Principles of Screening Programs

A few principles apply to all screening programs. First, all screening tests are subject to false-positive results because of normal biologic variation, genetic heterogeneity, and human error. Accordingly, all positive screening results must be confirmed by definitive analysis. It is important that all patients who require therapy receive it and, conversely, that patients who do not require therapy not be treated.

Second, all positive results must be considered medical emergencies. Many positive results turn out to be falsely positive, but the concept underlying newborn screening is that identification of the few affected patients is crucial. In addition to the potential tragedy of misdiagnosis of the individual neonate, lack of attention that permits delayed care of a single affected patient can seriously jeopardize the public’s confidence in and the cost-benefit structure of an entire statewide screening program and can compromise the continuation of such programs.

Third, the disorders that are part of newborn screening programs are the result of autosomal recessive traits, which exhibit variable clinical expression even within families. Thus, the siblings of a patient identified by a screening program should be biochemically evaluated for the same disorder because they could be affected although they appear free of symptoms.

Fourth, all patients should be referred to an experienced specialist for definitive diagnosis because these disorders are characterized by clinical and genetic heterogeneity, which can significantly affect care of the patient and genetic counseling for the family.

There is considerable variation in the screening programs of different states in the United States and in various nations. All states and US territories screen for PKU. Until relatively recently, most states performed newborn screening for three to six metabolic disorders (including PKU, homocystinuria, MSUD, and galactosemia), one endocrine disorder (congenital hypothyroidism), and the hemoglobinopathies. The requirements and procedures for the screening programs for congenital hypothyroidism and the hemoglobinopathies are discussed in Chapters 97 and 88, respectively.

Since the 1990s, intensive efforts have been made to expand the scope of newborn screening.58,61,82,127 These efforts started from the premise that the available newborn screening tests were relatively inefficient and not easily generalized to detect new diseases either within the current categories of disease or in new categories of metabolic disease. Testing programs employed separate tests for each disease of interest.

Screening Techniques

Most state screening programs had focused primarily on the classic disorders of amino acid metabolism, which can be evaluated by bacterial inhibition assays. These assays cannot be easily adapted to screen for the more recently described disorders of organic acid metabolism and fatty acid oxidation. Similarly, the standard methods being used to diagnose the organic acidemias and fatty acid oxidation defects—gas chromatography, or combined gas chromatography and mass spectrometry (GC/MS)—could not be upgraded to large-scale newborn screening programs because they require tedious sample preparation and long analysis times.

Tandem mass spectrometry (MS/MS) circumvents these limitations of the bacterial inhibition assay, gas chromatography, and GC/MS methods. In brief, MS/MS permits analysis of a broad range of metabolites in hundreds of blood samples per day. Most states have adopted, or are in the process of adopting, the MS/MS approach to newborn screening as part of their program.

Newborn screening by MS/MS starts, as did the traditional screening programs, by collecting by heel stick a small blood sample and applying it to a standardized paper card. The period for appropriate postpartum collection is 24 to 72 hours in the state of Ohio, and is similar in other states. Samples collected from either premature infants or sick newborns are potentially more difficult to interpret and are subject to greater false-positive and false-negative rates. The blood samples are shipped to a centralized laboratory where a standardized amount of the specimen card is punched out, following which the metabolites of interest are extracted from the punch, subjected to specific chemical modifications to make them compatible for subsequent MS/MS analysis, and automatically introduced into and analyzed by the MS/MS system.

As opposed to traditional screening protocols that required different analytic approaches for each disorder, the current MS/MS techniques permit analysis of a large number of metabolites belonging to a particular category of disease—hence, many disorders—in each sample. Hundreds of samples can be prepared for analysis, analyzed, and interpreted each day. The analysis is performed by state-of-the-art mass spectrometers that permit highly sensitive, accurate, and concurrent identification of multiple metabolites. Computer software permits pattern recognition using several related metabolites, thereby improving the reliability of the testing. In summary, the MS/MS technology is ideally suited for newborn screening of many samples for many possible disorders.

