Extremity, Neurologic and Peripheral Vascular Examination

Most extremity and neurologic items on Part III are really localization problems: a patient points to a painful shoulder, a numb hand or a weak foot, and you must decide which tissue, which nerve or which level is responsible, and whether the cause is in the limb, the spine or the blood supply. The tests in this chapter are the tools for that decision. Know how each is done and what it means, but above all know which test to choose next and how much a positive or negative result should change your mind.

Section 1Upper extremity orthopedic examination in cases

The shoulder

Start every shoulder examination by clearing the neck, because C5 radiculopathy, C5 to C6 facet pain and referred visceral pain (the heart, the diaphragm through C3 to C5, the gallbladder on the right, the spleen on the left) all present as "shoulder pain." Pain that does not change with shoulder movement, or that is reproduced by neck movement, is not from the shoulder. Then compare active and passive range. Loss of both active and passive motion, worst in external rotation, then abduction, then internal rotation (the capsular pattern), indicates adhesive capsulitis or glenohumeral arthritis. Loss of active motion with preserved passive motion indicates a weak or torn rotator cuff or a nerve lesion. A painful arc between about 60 and 120 degrees of abduction suggests subacromial impingement or cuff tendinopathy; pain at the top of the arc, from about 160 to 180 degrees, suggests the acromioclavicular joint.

Clusters outperform single shoulder tests. For subacromial impingement, a positive Hawkins Kennedy, a painful arc and weak resisted external rotation together give a positive likelihood ratio of about 10. For a full thickness rotator cuff tear, a painful arc, a positive drop arm and weak external rotation together, especially over age 60, make a tear very likely. A traumatic cuff tear with marked weakness in an active patient is a reason for early orthopedic referral, because repair results are best within weeks.

The elbow

Lateral elbow pain in a person who grips repeatedly is usually lateral epicondylalgia (tendinopathy of the extensor carpi radialis brevis origin). It is reproduced by Cozen test (resisted wrist extension with a fist, elbow extended, forearm pronated), Mill test (passive wrist flexion with elbow extended and forearm pronated), and Maudsley test (resisted middle finger extension). Medial epicondylalgia is reproduced by resisted wrist flexion and pronation, or passive wrist extension with the elbow extended. Ligaments are tested with valgus stress at about 20 to 30 degrees of flexion for the ulnar collateral ligament (throwing athletes) and varus stress for the lateral collateral ligament. The ulnar nerve is assessed with a Tinel sign in the cubital tunnel and the elbow flexion test (full elbow flexion with wrist extension for one to three minutes reproduces ulnar side paresthesia). Lateral elbow pain that also produces aching in the dorsal forearm and weakness of finger extension without sensory loss suggests the posterior interosseous nerve (radial tunnel).

The wrist and hand

Section 2Lower extremity orthopedic examination in cases

The hip

Hip joint pain is felt mainly in the groin and anterior thigh and may refer to the knee, so a child or older adult with "knee pain" and a normal knee needs a hip examination. Lateral hip pain with tenderness over the greater trochanter that is worse lying on that side suggests greater trochanteric pain syndrome (gluteus medius and minimus tendinopathy, with or without bursitis). Buttock pain is more often lumbar, sacroiliac or gluteal than intra-articular. The hip tests most worth knowing were introduced in Chapter 3: loss of internal rotation as the earliest sign of osteoarthritis; FADIR for impingement and labral tears (sensitive, so a negative result helps exclude); the log roll, which is specific for intra-articular pathology because it moves only the joint; FABERE; the Thomas, Ely and Ober tests for flexor, rectus femoris and iliotibial band tightness; and the Trendelenburg sign. Craig test estimates femoral anteversion by rotating the prone hip until the greater trochanter is most lateral; increased anteversion produces in-toeing. The piriformis or FAIR test (flexion, adduction, internal rotation in side lying) reproduces buttock pain in piriformis syndrome.

The knee

The mechanism usually tells you which structure is injured: a noncontact pivot with a pop and rapid swelling (hemarthrosis within two hours) is the classic anterior cruciate tear; a blow to the lateral knee stresses the medial collateral ligament; a dashboard injury or fall onto the flexed knee drives the tibia backward and tears the posterior cruciate; twisting on a loaded, flexed knee tears a meniscus, which swells more slowly over 6 to 24 hours and may lock or catch. Check for effusion first with the bulge (sweep) test for small effusions and patellar ballottement for large ones.

The Ottawa knee rule calls for knee radiographs after acute injury if any one of the following is present: age 55 or older, isolated tenderness of the patella, tenderness of the fibular head, inability to flex the knee to 90 degrees, or inability to bear weight for four steps both immediately and in the examination room (limping steps count as bearing weight).

