagnosis, because it directly measures CFTR function. Proper performance of the sweat chloride test, which is crucial for the accurate diagnosis of CF, requires skill and experience. The sweat chloride test should be conducted in accordance with established guidelines. It involves transdermal administration of pilocarpine by iontophoresis to stimulate sweat gland secretion, followed by sweat collection into a Macroduct coil, gauze, or filter paper, and analysis of chloride concentration. A sweat chloride level greater than or equal to 60 mmol/L is consistent with a diagnosis of CF. In this era of precision medicine, it is recommended that CFTR mutation analysis also be performed as part of the diagnostic evaluation. Identification of 2 CF-causing mutations in trans is consistent with the diagnosis of CF. Although ∼1800 mutations have been identified in the CFTR gene, to date the minority have been established to be disease causing. The CFTR2 Web site ( http://www.cftr2.org/ ) has the most up-to-date information on CFTR mutations and their phenotypic consequences (see Sosnay PR, Raraigh KS, Gibson RL: Molecular Genetics of CFTR: Genotype and Phenotype , in this issue).
Positive newborn screen
Or signs/symptoms suggestive of CF
Or positive family history in a parent or sibling
And:
Either a sweat chloride level greater than or equal to 60 mmol/L
Or identification of 2 CF-causing mutations in trans
Or nasal potential difference measurement consistent with CF
Diagnosing Cystic Fibrosis in Newborn-screened Infants
It is critical to acknowledge that newborn screening is only a screening test and does not establish the diagnosis of CF. Even infants with 2 CF-causing mutations identified on the dried blood spot need a sweat chloride test to establish the diagnosis, although they may carry a presumptive diagnosis of CF so that initiation of therapies such as pancreatic enzymes is not delayed. Infants with a positive CF newborn screen should be rapidly referred for sweat chloride testing in order to avoid delays in treatment. Sweat chloride testing can be performed in infants more than 2 kg and 10 days of age, and ideally should be performed in the neonatal period (ie, before 30 days of age).
- •
A sweat chloride value greater than or equal to 60 mmol/L in an infant with a positive newborn screen is consistent with a diagnosis of CF. Genetic testing should also be performed to confirm the diagnosis and potentially aid in treatment and discussions of prognosis.
- •
A value less than 30 mmol/L makes CF unlikely, although on rare occasions infants with 2 CF-causing mutations can have a normal sweat chloride value.
- •
A value of 30 to 59 mmol/L suggests possible CF and further testing is required, generally to include extended CFTR mutation analysis.
- ○
If 2 CF-causing mutations are identified, the diagnosis of CF can be established.
- ○
If 0 or 1 CF-causing mutations are identified, the infant is diagnosed with CFTR-related metabolic syndrome (CRMS)/CF screen–positive, inconclusive diagnosis (CFSPID), which is discussed later.
- ○
- •
All infants with a diagnosis of CF or an indeterminate diagnosis should be referred to a specialized CF center for ongoing monitoring and care. Guidelines for the care of infants with CF have been published.
- •
Meconium ileus may produce a false-negative CF newborn screen, but is highly likely to be associated with CF. Therefore, all infants presenting with meconium ileus, regardless of newborn screening result, should carry the presumptive diagnosis of CF until further testing can be accomplished.
Diagnosing Cystic Fibrosis in Symptomatic Individuals
Any child or adult presenting with signs of symptoms of CF ( Box 2 ) or a positive family history should undergo diagnostic testing, regardless of newborn screening results. In general, the sweat chloride test is the initial procedure.
- •
A sweat chloride level greater than or equal to 60 mmol/L is consistent with a diagnosis of CF. A second, confirmatory sweat chloride test is recommended unless 2 CF-causing mutations are identified by genetic testing.
- •
A sweat chloride level less than or equal to 39 mmol/L in individuals more than 6 months of age makes CF unlikely. If clinical suspicion remains high, genetic testing can be performed. Identification of 2 CF-causing mutations if there are symptoms or a positive family history is consistent with the diagnosis of CF.
