• On forward flexion, the lumbar spine should move from its normal lordotic curvature to at least a straight or slightly flexed curve. If this change does not occur, some hypomobility probably is present in the lumbar spine either from tight structures or muscle spasm.

• In a patient with no back pain, when returning to the upright posture from forward flexion, the person first rotates the hips and pelvis to about 45° of flexion; during the last 45° of extension, the low back resumes its lordosis.

• In a patient with back pain, most movement usually occurs in the hips, accompanied by knee flexion; in some cases, the patient also uses hand support, working up the thighs.

• The examiner must differentiate the movement that occurs in the lumbar spine from that occurring in the hips or thoracic spine. Some patients can touch their toes by flexing the hips, even if no movement occurs in the spine. The degree of injury also has an effect. For example, the more severely a disc is injured (e.g., if sequestration has occurred rather than a protrusion), the greater is the limitation of movement.

• Often, an “instability jog” may be seen during one or more movements, especially flexion, returning to neutral from flexion, or side flexion. An instability jog is a sudden movement shift or rippling of the muscles during active movement, which indicates an unstable segment.

• Similarly, muscle twitching during movement or complaints of something “slipping out” during lumbar spine movement may indicate instability.

• If the patient bends one or both knees on forward flexion, the examiner should watch for nerve root symptoms or tight hamstrings, especially if spinal flexion is decreased when the knees are straight.

• Bourdillon and Day² recommend having the patient do this movement in the prone-lying position to hyperextend the spine. They called the resulting position the “sphinx position.” The patient hyperextends the spine by resting on the elbows with the hands holding the chin and allows the abdominal wall to relax. The position is held for 10 to 20 seconds to see whether symptoms occur or worsen if already present.

• Although the patient usually is in the standing position, rotation may be performed while the patient is sitting to eliminate or decrease pelvic and hip movement. If the patient stands, the examiner must take care to watch for this accessory movement and try to eliminate it by stabilizing the pelvis.

• If a movement (e.g., side flexion) toward the painful side increases the symptoms, the lesion is probably intra-articular, because the muscles and ligaments on that side are relaxed.

• If a disc protrusion is present and lateral to the nerve root, side flexion to the painful side increases the pain and radicular symptoms on that side.

• If a movement (e.g., side flexion) away from the painful side alters the symptoms, the lesion may be articular or muscular in origin, or it may be a disc protrusion medial to the nerve root.

• In the spine, the movement of side flexion is a coupled movement with rotation. Because of the position of the lumbar facet joints, side flexion and rotation occur together, although the amount and direction of movement may not be the same.

• Patients often deviate into forward flexion instead of remaining in true side flexion. To prevent this, the patient can be cued to run the hand down the back of the thigh instead of the side of the thigh.

• This is, in effect, a modified Trendelenburg’s test. A weak gluteus medius muscle or a coxa vara (abnormal shaft-neck angle of the femur) on the stance leg side may produce a positive sign on the nonstance side.

•  The examiner should look for patient compensations when completing the test. The most common compensation is trunk side flexion toward the stance leg. The patient side flexes to move his or her center of gravity over the stance leg, thus decreasing the demand on the gluteus medius. Therefore, the test may appear normal because the ilium responds normally, but if the side flexion is corrected, the positive response occurs.

• This test places significant stress upon the lumbar discs, so if disc pathology is suspected, the examiner should consider whether it is advisable to do the test.

• This test also can be done dynamically, to test endurance, by counting the number of repetitions to each side the patient is able to do before cheating.

• This test should be performed only if the patient receives a grade of “normal” in the dynamic abdominal endurance test or the abdominal isometric test.

• This is an abdominal eccentric test that can put a great deal of stress on the spine; therefore, the examiner must make sure the patient is able to hold a neutral pelvis before doing the exercise. If the patient cannot hold neutral spine or if the patient is symptomatic with the test, the test should not be completed. This test also causes greater abdominal activation than curl-ups.

• The test may also be done dynamically, to test endurance, by having the patient repeat the side bridging as many times as possible on each side.

• McGill et al.¹⁵ reported that the side bridge should be able to be held 65% of the extensor time for men and 39% for women, and 99% of the flexor time for men and 79% for women.