Rudolph's Pediatrics, 22nd Ed.

CHAPTER 12. Screening

Ada M. Fenick

PRINCIPLES OF SCREENING

Much of the history and physical examination obtained at each health supervision visit is directed toward the identification of undetected problems or their risk factors in an effort at secondary prevention of undesirable outcomes. Screening for these conditions implies the presumptive identification of disease in an asymptomatic individual before it becomes clinically evident. Screening is not diagnostic; a patient with a positive screening test must undergo further evaluation for definitive diagnosis.¹ Pediatricians must be aware not only of current recommendations regarding screening and the specific tests available but also of the basic principles and concepts behind screening in order to evaluate whether a given program does more good than harm for their particular patients and community.

Screening assumes that identified persons will undergo definitive diagnostic testing and will subsequently benefit by earlier implementation of treatment or prevention programs. In deciding what conditions are worth screening for, the clinician must consider the following:

• Is the disease common in the population and serious enough to warrant screening?

• Is there an acceptable treatment for the patients who have the condition?

• Will early diagnosis favorably influence the outcome?

• Is there a good screening test available?²

The effectiveness of a given screening program can be demonstrated by performing a randomized clinical trial in which all pertinent outcomes are evaluated. Unfortunately, such data are often lacking or difficult to obtain. In the absence of such studies, the value of a given screening program must be defined in relation to the characteristics of the condition being screened for, the test being used, the population being evaluated, and the larger social context in which decisions regarding the value of detection and the allocation of resources are being made.

Identification of conditions for which no treatment exists or for which the benefit of existing therapy is unproven, is of questionable value, or is potentially harmful. Even if an effective intervention exists, the clinician must weigh the potential risks and benefits of the treatment itself with that of the identified condition and consider the impact of public acceptance on compliance with recommendations about screening and treatment.²

The costs associated with a screening program must be broadly defined. Costs include not only the screening itself but also the subsequently required diagnostic, therapeutic, and supportive services.² The psychological impact on individuals identified as false positives and the costs involved in definitive evaluation of these individuals may be significant and should not be underestimated.^3,4 (See also Chapter 98.)

Early identification through screening does not always imply a better outcome. If the health care system or community is unable to provide the necessary subsequent diagnostic and therapeutic services, the ultimate value of the screening program is questionable. In addition, if persons at greatest risk do not avail themselves of the screening program, or if individuals with abnormal screening tests do not follow through with subsequent diagnostic and therapeutic recommendations, the screening program will fail to achieve the benefits intended.

Besides not being harmful to the individual being tested, a good screening test must also be accurate. The accuracy of a test is described by its sensitivity and specificity when compared with gold-standard measures of the presence or absence of disease and by its positive and negative predictive value within a population with a given disease prevalence.² It is important to understand how these test characteristics affect the overall value of screening programs and strategies for their implementation.

If the population on which the test was standardized is sufficiently similar to the group to be screened, then measures of predictive ability will be comparable and the application of the instrument is appropriate. Clinically, the positive predictive value of a test may be of paramount importance to a patient and provider; the provider must know the prevalence of the condition in his practice in order to effectively interpret the result of the screen for his patient. Independent of the sensitivity and specificity of the screening test, diminishing population prevalence of the condition being sought diminishes the positive predictive value of the test by changing the proportion of true-positive to false-positive results. For most screening situations, it is important to know the predictive ability of the test in a population with low disease prevalence, because this is generally how screening tests are used. In many cases, selective testing of high-risk subgroups may make more sense than mass screening.²

Current recommendations regarding screening during routine health supervision visits reflect an increasing awareness of the importance of these issues in deciding the value of specific screening programs. Some screening programs are applicable universally because of a high prevalence in the population, while some screening should be done on the basis of selective factors; different strategies may be appropriate for different populations.^5,6 The following section addresses specific recommendations for screening during the health supervision visit. Areas of health screening unique to adolescents are discussed in Chapter 67.