As with traditional newborn screening programs, the current MS/MS screening programs must determine the normal range for the different metabolites they analyze in their system. More importantly, the programs must set cutoffs above or below which they identify a case as at-risk. This is a difficult, ongoing task. Programs that set their cutoffs too high have an unacceptable false-negative rate, and programs that set their cutoffs too low have an unacceptable false-positive rate.

Pilot programs in the United States and elsewhere have demonstrated that MS/MS programs can detect PKU, MSUD, and homocystinuria as well as or better than the traditional screening approaches.58,82,97,127 Nevertheless, the practitioner must still be aware that the MS/MS-based screening programs have similar problems with false-positive and false-negative results as their older counterparts, although they appear to have lower false-positive rates. The practitioner must still determine whether a particular result is truly positive or falsely positive as rapidly and safely as possible. The expanded newborn screening programs have found that approximately 1 in 4000 newborns have an identifiable inborn error of metabolism.

Screening for Disorders

Most states in the United States and many other countries have adopted MS/MS screening to analyze disorders of amino acid metabolism (including several urea cycle disorders), organic acid metabolism, and fatty acid oxidation. The amino acid disorders and urea cycle disorders are detected by analyzing for increased blood concentrations of specific amino acids or combinations of amino acids. Most programs do not screen for disorders that are associated with reduced concentrations of specific amino acids. Similarly, the organic acidemias and fatty acid oxidation disorders are detected by analyzing for increased blood concentrations of specific acylcarnitines, namely, the esters formed between carnitine and the accumulated acids in the various organic acidemias and fatty acid oxidation disorders. As in the case of the amino acid disorders, many programs evaluate samples for combinations of particular acylcarnitines to increase the reliability of their results. Screening for the plasma membrane carnitine uptake defect is an exception to the rule, because it looks for a reduced (rather than increased) concentration of free carnitine.

Table 99-1 lists abnormal laboratory findings, along with the disorders associated with those findings and the additional testing recommended to evaluate the significance of the findings. In the case of the amino acid disorders, a particular abnormal laboratory finding can be associated with more than one disorder because different enzymatic defects can lead to excessive accumulation of that metabolite.

This is also true for the disorders detectable by acylcarnitine analysis, but in addition, there is some ambiguity in identifying several of the acylcarnitines evaluated in the program. For example, C4 (an acylcarnitine that contains an acid group with four carbons) can be either butyrylcarnitine (wherein the four carbons are arranged in a linear pattern) or isobutyrylcarnitine (wherein the four carbons are arranged in a branched pattern). The diseases associated with these two acylcarnitines are quite different, and further studies are required to determine which metabolite, or which disorder, is present.

The confirmatory studies listed in Table 99-1 are readily available in most clinical settings, and discussed in detail later (see Specialized Biochemical Testing). The confirmatory studies cited include tests that have a relatively rapid turnaround time, generally 1 to 2 weeks, hopefully leading to rapid confirmation or elimination of a possible diagnosis. Additional, more refined studies, including specific enzyme analysis, whole cell studies, or genetic mutational analysis, are often required to definitively establish a specific diagnosis, but these generally have a longer turnaround time.

The abnormal laboratory findings listed in Table 99-1 permit the diagnosis of more than 20 genetic disorders, including amino acid disorders that encompass several urea cycle disorders, organic acidemias, and fatty acid oxidation disorders. The list of metabolites provided in Table 99-1 is not comprehensive. Many other metabolites have been identified or can theoretically be identified, but they are not listed because of the rarity or uncertain clinical phenotype of the associated disorder. Not all states test for this particular list of metabolites; some test for fewer and others for more. Practitioners should be familiar with the scope of their local newborn screening program. In any event, the laboratory findings listed in Table 99-1 should provide all practitioners with a foundation for interacting with their local program.