- •
If the sweat chloride level is in the intermediate range (40–59 mmol/L if aged >6 months and 30–59 mmol/L if aged <6 months), extended genetic testing should be performed, potentially including gene sequencing and evaluation for deletions/duplications.
- ○
If 2 CF-causing mutations are identified, the diagnosis of CF is established.
- ○
Individuals with no or 1 CF-causing mutation and clinical signs or symptoms of CFTR dysfunction may be diagnosed with CFTR-related disorder (CFTR-RD). Sweat chloride testing should be repeated. If the sweat chloride level remains in the intermediate range, a referral should be made to a specialized CF center for further evaluation, which may include expanded genetic testing, lung function testing, chest imaging, respiratory culture, fecal elastase to evaluate exocrine pancreatic function, genital evaluation in boys and specialized tests of CFTR function such as nasal potential difference and intestinal current measurements.
- ○
Nutritional and gastrointestinal:
- •
Nutritional/metabolic: failure to thrive, hypoproteinemia, hypochloremic dehydration, chronic metabolic alkalosis
- •
Intestinal: meconium ileus, rectal prolapse, distal intestinal obstructive syndrome, steatorrhea
- •
Pancreatic: exocrine pancreatic insufficiency, recurrent pancreatitis
- •
Hepatic: protracted neonatal jaundice, biliary cirrhosis
- •
Sinopulmonary:
- •
Chronic wet or productive cough
- •
Bronchiectasis on chest imaging
- •
Respiratory infection with Pseudomonas aeruginosa or other atypical gram-negative organisms
- •
Nasal polyposis in children
- •
Digital clubbing
- •
Allergic bronchopulmonary aspergillosis
- •
Obstructive azoospermia in boys
Cystic fibrosis newborn screening
Rationale for Screening for Cystic Fibrosis: the Impact of Early Diagnosis on Cystic Fibrosis Outcomes
The natural history of CF makes it distinct among disorders that are currently on the Recommended Uniform Screening Panel (RUSP) maintained by the US Secretary of Health and Human Services. Most of the disorders on the RUSP are responsive to treatment in the first weeks of life, resulting in a dramatic change in the natural history of the disorder, avoiding significant morbidity and even mortality. Although it is well known that pancreatic disease leading to malabsorption and malnutrition is present at birth, and lung disease starts early in infancy in CF, a definitive treatment has not been available that would stop disease progression in CF. Nonetheless, epidemiologic and clinical evidence accumulated throughout the 1980s and 1990s such that the Centers for Disease Control and Prevention (CDC) issued a statement in 2004 that newborn screening for CF is justified based on the benefits of early intervention and nutritional management. The strongest evidence of the benefit of early identification through newborn screening that supported the decision by the CDC was in the treatment of early malabsorption and improved growth. Additional reports provided evidence of screened infants experiencing decreased pulmonary complications, opportunities to address early vitamin E deficiency associated with lower cognitive function, and a potential survival advantage associated with an earlier diagnosis, although the CDC decision to recommend CF newborn screening was driven by improved nutritional outcomes. The long-term pulmonary benefits of CF newborn screening remain more controversial.
As newborn screening for CF has become universally adopted in the United States and internationally, studies on the early course of disease and evidence in support of early identification and treatment have become more prevalent. Epidemiologic and clinical studies continue to provide critical information about early growth restriction in CF while providing additional evidence of the early onset of lung disease, showing the need for early therapies before onset of irreversible stunting and lung damage ( Table 1 ). Despite catch-up growth in weight following newborn screening, linear growth impairment occurs during the first year, even among newborn-screened infants, suggesting that earlier and more aggressive nutritional therapy may be necessary to achieve normal growth. Further, it is clear that lung disease is present in the first months of life, and, despite guidelines for early care and more aggressive approaches to treat CF lung disease, the ability to change the early course of lung disease in CF has been limited. Recent studies have shown that CFTR modulators that are targeted toward specific CFTR mutations can improve CFTR function, with resulting improvements in weight and lung function in adults and older children with CF. These revolutionary new drug therapies may be able to prevent the onset of lung disease and other complications of CF if started early in life; however, safety and efficacy studies in younger children must be completed, and the clinical community is proceeding with appropriate caution.