SPECIFIC SCREENING AREAS

THE PHYSICAL EXAMINATION

During routine health surveillance visits, a physical examination should be performed for diagnostic and case-finding (screening) purposes; it also provides a useful framework for parent and child education and reassurance.⁶ Height, weight, head circumference (ages 0 to 3 years), and body mass index (over age 2) should be monitored sequentially as part of the physical examination as detailed in Chapters 10 and 28.⁵

NEWBORNS

The American College of Medical Genetics recommends that all states screen in the early neonatal period for a core panel of 29 metabolic diseases and hemoglobinopathies, with another 25 disorders recommended for inclusion either because of the benefit of early diagnosis or because of their potential for confusion with the core panel disorders.⁷ All states in the United States have initiated neonatal screening programs, and all screen for congenital hypothyroidism, phenylketonuria (PKU), and galactosemia, but the absence of federal guidelines has led to considerable state-to-state variability in other measures.⁸ The Maternal and Child Health Bureau has recently begun to move toward standardization of outcomes and guidelines for state newborn screening programs.⁷

To minimize the number of infants inadvertently missed by the screening program, a blood sample should be obtained on all full-term neonates just before hospital discharge. In no case should this be obtained later than 7 days of age.⁹ Special testing arrangements must be made if birthing takes place in a nontraditional setting. Identification of some disorders, such as phenylketonuria, requires sufficient buildup of metabolites to be detected; thus, if blood was drawn before the infant was 24 hours old (eg, due to early discharge), a second sample should be obtained when the child is 1 to 2 weeks old. Blood transfusions and dialysis, by introducing foreign blood cells and reducing concentrations of circulating metabolites, may result in both false-negative and false-positive results when newborns are screened for metabolic disorders and hemoglobinopathies.³ When feasible, samples should be obtained before these procedures. However, preterm and sick infants should be screened by 1 week of age regardless of the presence or absence of these or other factors (parenteral feeding, antibiotic use, prematurity) that may interfere with specific assays or the interpretation of test results. Where such concerns exist, a repeat sample should be obtained at a time interval appropriate to resolution of the confounding factors. Because of variability in disease presentation and the technical aspects of screening, some affected infants may test falsely normal on their initial screen. Therefore, regardless of the results of the newborn screening, specific diagnostic testing should always be performed when clinical suspicions warrant.

Because of the rapid pace of change regarding newborn screening, and because the disorders screened for are rare, the Committee on Genetics of the American Academy of Pediatrics periodically issues updated information for physicians regarding currently available tests and screening recommendations.¹⁰ Because the choice of screening test, threshold values, and implementation strategies vary in different states and countries, providers should be familiar with the methodology, standards, and follow-up procedures for their regional screening program.³

DEVELOPMENT

See Chapter 82, “Standardized Screening and Assessment Instruments.”

Vision Screening

Routine vision screening is an effective way to identify otherwise unsuspected problems that are amenable to correction. Because normal visual development depends on the brain’s receipt of clear binocular visual stimulation, and because the plasticity of the developing visual system is time-limited, early detection and treatment of a variety of problems impairing vision are essential to preventing permanent and irreversible visual deficits. An age-appropriate assessment should be incorporated into each health supervision visit beginning with the newborn examination. At all ages, the examination should include a review of relevant historical information regarding visual concerns and family history, gross inspection of the eye and surrounding structures, observation of pupillary symmetry and reactivity, assessment of ocular movements, elicitation of the red reflex (to detect opacities and asymmetries in the visual axis), and age-appropriate methods to assess ocular preference and alignment and visual acuity. A successful funduscopic examination should be attempted beginning at 3 years of age and can generally be accomplished by 5 years of age.¹³

It is especially important to assess the red reflex during the neonatal period. Identification of an absent, defective, or asymmetric red reflex should lead immediately to either a repeat examination following pupillary dilation or referral to an experienced pediatric ophthalmologist for definitive diagnosis.

In the infant, ocular preference and alignment and visual acuity can be grossly assessed by observing the baby’s ability to visually track an object, noting any behavioral cues of an eye preference by alternately covering each eye while presenting an interesting object and observing the position and symmetry of the light reflected off the corneas when a light is held several feet in front of the eyes (corneal light reflex).¹³ Ocular alignment (conjugate gaze) should be consistently present by 4 months of age; a child with ocular deviation by history or examination after this time should be referred for evaluation by an ophthalmologist.¹⁵

The toddler and preschooler should have ocular preference and alignment assessed via the corneal light reflex, but at this age, the cross cover test should be utilized as well. This test involves covering and uncovering each eye while the child is looking straight ahead at an object approximately 10 feet away. The observation of any movement of the uncovered eye when the opposite is covered or of the covered eye when the occluder is removed suggests potential ocular misalignment (strabismus). By 3 to 5 years of age, stereoscopic vision can also be assessed using the random-dot-E stereotest or stereoscopic screening machines. A positive test via any of these modalities warrants referral to an ophthalmologist for further evaluation.¹³ Regardless of the underlying etiology, strabismus that is left untreated will eventually result in cortical suppression of visual input from the nondominant eye and the absence of depth perception, making early detection and treatment critical.