Table 99-2 provides basic information about the disorders cited in Table 99-1, including the name of each disorder along with its common abbreviation (if one is available), the underlying enzymatic defect, the clinical features and natural history, the general approach to treatment, and the prognosis. The frequency of these disorders ranges between approximately 1 in 10,000 for PKU to 1 in 200,000 for MSUD; some disorders have been reported in only a few single-case reports. In addition to their rarity, most of these disorders are characterized by a high degree of clinical variability, making it difficult to provide a succinct but accurate summary. Hopefully the information will provide the practitioner with a reasonable place to start when confronted with a patient who has an abnormal newborn screening result, following which he or she can turn to other resources after a diagnosis is established.

Handling Test Results

The first obligation of the practitioner who receives an abnormal newborn screening report is to inform the parents of the result. The practitioner should explain that the results are provisional and that confirmation is required. The physician must aim for an appropriate balance between his or her own natural desire to reassure the parents that the result might be falsely positive and the desire to instill a sense of appropriate concern in the parents so that they can carry through with appropriate follow-up evaluation. The practitioner’s burden is generally more straightforward when the abnormal metabolite is associated with only a single disorder, but the principles of reassurance and follow-up are the same for metabolites that can be found in more than one disorder (see Table 99-1). The physician should then assess the newborn’s clinical status and arrange to see the family as expeditiously as possible.

The primary physician should either see the patient or refer the patient to a metabolic disorders specialist for further evaluation and care. It is often best that the primary physician see the patient as soon as possible to assess the patient’s status and then work with the metabolic disorders specialist to develop an expeditious plan for evaluation. Confirmatory testing should be initiated as soon as possible.

The decisions about when to initiate treatment and how to treat are based on the nature of the laboratory abnormality found, the quantitative degree of the abnormality, the program’s prior experience with false-positives for that metabolite, and the patient’s clinical status. In general, starting a treatment immediately after the initial confirmatory studies are initiated is both safe and unlikely to compromise the ability to establish a diagnosis. However, this option is predicated on the ability of the physician to make certain that the diagnostic samples are collected properly, sent to the appropriate laboratory, and received in satisfactory condition by the laboratory. Failure to do this before starting treatment might significantly delay the time required to establish a diagnosis and initiate appropriate treatment.

Most of the disorders identified are treated, at least in part, by some form of dietary restriction. A family’s desire to continue breastfeeding while the diagnostic studies are in progress should be carefully considered in all cases. However, depending on the disorder under consideration and the patient’s clinical status, the default position should be in favor of stopping, or at least interrupting, breastfeeding until a provisional diagnosis has been established. It is generally a matter of days to a week before the results of the initial confirmatory studies are available, when a more definitive decision can be made about the advisability of breastfeeding. Similar reasoning should be exercised about starting vitamin or cofactor supplementation.

Effect of Screening Programs

The impact of the expanded newborn screening programs is still being determined. There have clearly been many instances when the programs have led to the early recognition of an as-yet-unaffected newborn, followed by the introduction of appropriate treatment. In some cases, this has meant that a newborn with one of the organic acidemias or urea cycle defects that can manifest with an acute neurologic intoxication syndrome in the first few days of life does not suffer an insult that produces severe, irreversible neurologic damage. In other cases, the newborn screening result becomes available after a newborn is already ill, but the result provides a rapid diagnosis for the illness and leads to earlier introduction of appropriate therapy, thereby improving the patient’s outcome.

However, it is not yet clear whether early recognition and institution of appropriate treatment changes the long-term prognosis for many of these diseases, such as recurrent hyperammonemic crises in the urea cycle defects or renal failure in methylmalonic acidemia. There may also be negative consequences to these new programs. For example, the screening programs could produce undesirable effects on the family of a child with a false-positive result, including increased hospitalization of the child, parental stress, and parent-child dysfunction.¹²² Carefully organized multicenter studies are needed to determine the long-term benefits of the expanded newborn programs.

In addition to the current MS/MS newborn screening programs for amino acid and acylcarnitine analysis, new methods for evaluating other groups of inborn errors of metabolism, including the lysosomal storage disorders, are under development. It seems reasonable to anticipate that many of these methods will be introduced over the next several years, further expanding the responsibility and role of the pediatrician and neonatologist in caring for children with metabolic disorders.

