INTRODUCTION

Lumbar segmental instability is an important but often unrecognized cause of chronic low back pain (LBP). It has been a controversial and poorly understood topic, primarily because of the varying definitions and usage among the several disciplines involved in the treatment of spinal disorders.¹

Traditionally, one of the most obvious manifestations of lumbar instability has been the radiographic diagnosis of spondylolisthesis,^2,3 with increased segmental motion reported in patients with chronic LBP.^4–7 However, lumbar segmental instability (‘microinstability’) also has been cited as a cause of chronic LBP in patients without an injury to the osseous spine,⁸ with a number of studies reporting increased and abnormal intersegmental motion in these patients.^9–11

DEFINITION OF SPINAL INSTABILITY

Numerous biomechanical and clinical definitions have been developed to describe spinal instability; however, most, including the one adopted by the American Academy of Orthopaedic Surgeons, imply excessive motion beyond normal constraints.^12–16 Despite a multitude of biomechanical, radiographic, and clinical studies, ‘normal’ motion of the lumbar vertebral segments continues to be a topic of debate because of the lack of standardized measurements in normal subjects.¹⁷

Correlating clinical instability and radiographic instability also has been difficult because of the overlap of symptomatic and asymptomatic motion patterns. Additionally, conventional radiography often is insensitive and unreliable in detecting abnormal or excessive intersegmental motion.^18,19 Conceptually, therefore, as with other spinal conditions, radiographic abnormalities (more specifically, abnormal motion of a single motion segment) are considered significant only if they confirm the clinical finding (of lumbar segmental instability) at the corresponding symptomatic level.¹⁴

As the pathomechanics of the lumbar spine have become better understood, definitions of spinal instability have been developed that incorporate both the mechanical aspect and the clinical consequence. In 1992, Panjabi defined spinal instability as a region of laxity around the neutral position of a spinal segment, called the neutral zone, and more precisely as a decrease in the capacity of the stabilizing systems of the spine to maintain intervertebral neutral zones within physiological limits within which there is no major deformity, no neurological deficit, and no incapacitating pain.²⁰ This definition is useful because it describes the quality of motion throughout the range of motion rather than relying solely on the total range of motion values for diagnosis.²¹

THE NEUTRAL ZONE CONCEPT AND THE SPINAL STABILIZING SYSTEM

The neutral zone concept of segmental instability is based on the observation that the load-displacement curve of the spine is nonlinear (i.e. the ratio of the load applied to the displacement produced is not consistent), with minimal resistance to intervertebral motion occurring within the neutral zone and increased resistance to intervertebral motion occurring at the end-ranges of spinal motion.²² This concept has been described as a ‘ball in a bowl’ (Fig. 101.1). The ball rests in a bowl created by flipping the extension part of the load-displacement curve around the displacement axis. The ball moves easily within the base of the bowl (the neutral zone), but requires greater effort to move in the steeper part of the bowl (the end-ranges of motion). The shape of the bowl is analogous to spinal stability: a deeper bowl, such as a wine glass, represents a more stable spine, while a shallower bowl, such as a soup plate, represents a less stable spine (Fig. 101.2).²² The total range of motion of a spinal motion segment, therefore, can be divided into a neutral zone and an elastic zone. The neutral zone is the initial spinal motion which is produced against minimal internal resistance, whereas the elastic zone is motion nearer to the end-range of movement that is produced against significant internal resistance.²³ In order for the stabilizing system of the spine to be effective, it must limit the excursion of spinal motion within a segment and maintain the proper ratio of neutral to elastic motion.²³

Fig. 101.1 Load-displacement curve. (A) Spinal segment subjected to flexion and extension loads exhibits a nonlinear load displacement curve, indicating a changing relationship between the applied load the displacements produced. (B) A ‘ball in a bowl’ is a graphic analogy of the load-displacement curve.

(Adapted from Panjabi MM. Clinical spinal instability and low back pain. J Electromyogr Kinesiol 2003; 13:371–379.)

Fig. 101.2 Representation of different spinal stabilities. Using the analogy of a ball in a bowl to represent the load-displacement curve of the spine (Fig. 101.1), a deep champagne glass and a shallow soup plate represent a more and a less stable spine, respectively.

(Adapted from Panjabi MM. Clinical spinal stability and low back pain. J Electromyogr Kinesiol 2003; 13:371–379.)

Biomechanical studies of the stabilizing system of the spine have provided insight into the roles of the various components providing spinal stability. The stabilizing system of the spine is composed of three subsystems: (1) the spinal column or passive subsystem, (2) the spinal muscles or active subsystem, and (3) the neural control unit.^20,22 Because, under normal conditions, the three subsystems work in harmony to provide mechanical stability, their functions are interrelated, and decreased function of one subsystem may result in increased demands on the others to maintain stability.^24,25 The various components of the spinal column generate transducer information about the mechanical status of the spine, such as position, load, and motion of each vertebra, and the neural control unit computes the needed stability and generates the appropriate muscle pattern (Fig. 101.3).²³ A comprehensive knowledge of the three subsystems of the lumbar spine is essential in understanding spinal stabilization and clinically evaluating patients with LBP.