Formal visual acuity testing should begin at 3 years of age using age-appropriate methods.¹³ Approximately 5% to 10% of all preschoolers have refractive errors.¹⁶ While picture tests such as the Lea Hyvärinen (LH) test and Allen picture cards are most effective for screening preschoolers, by 5 years of age, most children can be successfully screened using a wall chart with the standard Snellen alphabet chart, the tumbling-E test, or the HOTV test. School-aged children, including adolescents, should have their acuity checked yearly. Preschoolers should be referred for further testing if the acuity in either eye is 20/40 or worse. In children over age 6, inability to read the majority of a 20/30 line warrants referral. At all ages, a difference of more than one line in the acuity measurements between eyes necessitates further evaluation.¹³

For more detailed discussions of the approach to office evaluations of the eyes and tests of vision, refer to Chapters 580 and 581.

Hearing Screening

Approximately 1 to 3 of every 1000 infants are born deaf, and many children develop hearing deficits during childhood.¹⁷ Timely detection of these problems allows for earlier initiation of interventions aimed at enhancing the communication, social, and educational skills of these children. When deaf or hard-of-hearing children are identified earlier than 6 months of age, they perform as much as 20 to 40 percentile points higher on school-related testing.¹⁸

Hearing of all infants should be screened by 1 month of age.¹⁸ In the United States, 31 states require newborn hearing screening, while 17 more offer it to all newborns.⁸ At this time, otoacoustic emissions (OAE) and automated auditory brainstem response (AABR) testing are available for testing in the newborn period. In both these modalities, a series of stimuli is presented to the infant via a probe in the ear canal. However, OAE measures cochlear responses to an acoustic stimulus via a probe in the canal, while AABR measures neural activity in response to a series of acoustic stimuli via surface electrodes. While both modalities can detect sensory hearing loss, only the AABR may detect hearing loss associated with neural dysfunction. AABR can be used to retest a child who has not passed the OAE, as it can verify an intact pathway. However, the opposite is not true; a child who has not passed the AABR because of neural dysfunction may have an intact sensory system, pass an OAE examination, and not receive appropriate referral for the root cause. Any child who does not pass a newborn hearing screen should receive a comprehensive evaluation by 3 months of age by an audiologist who has the necessary expertise and equipment.¹⁸

Because not all hearing loss will be uncovered by newborn screening, and because some hearing loss is later in onset, all children with risk factors listed in Table 12-1 should have their hearing evaluated at least once by 2 to 2½ years of age.¹⁸ In addition to performing a gross hearing assessment and inquiring about hearing concerns at each well-child visit, the American Academy of Pediatrics endorses a policy of formal hearing screening for all children at 4, 5, 6, 8, and 10 years old.⁵ A variety of transient conditions as well as testing problems can affect the hearing evaluation of older, otherwise healthy children, and so the results of audiologic screening must be interpreted within the context of the child’s ear-disease history and physical findings.

For further discussion of hearing evaluation see Chapter 369.

BLOOD PRESSURE SCREENING

Routine blood pressure screening during the well-child visit allows for the identification and potential treatment of children with persistently elevated blood pressure who are at increased risk for hypertension and its subsequent complications as adults. In a minority of patients, an underlying medical etiology may be found. Screening also provides an opportunity to evaluate and potentially modify additional cardiovascular risk factors and to provide education regarding prudent dietary and lifestyle choices.

Table 12-1. Risk Indicators Associated with Hearing Loss

*These risk indicators are of greater concern for delayed-onset hearing loss.

Source: Joint Committee on Infant Hearing. Year 2007 position statement: Principles and guidelines for early hearing detection and intervention programs. Pediatrics. 2007;120:898-921.

Blood pressure standards vary with age, gender, and height.¹⁹ Routine blood pressure screening at least once a year is recommended for all otherwise well children 3 years of age and older.^5,6 Blood pressure measurements should also be taken in ill and potentially symptomatic children as well as in children younger than 3 years who are believed to be at increased risk for hypertension because of coexisting medical conditions (see Table 12-2).¹⁹

Table 12-2. When to Measure Blood Pressure in Children Under 3 Years

Source: National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics. 2004;114:555-576.

Normal blood pressure is defined as systolic and diastolic readings less than the 90th percentile for age and sex. Prehypertension (formerly known as high-normal) and hypertension are defined, respectively, as systolic and/or diastolic readings between the 90th and 95th percentiles and greater than or equal to the 95th percentile for age and sex, found on at least 3 separate occasions. Children with persistently elevated blood pressure readings (> 90th percentile) warrant a thorough history and physical examination to identify underlying causal factors, end-organ damage, and concomitant cardiovascular risk factors, as well as a long-term surveillance and/or treatment plan¹⁹ (see Chapter 479).

