Vision Screening in Young Children: Best Practices

Joanne Waeltermann, MD; Laurie Scudder DNP, PNP


January 05, 2012

Editor's Note:
Providing the best screening that ensures optimal visual development in children is an enormous responsibility and one that pediatric primary care providers take very seriously. Over the last year, The Multi-Ethnic Pediatric Eye Disease Study and Baltimore Pediatric Eye Disease Study, the largest studies to date of children's eye disorders, have released 2 important papers examining risk factors for common pediatric eye disorders. The US Preventive Services Task Force (USPSTF) released new vision screening recommendations. Finally, a new study examined the risk for astigmatism in children with retinopathy of prematurity (ROP), even if these children do not develop the more severe complications associated with ROP.[1] Pediatric practices include an increasing number of children born preterm with a history of ROP. Medscape spoke with Joanne Waeltermann, MD, a pediatric ophthalmologist and clinical assistant professor at University of Maryland School of Medicine Department of Ophthalmology. Dr. Waeltermann is Chief of Pediatric Ophthalmology at St. Agnes Hospital in Baltimore, Maryland, and specializes in the eye problems of premature infants. We asked Dr. Waeltermann to discuss the implications of recent research for the pediatric primary care setting.

Medscape: Dr. Waeltermann, the recently revised USPSTF vision screening recommendations suggest that a visual exam be obtained at least once between the ages of 3 and 5 years and note that there is insufficient evidence to recommend screening in children younger than 3 years of age. This is a change from previous USPSTF guidelines, which endorsed visual screening for these younger children. Can you discuss some of the research that led to this change?

Dr. Waeltermann: The new guideline is the result of a review of a large number of studies for the treatment of amblyopia, including the PEDIG (Pediatric Eye Disease Investigator Group) studies. One of the conclusions of the reviewers was that amblyopia could be detected and corrected just as well in children 3 or 5 years of age as it could be in younger kids and that no strong evidence showed an advantage to treating prior to age 3 years. The new recommendation is that primary care providers should offer screening for the 3- to 5-year age group. Earlier screening (< 3 years) resulted in more false positive results for children screened in the primary care or school setting. There was concern about expenditures for exams and glasses in these children as well as potential negative effects on vision and social impact from unnecessary use of corrective lenses. The panel also could not endorse any particular combination of tests, type of tests (line ototypes, cross-cover, photoscreener, autorefractor, etc.), or screening age for use of these tests. None of the individual testing methods had outstanding sensitivity and specificity for amblyopia. There wasn't appropriate literature out there from which to make a definitive decision on these issues.

Medscape: The recommendations have generated some controversy. In an invited commentary to the guidelines published in Pediatrics, a leading group of experts voiced concern about the finding of "insufficient evidence" for screening children under the age of 3.[2] Other experts have noted that the guideline does not address the issue of red-reflex tests in young children and warned that omission of this test could cause clinicians to miss cataracts and retinoblastomas.[3] Are there other screening tools that should be utilized in primary care to better screen for these serious eye disorders?

Dr. Waeltermann: It was interesting that the USPSTF did not address red-reflex testing. The commentary to which you refer did suggest that the red reflex should have been included in these new recommendations. Its use is appropriate in the primary care setting because it's simple. It requires no equipment except a direct ophthalmoscope and it can be used on children of any age. However, the interpretation of the red reflex is affected by skill and experience. You have to know what you are doing to interpret a red reflex properly.

Medscape: I was taught that any child with an unequal red reflex should be referred to an ophthalmologist for evaluation.

Dr. Waeltermann: I agree. As long as the infant is awake and willing to fixate on your direct ophthalmoscope light, the red reflex is a very informative tool for the primary care provider. I, too, was surprised by the decision not to recommend this evaluation in primary care.

Medscape: The new recommendations do not discuss a more comprehensive assessment of a child, including personal and family history. Can you speak to this issue?

Dr. Waeltermann: When you are trying to determine whether or not a child actually has a problem -- such as strabismus (crossing or turning out of the eyes) or amblyopia (loss of vision in one eye vs the other) -- there are certain questions that you can ask families that give you a really high yield in identifying a true problem as opposed to something benign such as pseudostrabismus (large epicanthal folds). The 2 questions that I ask a parent are:

  • Have any of the family members -- aunts, uncles, cousins, or parents -- had glasses before the age of 5 years?