| System | Key Findings |
|---|---|
| Pulmonary | Respiratory infection in infancy is associated with poor pulmonary function later in childhood |
| Lung function is abnormal early in infancy, and early abnormalities are associated with poor lung function at 1 y | |
| Structural lung disease is present in infants and progresses through childhood | |
| Growth and nutrition | Length, but not weight, in infants is stunted at 1 y, despite early interventions |
General Principles of Newborn Screening
Newborn screening is based on the early tenets of Wilson and Jungner that there must be (1) an acceptable treatment or intervention, (2) a cost-effective screening method, and (3) appropriate confirmatory testing. More recently, revised guidelines have been published to help guide decision makers about the inclusion of new disorders on the newborn screening panel in the era of rapid expansion of newborn screening panels caused by multiplex testing and genomic screening. In the United States, the public health system within each state is responsible for implementing newborn screening at the state level. Adoption and implementation of screening for disorders within each state can vary widely because of local resources and legislative and regulatory constraints. Each state also determines the specific processes for follow-up on abnormal results.
History of Newborn Screening for Cystic Fibrosis
Crossley and colleagues from Auckland, New Zealand, established the foundation for CF newborn screening in the 1970s by measuring immunoreactive trypsinogen (IRT) on a dried blood spot specimen using a radioimmunoassay. Screening for CF was adopted internationally through adaptation of this technique, beginning in East Anglia, United Kingdom, and New South Wales, Australia. In 1982, Colorado became the first US state to initiate CF newborn screening, followed shortly thereafter by Wisconsin. The Wisconsin Cystic Fibrosis Neonatal Screening Project was a randomized trial of newborn screening for CF, showing both its feasibility and the long-term efficacy of newborn screening for nutritional outcomes.
Initially, uptake of CF newborn screening was slow internationally as well as within the United States; however, systematic evaluation of the evidence to support the benefit of newborn screening led initially to recommendations for more focused studies, and eventually to the CDC’s statement that newborn screening for CF was justified. Rapid uptake of CF newborn screening followed this recommendation, moving from only 8 states at the time of the recommendation in 2004 to all 50 states by 2010. Similarly, international adoption of CF newborn screening has been widespread, and most Australian, European, and North American newborn screening programs have adopted CF newborn screening.
Cystic Fibrosis Newborn Screening Algorithms
Screening for CF in the United States is implemented at a state level through 4 algorithms. The algorithms and their relative strengths and weaknesses are presented in Fig. 1 . All the algorithms use an increased IRT level from the dried blood spot specimen as the first stage. IRT level is increased in infants with CF presumably because of pancreatic duct dysfunction in both pancreatic-sufficient and pancreatic-insufficient infants. However, increased IRT levels can also be seen in preterm infants, associated with perinatal stress, low Apgar scores, and African American origin. Thus, all states use a multistage algorithm to minimize false-positives. IRT cutoffs are typically set such that 1% to 5% of samples tested are positive at the first stage and move on for further testing; the selection of the cutoff for the IRT concentration is crucial for balancing the sensitivity and positive predictive value of the screen. All CF newborn screening algorithms in the United States have a second stage of testing, and some have 3 stages, all of which are designed to improve the positive predictive value of the screen. Most frequently, the second stage incorporates molecular testing for CFTR mutations on the first specimen (IRT/DNA). In some states, IRT concentrations are analyzed on a second specimen collected between 1 and 2 weeks of life (IRT/IRT), whereas other programs combine the two strategies, following 2 increased IRT results with CFTR mutation analysis performed on the second specimen (IRT/IRT/DNA). The most recent developments in CF algorithms incorporate an extended gene analysis following an increased IRT level and identification of 1 CFTR mutation, allowing a more specific screening result (2 mutations).