ORAL HEALTH

In addition to providing primary prevention via anticipatory guidance, referral to a dental home, and fluoride supplementation if necessary, the pediatrician should perform a dental screening examination of children aged 1 to 4 as part of an oral health risk assessment.⁶ See Chapter 13 and Section 20 for more detailed information.

CHOLESTEROL AND LIPIDS

Epidemiologic data support the hypothesis that atherosclerosis and coronary heart disease have their precursors in childhood and that identifiable risk factors such as hypertension, obesity, and hyperlipidemia are associated with an increased incidence of atherosclerotic disease.²⁰ Serum cholesterol as well as other cardiovascular risk factors can be influenced significantly by dietary and life-style choices, and although long-term pediatric data are lacking regarding the risks and benefits of following prudent life-style recommendations during childhood, until more definitive information is available, it seems reasonable that pediatricians should provide primary preventive counseling to all their patients and families regarding these areas.

Because of the current paucity of information regarding the risks and benefits of treatment for hyperlipidemia in childhood, the costs and limitations of available screening tests, and the potential benefit of promoting healthy life-style and dietary choices to all families, the American Academy of Pediatrics and the American Heart Association do not support universal cholesterol screening for children.²⁰ Rather, children with a body mass index over the 85th percentile should be screened, as should children who are at high risk for hyperlipidemia (see Table 12-3).^20,21

LEAD

Although the use of lead-based paint was effectively banned in the 1970s, ingestion of lead-containing paint chips and dust created by the deterioration or renovation of older homes remains the primary source of lead contamination in children in the United States. Considerable attention has been focused on this issue recently because of a growing body of evidence that suggests an association between subtle neurobehavioral effects and blood lead levels previously felt to be innocuous.^22,23

In the United States, screening is recommended for children at highest risk. All children receiving or eligible for Medicaid should be routinely screened at 9 to 12 and 24 months of age.²² In addition, local health departments may make recommendations, based on suggestions from the Centers for Disease Control or on local data, to either screen all children in their jurisdiction universally or to not do so. Because there are no well-validated written risk assessments available, it is additionally recommended that screening be performed for all children who are not Medicaid eligible and who live in an area with no local health department screening recommendations.²² Blood lead level testing should be performed at 1 and 2 years of age and at 36 to 72 months of age among children who have not previously been screened. Because levels usually peak at about 18 to 24 months of age, a single screening at 12 months of age is not sufficient. If either of these levels are greater than 10 μg/dL, follow-up with additional testing is indicated. Additional screening is indicated for any child up to 6 years of age with suspected increased risk of exposure and for children with developmental delays, particularly if they exhibit pica. Because of the increased potential for contamination from environmental sources, elevated values obtained from capillary specimens should be confirmed using venous blood testing.²⁴ The approach to management of elevated lead levels is provided in Chapter 17.

Table 12-3. Cardiovascular Disease Stratification by Risk

Table 12-4. Conditions Prompting Evaluation via Urinalysis with Microscopy

Source: Bock GH. Screening UA should be based on specific conditions. AAP News. 2006;27:18.

IRON-DEFICIENCY ANEMIA

Primary prevention of iron-deficiency anemia in children is largely accomplished via dietary recommendations, such as the early introduction of iron-containing complementary foods to infants.^25,26 However, iron deficiency is still a relatively common condition, especially in high-risk groups. Analysis of the fourth National Health and Nutrition Examination Survey, which collected data from 1999 to 2002, showed an overall prevalence of iron deficiency in 8% of all 1-year-old to 3-year-old children.²⁷

Current recommendations for screening involve both universal and selective measures. All infants should be screened once, via measurement of hemoglobin or hematocrit, between 9 and 12 months of age.^6,25,26 In addition, prematurity, low birth weight, early introduction of cow’s milk, and use of low-iron formula are risk factors that should prompt an assessment at 4 months.^6,25 Preschool children who live in poverty, have limited access to food, have a diet low in iron, or are at risk of iron deficiency because of special health needs should be screened at 18 months and 2, 3, 4, and 5 years of age. School-aged children who consume a strict vegetarian diet and are not receiving an iron supplement should be screened.^6,25 Lastly, all nonpregnant women in their childbearing years should be screened every 5 to 10 years; this applies equally to adolescent females.^6,26

Because of the frequent occurrence of mild transient anemia with acute illness, hemoglobin screening should not be done while the child is ill or within several weeks of a fever or infection. Although measurements obtained by skin puncture can be used for initial screening, measurements obtained by venipuncture are more accurate and reproducible and should be used for confirmation.²⁵ Abnormally low values are defined as being more than 2 standard deviations below the mean for children of similar age and same sex (see Chapters 431, “Anemia: Definition, Pathophysiology, and Classification,” and 432, “Iron Deficiency”).