Separate summaries of several disorders that were part of the traditional screening programs and that are now evaluated by MS/MS programs (e.g., homocystinuria, MSUD, PKU) are provided next because they are useful paradigms for understanding the benefit of the newborn screening programs and how they work.⁴⁸ A summary is also provided for MCAD deficiency because it is the most common of the fatty acid oxidation disorders that are now evaluated by MS/MS programs, and it is one of the paradigms for this group of disorders. Summaries are also provided for biotinidase deficiency and galactosemia, which are disorders that are primarily evaluated by methodologies other than MS/MS. Other disorders that are now part of expanded newborn screening programs are discussed elsewhere in this chapter, including fatty acid β-oxidation disorders (see Hypoglycemia), nonketotic hyperglycinemia (see Metabolic Seizures), organic acidemias (see Metabolic Acidosis), tyrosinemia type I (see Hepatic Dysfunction), and urea cycle defects (see Hyperammonemia).

Galactosemia

Classic galactosemia is the consequence of galactose-1-phosphate uridyltransferase (GALT) deficiency. Classic galactosemia can manifest in the newborn period with lethargy, poor feeding, jaundice, cataracts, and in some cases, Escherichia coli sepsis.23,48 If unrecognized, this disorder can lead to early death or a chronic course characterized by cirrhosis, cataracts, seizures, and mental retardation. The mainstay of therapy for classic galactosemia is strict dietary lactose restriction.^23,48 Diet therapy is difficult to sustain because lactose is a ubiquitous food additive. Dietary galactose restriction should be started as early as possible (preferably within the first few days after birth) to have the best chance of precluding the development of speech and learning problems. However, even children treated early often have mild growth failure, learning disabilities, and verbal dyspraxia. Affected girls almost invariably develop premature ovarian failure.^39,98 This observation serves as a caution to those caring for children with galactosemia that long-term follow-up is mandatory and further improvements in treatment are required.

There are two other forms of galactosemia: uridine diphosphate galactose-4′-epimerase deficiency and galactokinase deficiency. In most cases, epimerase deficiency is a benign condition that does not require treatment. The rarer, systemic form of epimerase deficiency produces a clinical picture similar to classic galactosemia. Galactokinase deficiency is also rare, and produces nuclear cataracts but none of the other manifestations of classic galactosemia. Early recognition and treatment of this disorder are generally successful.

One approach to screening measures GALT activity. This assay can detect transferase deficiency without regard to prior dietary intake of galactose. It does not evaluate for either epimerase deficiency or galactokinase deficiency. Another approach is to measure galactose and galactose-1-phosphate (the substrate for GALT), which depend on prior dietary galactose intake, and evaluate for all three enzyme deficiencies. Most US states use a combination of these approaches. Because of the rapid onset of symptoms of classic galactosemia and the presence of lactose in breast milk and most artificial formulas, screening programs for galactosemia must provide rapid results. However, the screening results are not always available before the affected neonate becomes ill; initial evaluation of a sick newborn should, therefore, include testing for the presence of urinary reducing substances (see Specialized Biochemical Testing). A newborn identified by newborn screen as possibly having classic galactosemia should have definitive biochemical testing by measuring whole blood or erythrocyte GALT activity and erythrocyte red cell galactose-1-phosphate. In addition, genetic analysis for the common GALT mutations is often helpful in interpreting the results of the GALT activity measurements and making treatment decisions. Following initiation of these studies, lactose should be withdrawn from the diet pending results of the laboratory investigations.

Widespread neonatal testing of erythrocyte transferase activity in various populations has revealed considerable genetic heterogeneity of this enzyme deficiency.²³ Some individuals have a partial enzyme deficiency that does not result in significant impairment of galactose metabolism or any discernible clinical disorder; there is no evidence of a need for dietary treatment of these cases. In other cases with partial enzyme activity, erythrocyte galactose-1-phosphate concentrations are increased, and minimal symptoms can develop. These cases can be managed with less severe restriction of dietary lactose intake.