The ankle and foot

The common inversion sprain injures the anterior talofibular ligament first, then the calcaneofibular ligament. The anterior drawer of the ankle (pull the heel forward with the tibia stabilized, in slight plantar flexion) tests the ATFL, and the talar tilt (inversion stress in neutral) tests the CFL. A high ankle sprain of the distal tibiofibular syndesmosis follows external rotation of the foot, is painful above the joint and recovers slowly; it is reproduced by the squeeze test (compressing the tibia and fibula at midcalf produces distal pain) and the external rotation (Kleiger) test. The Thompson test squeezes the calf of the prone patient: absence of plantar flexion means an Achilles tendon rupture (sensitivity about 0.96, specificity about 0.93), often in a middle aged recreational athlete who felt as if kicked in the heel, sometimes after a fluoroquinolone. Morton neuroma (usually the third web space) gives burning forefoot pain reproduced by squeezing the metatarsal heads, sometimes with a click (Mulder sign). Plantar fasciitis produces first step heel pain with tenderness at the medial calcaneal tubercle, reproduced by passive great toe extension (windlass test). Tarsal tunnel syndrome gives a Tinel sign behind the medial malleolus with burning in the sole.

Section 3Neurologic screening in cases

Root level screening

A neurologic screen of the limbs tests each root three ways: a dermatome (the skin supplied by one root), a myotome (a muscle group supplied mainly by one root) and a deep tendon reflex. Agreement of all three at one level localizes a radiculopathy; disagreement suggests a peripheral nerve, a plexus or a central lesion. Because adjacent dermatomes overlap, sensory loss from a single root is often patchy; motor and reflex findings are more objective.

Muscle strength is graded 0 (no contraction), 1 (flicker), 2 (full range with gravity eliminated), 3 (full range against gravity), 4 (against some resistance) and 5 (normal). Reflexes are graded 0 (absent), 1+ (diminished), 2+ (normal), 3+ (brisk, possibly with spread) and 4+ (with clonus). Diminished reflexes on one side point to a root or peripheral nerve; symmetrically diminished reflexes may be normal or reflect a peripheral neuropathy (ankles first); brisk reflexes with spread, sustained clonus (more than three beats) or pathologic reflexes indicate an upper motor neuron lesion. Reinforcement (the Jendrassik maneuver, pulling clasped hands apart) helps bring out a hard to elicit reflex.

Upper versus lower motor neuron lesions

The upper motor neuron runs from the motor cortex through the internal capsule and brainstem, crosses at the pyramidal decussation of the medulla and descends in the lateral corticospinal tract. The lower motor neuron begins at the anterior horn cell and travels through the root, plexus and peripheral nerve to the neuromuscular junction. A lesion above the anterior horn releases the spinal reflex arc from cortical inhibition; a lesion at or below it interrupts the arc itself.

Pathologic reflexes are the bedside markers of an upper motor neuron lesion. The Babinski sign is extension of the great toe with fanning of the other toes when the lateral sole is stroked from heel to forefoot; it is normal in infants up to about 12 to 24 months. Oppenheim (stroking down the tibial crest) and Chaddock (stroking around the lateral malleolus) elicit the same response. In the upper limb, Hoffmann sign is the equivalent. Superficial reflexes are lost in upper motor neuron lesions and with damage to their own segments: the abdominal reflexes (T8 to T12), the cremasteric reflex (L1 to L2) and the anal wink (S2 to S4).

Cranial nerves and the cerebellum

Cerebellar function is screened with finger to nose and heel to shin (dysmetria and intention tremor), rapid alternating movements (dysdiadochokinesia), tandem gait and observation for nystagmus and scanning speech. Cerebellar signs are ipsilateral to a hemispheric lesion, while midline (vermis) lesions mainly produce truncal and gait ataxia. The Romberg test (feet together, eyes closed) assesses proprioception rather than the cerebellum: a patient who is steady with eyes open but falls with them closed has a dorsal column or large fiber sensory deficit (B12 deficiency, diabetic neuropathy, tabes dorsalis). A patient with cerebellar ataxia is unsteady with eyes open as well.

Section 4Peripheral vascular procedures

Pulses, color, temperature and refill

The peripheral vascular examination begins with inspection and touch: compare color, temperature, hair distribution, skin and nail texture, swelling and ulcers side to side. Chronic arterial insufficiency produces cool, pale, shiny, hairless skin, thickened nails and painful ulcers on the toes and pressure points; chronic venous insufficiency produces edema, brown hemosiderin staining, stasis dermatitis and shallow, less painful ulcers above the medial malleolus. Palpate the radial, brachial, femoral, popliteal, dorsalis pedis and posterior tibial pulses and grade them: 0 absent, 1+ diminished, 2+ normal, 3+ bounding (some scales add 4+). The dorsalis pedis is congenitally absent in a minority of people, so an isolated absent dorsalis pedis with a normal posterior tibial pulse is less meaningful than loss of both. Capillary refill after nail bed blanching should be under about 2 seconds (up to 3 seconds is often accepted in adults and the elderly); a prolonged refill suggests poor perfusion, though cold and dehydration slow it too.