  • Have any of these family members had an eye that turns in or turns out or worn a patch?

Using that wording is important because amblyopia, strong glasses, and strabismus all run in families. I helped by contributing families from my practice to the study that identified the chromosomal location of these factors. If the child has a close family member with one of these conditions, the odds of you finding something in that child are much higher than if that piece of history is negative.

Medscape: So, a primary care provider who is uncertain, based on office-based screening, about the presence of a visual disorder should err on the side of caution and refer a child with that history?

Dr. Waeltermann: Yes, especially if you have a positive family history for glasses at a very young age, patching, or drifting of the eyes. Those kids should be seen because one of the most common ages for crossed-eye children to be first identified is around the age of 2 (prior the guidelines for screening).

Medscape: Is it still the case that these children may be identified by examining photos of the child?

Dr. Waeltermann: That is an interesting question because the widespread use of newer digital cameras has really changed what parents observe with respect to the red reflex. If the child is not looking directly at the camera, the pupillary image may appear like a white reflex. Occasionally, a parent will note the pupil has a white appearance in a photo, do some research on the Internet, and discover that retinoblastoma appears as a white reflex, so they'll tend to be frightened. A more common thing seen in photos is a child who might look cross-eyed as a result of wide epicanthal folds, which can be more pronounced in photos if the child is not looking straight into the camera or has his or her head partially turned.

Medscape: As we briefly discussed earlier, some experts have voiced concern that omission of red-reflex testing could lead to children with cataracts or retinoblastomas being missed in the primary care setting. Are there other screening tools that should be utilized in primary care? The USPSTF did not make a recommendation for a specific type of screening tool, but are there particular techniques that you would recommend primary care providers utilize to screen for these serious eye disorders?

Dr. Waeltermann: It depends on the age of the child. The younger the child, the more useful red-reflex or Bruckner testing is. As the child gets a little older, visual testing using a cover for one eye and an age-appropriate chart is most important. The American Academy of Ophthalmology clinical statement on Eye Examination in Infants, Children, and Young Adults by Pediatricians discusses a number of different tools. The easiest charts are called Allen picture or LEA symbol charts and utilize pictures of familiar objects such as a house or an apple and ask the child to identify the image. The HOTV test consists of a wall chart composed of Hs, Os, Ts, and Vs and requires that the child point to the letter that corresponds to the letter on the chart. This should be the next test used after the picture type of acuity test. The illiterate or tumbling E test asks the child to match their hand to which way the E is pointing. In my experience, that test is much more difficult for the child to do than the HOTV test.

There are several commercial tools available for primary care offices. One is an automatic refractor. SureSight™(Welch Allyn, Skaneateles Falls, New York) vision screener is an example of this type of product. These systems utilize a light that you shine at the child that very quickly refracts the child's eyes to check them for glasses. Guidelines for what is abnormal and normal are built into the machine. The children don't have to do anything. All they have to do is look at the light.

Another tool used for preschool vision screening is the photo screener, which also utilizes the red reflex. Photoscreener (Feakins Howson Partnership, Lancaster, Pennsylvania) is the most common of those. These machines take a flash photo of a child to produce a red reflex. Interpretation of results does require some training.

A last strategy is visual evoked potential testing for infants, which requires the application of 3 electrodes to the child's head and then flashing patterns or lights at the child and studying the brain's reaction. Enfant® Pediatric VEP vision testing system (Diopsys, Pine Brook, New Jersey) is being promoted to primary care practices for use in testing vision in preverbal children.

Medscape: The Bruckner test has also been suggested as a screening option for preschool-aged children. Could you discuss this exam and its use?

Dr. Waeltermann: Bruckner reflex testing is a very good option. You already have the instrument -- a direct ophthalmoscope -- in your office. Generally, children are quite cooperative with this testing, which can often be turned into a peek-a-boo game with the youngest of children. Bruckner testing allows the screener to assess visual acuity, alignment, refractive errors, and media opacities. The exam requires that the observer assess not just for the presence of a red reflex but also for color and symmetry. A strong ambylogenic refractive error will give a dull or gray reflex instead of the more typical rich reds and oranges. In children with darker colored eyes the "red" reflex may be greenish. Children who are not equally fixing on the light because of strabismus will appear to have an unequal red reflex; the reflex in the non-fixing eye will appear a lighter color. My Website provides a tutorial on use of this screening tool.

