Results for unilateral astigmatism showed also the higher detection rates at 4 m distance (Table 9.3). On the basis of these results, it is recommendable to perform the test also at a distance of 4 m to detect refractive error more sensitively [34].

Paysse et al. compared the ability of paediatric residents to di erentiate asymmetric from symmetric red reflex in ten patients and six control subjects. Four patients were anisometropic by 2.25–5.5 dioptres without strabismus. In the entire group, paediatric residents achieved a test sensitivity of 61% and a specificity of 71% [3]. Gole and Douglas reported a test sensitivity of 86% and a specificity of not more than 65%. The Brückner test was performed by a medical student [20]. In these two studies, the test distance was 1 m. In a group of anisometropic patients, we achieved a sensitivity of 32.5% at that distance, and a specificity of 93.3%. At a distance of 4 m, sensitivity increased to 77.5% and specificity was 80%

[34]. These rates that depend on patient selection and observer experience are not representative for a real screening situation in early infancy.

9.3.2Performance

It is commonly recommended to perform the test at a distance of about 1 m or less (‘arm’s length distance’) by simultaneously illuminating both eyes of a patient, and to compare colour and brightness of the pupillary red reflexes for symmetry [3, 18, 19, 35, 37, 38, 40, 41]. The room light should be dimmed but the room should not be completely dark [18].

Using a direct ophthalmoscope is mandatory.Otoscope or flashlight illumination will not yield the same optical phenomena because the characteristic of the emitted light is di erent. The light beam must be directed simultaneous

into both eyes to enable accurate comparison of the red reflexes. Light intensity can be varied during the examination. It should not be too high to avoid glare. Detection of small media opacity is easier at 0.5–0.1 m, examining each

9eye separately and using a convex lens in the ophthalmoscope, if necessary. To improve detection of refractive error, the examiner should then go 4 m backwards continuously observing the pupils simultaneously for luminance of the red reflexes. By the same way, it is possible to check for correct spectacle correction.

9.3.3Possibilities and Limitations

Severe media opacity is visible by lacking or dark fundus red reflex, regardless of distance. Small media opacity and pathologies of the fundus are best visible at a short distance. Any asymmetry in brightness and colour of the reflexes is predictive of amblyogenic risk factors [18, 19, 37, 38].

While the test is very sensitive to detect media opacity, detection of small-angle strabismus is limited. Detection of ametropia (particularly myopia) and anisometropia can be improved by extending the test distance, but nevertheless, isometropic hypermetropia cannot be detected reliably. Inter-ocular asymmetry in the red reflex can be caused by anisocoria and is also frequent in the age range below 8 months.

a

b

c

d

Summary for the Clinician

■The fundus red reflex test is an excellent complement and a possibility for ophthalmologists, orthoptists, paediatricians, and general practitioners to recognize various eye disorders very early. It is also a valuable tool for use in developmental countries. At a short distance relevant media opacity can be reliably detected by darkening of the red reflex. Testing for refractive error is better performed at an extended distance. Unior bilateral partially or completely weak or lacking red reflex is always pathological.

e

f

Fig. 9.5 Brückner reflex at 4 m distance in case of emmetropia OU (a) and simulated hypermetropia OS (anisometropia) of 1 diopter (b), 2 diopters (c), 3 diopters (d), and 4 diopters (e). For comparison, simulated myopia OS of 1 diopter (f) [34]

9.4Pupillary Light Reflexes

Pupillary light reflexes are being tested to detect pathologies in the iris, in the e erent branch of the pupillary light reflex loop, in the midbrain, or in the a erent branch of the reflex loop.Regarding strabismus diagnostic,Brückner described two criteria:

1.Dimming of the red reflex when the child is centrally fixating the ophthalmoscope light.

2.Reduced direct pupillary light reflex in the amblyopic eye compared with the direct light reflex in the nonamblyopic eye.

This step requires monocular illumination of the pupils. Brückner reported pupillary constriction in the deviated eye when the light beam was changed from the fixating eye onto the strabismic eye, as soon as this eye took up fixation. Permanent fixation with the previously illuminated dominant eye yields an eccentric retinal image of the ophthalmoscope light in the deviated eye. Despite dark adaptation of the deviated eye, the pupillomotor e ect of the eccentric illumination can be weaker than that of the central illumination in the fellow eye. So, the response to alternating illumination may either look like relative a erent pupillomotor deficit or dimming of the red reflex in the amblyopic eye occurs after some latency when the amblyopic eye takes up fixation.

9.4.1Physiology

Illumination of one eye causes symmetric constriction of both pupils [42–46]. In unilateral amaurosis, pupillary constriction is lacking in both eyes when only the blind eye is being illuminated. Illumination of the other eye causes normal constriction of the pupils in both eyes. Less severe a erent disorders show a similar pattern except residual reaction to illumination of the (more severely) concerned eye. Discreet a erent disorders can be found by the swinging flash light test [42–44]. Aiming at strabismus diagnostic, the observer has to watch any eye movement occurring after the change of the illumination to the other eye. If the previously deviated eye which is now being illuminated takes up fixation, the movement of this eye may be visible, and the pupils will constrict because foveal illumination

has a stronger pupillomotor e ect compared with paracentral illumination. Pupillary constriction is also induced by the increased light sensitivity of the ‘dark-adapted’ eye. In the clinical situation, it is hardly possible to discriminate between these two mechanisms. If the strabismic eye fails to take up central fixation, an a erent pupillomotor defect component may be simulated when this eye is being illuminated or there is in fact a relative a erent pupillary defect (RAPD) due to amblyopia [47–50]. Figure 9.6 shows that already minimal eccentricity of illumination reduces pupillary constriction compared with a central illumination.

9.4.2Performance

The examiner directs the light cone on the patient’s right eye and observes constriction of each pupil. The procedure is repeated illuminating the patient’s left eye. If both pupils are normally reactive, which is mostly the case, comparison of the direct light reflexes of both eyes will be su cient [51–52]. If only one pupil is reactive, this pupil can be used to compare the constriction with subsequent illumination of the right and the left eye. The pupillary constriction has to be equal in latency, speed and amplitude, regardless of the eye illuminated.

9.4.3Possibilities and Limitations

RAPD is typical of severe asymmetric retinal lesion or asymmetric lesion of the optic nerve including the optic chiasm. Amblyogenic disorders, such as refractive error,

Fig. 9.6 Video-oculographic registration of the change in pupil diameter with alternating fixation of the ophthalmoscope light and low illuminated visual targets 2.5, 5, 7.5, and 10° right (positive values) and left (negative values) of the ophthalmoscope light. Fixation of the ophthalmoscope light induced more pupillary constriction than fixation of a target as few as 2.5° beside

pupil diameter

6 mm

4 mm

2 mm