02 The examination as an instrument

To an onlooker, a neurologic examination can look like a ritual, with the same hammer taps, the same light shone in the eyes and the same request to touch your nose. Performed as a ritual it is slow and nearly useless. Performed well, each maneuver is a small experiment that tests whether a particular circuit is intact.

The idea fits in two words, think circuitry, and it comes with a procedure (DeMyer, p. 36). First you record what you actually observed, in plain terms. Only then do you translate it into what it means about the nervous system. Weakness and brisk reflexes on one side, for example, become a hemiplegia from interruption of the pyramidal tract, which places the lesion in a circuit (DeMyer, p. 36). A conclusion written without its observation can't be checked, either by the next examiner or by you when the picture changes.

There are three standards for deciding whether something is abnormal. You compare it with the norm for a person of the same age and sex, with its mate on the other side, and with the family (DeMyer, p. 36). The third is the easiest to forget and is especially useful in children. A large head in a child whose father also has a large head is often a harmless family trait, while a wide-based gait that matches a parent's points toward an inherited cause.

Screening and targeting

Nobody performs every test on every patient. An efficient neurologic examination chooses the parts that bear on the problem the history has uncovered (Morris 2012, p. xxv). A brief screening examination is done for everyone and then expanded or trimmed to fit the history, and the better the history, the briefer the examination needs to be (DeMyer, p. 604). The screen asks whether anything is wrong, and the targeted examination asks exactly where.

Recording the examination

"Neuro exam WNL" will not do as a note (DeMyer, p. 604). Those three letters record a verdict and leave out the evidence for it. A useful note says what you tested and what you found, especially the normal findings that bear on the question at hand. A line such as "walks, runs and hops on each foot; reflexes symmetric; toes down" tells the next reader far more.

The most underused instrument in the pediatric neurologic examination is the examiner's patience, because most of the examination can be done without touching the child. Before you reach for a hammer, you can see whether the child is alert and engaged, whether both sides of the body move equally, and how each hand is used. You can also see whether the face moves symmetrically during a smile or a cry, whether there are extra movements, and how the child sits, stands, walks and plays.

Watching first is good practice with every patient, since unobtrusive observation of how someone sits, stands, walks and gestures often reveals more than formal tests, particularly in infants (DeMyer, p. 239). With a child the reason is also practical. Formal testing needs cooperation, and a young child's cooperation soon runs out. Every finding you collect by watching is one you won't have to wring from a frightened toddler later.

What to watch for

Watch the child for five things as you take the history from the parents.

Undressing the child

It is tempting to skip undressing out of politeness, but if a third of the body stays covered, you can only do two thirds of the examination (DeMyer, p. 239). For a child that means undressing down to the underwear, with a parent present, so that you can see the back, the spine, the calves and the skin. Café-au-lait spots, a tuft of hair over the lower spine, wasted or enlarged calves, and a difference in size between the limbs of the two sides are all easy to miss through clothing.

Ask a school-age child to push against your hand and you may get a polite nudge. Tell her "don't let me win," and you are far more likely to see what she can really do. Turning tests into a friendly contest works even with adults, and doubly well with children. Ask whether the patient can see out to the side as far as you can, or how light a touch they can feel. Challenges like these draw out the patient's best performance and keep the examiner interested too (DeMyer, p. 38). It also helps to expect the abnormal, because a mind that expects normal findings stops looking carefully (DeMyer, p. 38). The form the game takes changes as the child grows.

Newborn. Examine between feeds. Interrupting a feed produces a crying baby, and a full stomach produces a sleepy one. Watch spontaneous movements, posture and alertness before you handle the baby at all.

Infant. Faces, voices and bright objects test vision, hearing and eye movements, while pull to sit, ventral suspension and supported standing test tone and strength. The reflexes that should come and go are covered in section 5.

Toddler. At this age everything is play. A toy to follow tests eye movements, blowing a kiss and showing teeth test the face, and stacking blocks or picking up small objects tests fine motor skill and hand preference. To screen gait, ask the child to run a short distance to fetch something (Brazis 2011, p. 17). Videos of these tasks at each age are on the Utah PediNeuroLogic Exam site.

School-age child. Most formal tests become possible once they are framed as challenges, such as hopping on one foot, walking heel to toe like a tightrope walker, or "push me away". Children will refuse pinprick, but those as young as 3 will play "the temperature game" with a cool object instead (DeMyer, p. 617).

Adolescent. The examination is close to an adult's, with more attention to privacy and to asking the teenager alone about symptoms they may not raise in front of a parent.

Try the figure above, which asks you to put eight steps of a toddler examination in order. Any order within a tier counts as correct. It checks that you watch first, then play, then move to hands-on testing, and leave the things a toddler hates until last.