Medscape: It is well known that early recognition of refractive errors in children is critical in preventing childhood vision loss. The Multi-Ethnic Pediatric Eye Disease Study and Baltimore Pediatric Eye Disease Study screened almost 10,000 children and found startling differences in the prevalence of these disorders in children of varying ethnic groups. Could you discuss this finding? Do these findings suggest that some groups of children should be screened earlier?

Dr. Waeltermann: These large studies delved into the incidence and epidemiology of refractive errors, strabismus, and amblyopia in children. One of the findings was that astigmatism (where the cornea is shaped more like a football than a baseball) is associated with nearsightedness much more often than it is with farsightedness. Astigmatism was also found to be more common in Hispanic and African American populations. It is associated with maternal smoking. Younger children were found to have the most significant astigmatism, though this tends to resolve as the child becomes older.

Strabismus -- drifting, crossing, or turning out of the eye -- was found to be associated with prematurity as well as, again, maternal smoking. Additionally, children with strabismus were more likely than nonstrabismic children to have significant amounts of hyperopia (farsightedness), myopia (nearsightedness), and anisometropia (unequal need for glasses). The higher the degree of hyperopia, the higher the likelihood that the child would also develop crossed eyes.

Amblyopia with the potential for permanent decreased vision is generally associated with esotropia or crossing in of the eyes. When one eye crosses in, the brain suppresses the image from that eye so that the person does not see double. Amblyopia develops when that pattern of suppression becomes established in the brain. Anisometropia, the unequal need for glasses, is another important cause of amblyopia. Bilateral amblyopia can be due to very strong astigmatism or farsightedness in both eyes with the result that the brain never receives clear enough images to develop good vision. Severe astigmatism is very amblyogenic.

In unpublished work, another ethnic difference in visual disorders was the finding that Chinese populations have a much higher rate of high myopia or significant nearsightedness than other populations. Ethnically Chinese children are 6 times more likely to be nearsighted than other ethnicities (30% vs 5%).

Medscape: These studies identified maternal smoking as a specific risk factor associated with strabismus. Are there other modifiable risk factors for visual disorders?

Dr. Waeltermann: Lack of insurance was also noted. Even children who are covered by public insurance or limited private insurance plans often are allowed only 1 pair of glasses a year and they may not be of the highest quality. If the glasses break often the child goes without. Finally, prematurity is also an obvious risk factor.

Medscape: Your mention of prematurity is a good segue to my next question. A recent study concluded that children with ROP are at higher risk for astigmatism. Presumably those children with identified ROP are already receiving care from an ophthalmologist, but what about preterm children without recognized ROP? What are best primary care practices for screening in this growing group of children?

Dr. Waeltermann: Astigmatism is a major issue in very small, preterm infants. Babies who weigh less than 750 g at birth have a 90% chance of developing ROP. The risk for infants weighing less than 1000 g at birth is approximately 80%. Almost 65% of the children weighing less than 1250 g at birth will have some form of ROP. The vast majority of ROP will heal without any sequelae. Children with ROP or near-threshold ROP should be followed by an ophthalmologist until the age of 5-6 years, which is the age by which astigmatism will develop. However, preemies without ROP should be screened pretty much as other children. Assuming the child is growing appropriately and development is age-appropriate, their risk is not any different. It is the presence of ROP, not the prematurity itself, that increases risk for myopia, astigmatism, or both.

Medscape: Dr. Waeltermann, can you summarize the most important points that you would like to emphasize for our primary care members?

Dr. Waeltermann: Certainly. I think the most important points are as follows:

  1. The first is to realize that strabismus and amblyopia run in families. If a child has a first-degree relative with either crossed or drifted eye or who wore a patch as a child, that child is more likely to be affected.

  2. Refractive errors are more common in Hispanic, Chinese, and African American populations and in children with ROP. These refractive errors can contribute to strabismus and loss of vision and these children require particular vigilance.

  3. Screening should be attempted at any age with appropriate techniques, which means red-reflex testing in younger children and visual acuity screening in the older child (> 3 years).

  4. When a premature infant enters your practice, the presence and status of ROP should be determined in order to ensure appropriate follow-up.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.