URINALYSES

Although they are frequently obtained, many studies have shown that in the absence of clinical concerns or risk factors, routine surveillance urinalyses are not cost effective. Universal screening rarely leads to detection of significant asymptomatic renal disease,²⁸ and when it does, one must ask whether early detection benefits the patient more than diagnosis with the onset of symptoms. The relatively frequent occurrence of minor abnormalities, such as microscopic proteinuria, are of questionable significance but, along with contaminated culture specimens, often necessitate costly and inconvenient repeat studies. The 2007 American Academy of Pediatrics’ recommendations for preventive care have eliminated the former recommendations for urinalysis.^5,6 Urine studies should be obtained when disease is suspected or when the child is at increased risk for specific renal problems. The Expert Committee on Nephrology of the American Academy of Pediatrics has published risk factors that should prompt the pediatric provider to obtain a urinalysis, including microscopy (see Table 12-4).²⁹ (See also Chapter 468.)

Table 12-5. Questions for Determining Risk of Latent Tuberculosis Infection

¹ These definitions apply regardless of previous bacille Calmette-Guérin immunization; erythema at tuberculin skin test site does not indicate a positive test result. Tests should be read at 48 to 72 hours after placement.

² Evidence by physical examination or laboratory assessment that would include tuberculosis in the working differential diagnosis (eg, meningitis).

³ Including immunosuppressive doses of corticosteroids.

Source: American Academy of Pediatrics. Tuberculosis. In: Pickering LK, Baker CJ, Long SS, McMillan J, eds. Red Book: 2006 Report of the Committee on Infectious Diseases. 27th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2006:678-698.

TUBERCULOSIS

Yearly tuberculin testing is no longer recommended for all children. Although the number of cases of tuberculosis in the United States has risen in recent years, these cases continue to occur primarily within previously identified high-risk groups. In populations with a low prevalence of tuberculosis, most reactive tests reflect false-positive results, often because of cross-reactivity with nontubercular mycobacteria, leading to unnecessary treatment with isoniazid. The American Academy of Pediatrics Committee on Infectious Diseases and the Pediatric Tuberculosis Collaborative Group currently endorse a selective screening strategy based on the presence of risk factors for exposure to tuberculosis (Table 12-5).^30,31 All children for whom these questions are answered positively should receive a tuberculin skin test, as should children who have been living in a homeless shelter, children with radiographic or clinical findings suggesting tuberculosis, and children who are about to initiate immunosuppressive therapy.³¹ In addition, children infected with human immunosuppressive virus and incarcerated adolescents should have an annual tuberculin skin test.³¹

Prior bacillus Calmette-Guérin vaccination is not a contraindication to tuberculin skin testing. Individuals who have received this vaccine can still acquire tuberculosis. Although some previously vaccinated individuals have a positive tuberculin skin test result, there is no reliable way to differentiate this reaction from that resulting from a natural infection with Mycobacterium tuberculosis, and recommendations regarding screening, test interpretation, and subsequent evaluation and treatment remain the same.³¹

Tuberculin testing relies on the presence of skin hypersensitivity to indicate subclinical or clinical infection. Reactivity generally develops within 2 to 12 weeks of infection.³¹ In all screening situations, Mantoux testing, which is more accurate and has the advantage of allowing quantification of the subsequent response, should replace multipuncture or Tine testing, and results should be read by qualified medical personnel.³⁰ In Mantoux testing, a standardized dose of tuberculin (0.1 mL of 5 tuberculin units of purified protein derivative tuberculin) is delivered intradermally, using a 26-gauge needle, to raise a wheal 6 to 10 mm in diameter.³⁰ The Mantoux test should be read at 48 to 72 hours by tactile measurement of the margins of induration. Erythema alone does not signify a positive reaction. Test interpretation is based on the size of induration, reason for testing, and presence or absence of other risk factors. Guidelines for interpreting test findings have been defined by the American Academy of Pediatrics, the American Thoracic Society, and the Centers for Disease Control (Table 12-5).³¹ The classification presumes the physician’s knowledge of the child’s and family’s risk factors as well as the background prevalence of tuberculosis in the community. One should remember that skin testing may be negative early in the course of the disease or in the presence of anergy. Criteria for prophylactic and therapeutic treatment of tuberculosis are presented in Chapter 269.