A toddler who has cheerfully stacked blocks for ten minutes can be lost for the rest of the visit the moment a bright light comes close to his face. So do the hands-off parts first, the games next, the hands-on parts after that, and the things a child hates last. A tape measure round the head, a bright ophthalmoscope light and a tongue depressor can each end cooperation, so they wait until everything else is done. Reflexes are best done late, with the child on a parent's lap. Turning the reflex hammer into an imaginary horse with sound effects stops it from looking like a weapon (Utah PediNeuroLogic Exam).

Swinging the hammer

The commonest student error is hammer pecking, which means gripping the handle tightly and jabbing at the tendon with no wrist action. Hold the handle loosely between thumb and finger and let the hammer swing like a pendulum (DeMyer, p. 252). A reflex that seems absent after a peck is often present after a swing.

Before you call a reflex absent

Reposition the limb to change the tension on the tendon, or adjust the pressure of your thumb. Use reinforcement, for instance by asking the patient to lock their fingers together and pull hard while you tap (DeMyer, p. 258). In infants the biceps tendon can be hard to find, and the knee jerk is easiest to elicit with the leg nearly straight (DeMyer, p. 258). A child who is tense, cold or frightened will often show reflexes once relaxed.

Testing tone

Tone is the slight resistance you feel when you move a relaxed joint (DeMyer, p. 270). A child who "helps" you move the limb, or who stiffens in fear, will feel stiff for reasons that have nothing to do with the nervous system. Distract the child, wait, and then move the limb both slowly and quickly.

The plantar response in a ticklish child

Stroking the sole of a ticklish child produces a withdrawal that can look like an upgoing toe. Ticklish young patients tolerate stimulation on the lateral side of the foot better, so start there (DeMyer, p. 616).

Throughout, keep comparing one side with the other and the child with the family (DeMyer, p. 36). An asymmetry usually tells you more than any single absolute value.

An infant's examination includes findings that are supposed to disappear and others that are supposed to appear, each on its own schedule. That schedule is a powerful localizing tool in early childhood.

Primitive reflexes such as the Moro, the palmar grasp and the Galant are normally present in the term newborn and diminish over the next 4 to 6 months. Postural reactions such as the Landau, lateral propping and the parachute emerge between about 3 and 8 months. Persistent primitive reflexes, together with postural reactions that fail to develop, are the hallmark of an upper motor neuron abnormality in an infant (Utah PediNeuroLogic Exam). Examining the two together is a simple, quick and predictive screen for infants at risk of cerebral palsy (Zafeiriou 2004).

Why they come and go

The primitive reflexes are generated low in the nervous system and are present at birth. As the cortex and its descending pathways mature, they suppress these reflexes, and when descending control is damaged the reflex persists. The postural reactions depend on the same maturing control, so the same damage leaves them late or absent, and one lesion produces two signs pointing the same way.

Use the slider in the figure above to move from birth to 12 months. The ages come from the Utah PediNeuroLogic Exam.

Asymmetric reflexes

A symmetric delay suggests a global problem, and an asymmetric reflex is a localizing sign. A parachute that extends only one arm, or a grasp that persists in one hand, usually points to the motor pathway of the opposite hemisphere, though a brachial plexus injury on the same side can look similar. An absent plantar grasp in the first 6 months correlates with poor neurologic development (DeMyer, p. 282). Short videos of each reflex are on the Utah site at 3 months, 6 months and 12 months.

Stroke a newborn's palm and her fingers close round yours, which is exactly what they should do. Draw the same grip from an adult whose frontal lobes are failing and it means something else entirely. Across a lifetime the palmar grasp goes from present and normal, to absent and normal, to present again and abnormal. In the newborn the grasp shows that the lower parts of the motor system work. Its fading shows descending control arriving, and its return shows that control failing.

A lesion's place in the nervous system needs three spatial coordinates, and in a child it needs a fourth, time since birth. In an adult an upgoing toe is an upper motor neuron sign whenever you find it. In a child the same observation can be a normal stage, a warning, or a leftover from infancy. The basic unit of the pediatric exam is the pair (finding, age). The first standard of abnormality is comparison with the norm for the patient's age (DeMyer, p. 36), and in a child that comparison is half the work.

Findings whose meaning depends on age

Reactions that fail to appear

A postural reaction that should have arrived and hasn't is as informative as a reflex that should have gone and hasn't (Utah PediNeuroLogic Exam). A Moro still present at 6 months and an upgoing toe at 6 years say the same thing, that the descending motor pathway is not doing its job. Age only moves the line where normal ends. When Module 3 asks whether a weakness is upper or lower motor neuron, part of an infant's answer comes from reflexes that failed to disappear, so record the age beside every finding.