The ankle brachial index

The ankle brachial index (ABI) is the ratio of the higher systolic pressure at the ankle (dorsalis pedis or posterior tibial, measured with a Doppler probe) to the higher of the two brachial systolic pressures. Each leg has its own ABI. It is the standard office test for peripheral arterial disease, with about 90 percent sensitivity and 95 percent specificity at a cutoff of 0.90. Calcified, noncompressible arteries in diabetes and chronic kidney disease falsely raise the ankle pressure, giving a value above 1.40 that hides disease; a toe brachial index is then used.

Vascular versus neurogenic claudication

Leg pain with walking in an older adult has two common causes, and the distinction is a favorite board item because both patients may also have back pain.

Deep vein thrombosis

The classic Homans sign (calf pain on passive dorsiflexion) is neither sensitive nor specific and should not be used to rule DVT in or out. Probability is estimated with the Wells criteria: one point each for active cancer; paralysis, paresis or recent plaster immobilization of the leg; bed rest over 3 days or major surgery within 12 weeks; localized tenderness along the deep veins; swelling of the entire leg; calf circumference more than 3 cm greater than the other side (measured 10 cm below the tibial tuberosity); pitting edema confined to the symptomatic leg; collateral superficial (nonvaricose) veins; and previously documented DVT; minus two points if an alternative diagnosis is at least as likely. A score of 2 or more means DVT is likely and the patient needs same day compression ultrasound; a score below 2 means DVT is unlikely and a negative D-dimer can exclude it. A suspected DVT is never massaged, manipulated or treated with soft tissue work.

Section 5Putting the NMS exam together

Choose tests by hypothesis

An efficient examination is not a checklist of every eponym in the region. By the end of the history you should hold two or three ranked hypotheses, including at least one that must not be missed. Each test is chosen because its result will change that ranking. The usual sequence is: screen for red flags and safety; observe; take active and passive range of motion and resisted testing to identify the tissue type (contractile, inert or neural); do the neurologic screen if there are limb symptoms; then use a small number of provocative or special tests targeted at the leading hypotheses; palpate last, so that tenderness does not color the earlier findings. Stop when the answer is clear enough to act on.

Sensitivity, specificity and likelihood ratios

Sensitivity is the proportion of people with the condition who test positive; specificity is the proportion without it who test negative. A highly sensitive test rarely misses disease, so a negative result rules it out (SnNout): the negative SLR for disc herniation, the negative upper limb tension test for cervical radiculopathy, the Ottawa rules for fracture. A highly specific test rarely labels healthy people, so a positive result rules disease in (SpPin): the crossed SLR, the drop arm test, the pivot shift, Spurling. Predictive values depend on prevalence and so change from one clinic to another, which is why likelihood ratios are more useful.

The positive likelihood ratio (LR+) equals sensitivity divided by (1 minus specificity); the negative likelihood ratio (LR minus) equals (1 minus sensitivity) divided by specificity. An LR+ above 10 or an LR minus below 0.1 produces a large, often conclusive change in probability; 5 to 10 and 0.1 to 0.2 produce moderate changes; 2 to 5 and 0.2 to 0.5 small changes; values near 1 change nothing. A convenient approximation for pretest probabilities between about 10 and 90 percent: an LR of 2 adds about 15 percentage points, 5 adds about 30 and 10 adds about 45; an LR of 0.5 subtracts about 15, 0.2 about 30 and 0.1 about 45.

From complaint to localized lesion

Localization works by asking where one lesion could explain every finding. Weakness and sensory change confined to one root's territory with a matching reflex change localize to the root. Findings that follow a peripheral nerve's map but cross root boundaries localize to the nerve. Findings in more than one root and nerve in one limb suggest the plexus. Bilateral findings with upper motor neuron signs place the lesion in the cord or brain, and a sensory level on the trunk places it in the cord. Stocking and glove sensory loss with reduced ankle reflexes describes a length dependent polyneuropathy (diabetes, alcohol, B12 deficiency, chemotherapy). Fatigable weakness with ptosis and diplopia and normal sensation points to the neuromuscular junction (myasthenia gravis); symmetric proximal weakness with normal sensation and reflexes points to muscle.

Words worth knowing