Fig. 101.3 Spinal stabilization system has three subsystems: spinal column, muscles surrounding the spine, and motor control unit.

(Adapted from Panjabi MM. Clinical spinal instability and low back pain. J Electromyogr Kinesiol 2003; 13:371–379.)

A HYPOTHESIS THAT RELATES MOTION TO PAIN

The relationship between abnormal intervertebral motion and low back pain is inherent in the definition of clinical spinal instability. In theory, therefore, if intervertebral motion is decreased in a patient with LBP, pain also should be decreased. This assumption is the basis for low back treatments involving surgical fusion, muscle strengthening, and muscle control training.²³ Based on the hypothesis that severe LBP is caused by instability between lumbar segments, Olerud et al.⁵³ used external fixation to produce ‘instantaneous fusion’ of selected lumbar spinal segments in 18 patients with chronic severe LBP; 17 patients experienced ‘remarkable relief’ of pain. The authors suggested that this test could be used to identify unstable segments before surgery. Panjabi et al.,⁵⁴ in a biomechanical cadaver study of the cervical spine, found that after stabilization with external fixation the average range of motion decreased by 39.3%, while the neutral zone decreased by an average of 68.8%.

Using the ‘ball in a bowl’ analogy, Panjabi²³ formulated a hypothesis to relate motion to pain (Fig. 101.4). He postulated that in people without spinal pain the neutral zone and range of motion are normal and the ball moves freely within the pain-free zone (Fig. 101.4A). Injury to or degenerative changes in a spinal column component result in an increase in the neutral zone and the ball moves freely over a larger distance, beyond the pain-free zone (Fig. 101.4B). The spinal stabilizing system reacts to decrease the neutral zone by activating the muscles or by adaptive stiffening of the spinal column over time (e.g. formation of osteophytes) (Fig. 101.4C). The spine also may be stabilized by surgical fusion, muscle strengthening, and retraining of the neuromuscular control system. In the analogy, the ball is now anchored and the spine is again pain free. This hypothesis is unproven and must be validated by future clinical studies.²³

Fig. 101.4 Hypothesis to relate motion to pain. A ball-in-a-bowl analog representing the motion-pain hypothesis. (A) Control spine with neutral zone (NZ) within pain-free zone. (B) Painful spine has greater NZ bringing the pain-free zone within it. (C) Stabilized spine has decreased NZ and, therefore, is pain-free.

(Adapted from Panjabi MM. Clinical spinal instability and low back pain. J Electromyogr Kinesiol 2003; 13:371–379.)

CLINICAL ASSESSMENT OF LUMBAR INSTABILITY

From: Hazlett JW, Kinnard P. Lumbar apophyseal process excisions and spinal instability. Spine 1982; 7:171–176.

Table 101.3 Checklist for the diagnosis of clinical instability in the lumbar spine

From: White AA, Panjabi MM. Clinical biomechanics of the spine. Philadelphia: JB Lippiincott; 1990.

Traditionally, the diagnosis of instability has been based on radiographic observations and measurements. Although disc space narrowing, air within the disc, and calcification of the disc have been observed with spinal instability,^57–60 as have hypertrophic changes of the facet joints and claw osteophytes, these are signs of degeneration, not specifically of instability, and are not consistently correlated with symptoms. Macnab⁶¹ described the traction spur, which is a horizontal osteophyte arising at the attachment of the outer fibers of the anulus on the anterior and lateral aspects of the vertebral body, 2 mm away from its disc surface. This spur is thought to result from excessive tensile stresses applied to the outer vertebral body through the annular fibers as a consequence of segmental instability. No direct, statistically valid, clinical significance of the traction spur, however, has been established.

Knutsson⁶² described a method for diagnosing segmental instability based on lateral radiographs of the lumbar spine taken with the patient in maximal active extension while standing and in maximal active flexion while sitting. He defined instability as 3 mm or more of anterior translation measured between flexion and extension radiographs. Flexion–extension radiographs have now become the standard by which segmental instability is diagnosed. A variety of methods have been developed to more precisely quantify these displacements^63–66 either on neutral views or by comparison of flexion and extension views. The amount of anterolithesis considered indicative of instability varies from 3 mm to 6 mm or 6–15% of the width of the adjacent vertebra (Fig. 101.5).^19,55 Suggested criteria for instability include a relative sagittal angulation of more than 22 degrees on resting radiographs (Fig. 101.6) or more than 15 degrees at L1–2, L2–3, and L3–4 levels, more than 20 degrees at the L4–5 level, and more than 25 degrees at the L5–S1 level on flexion–extension films (Fig. 101.7).⁶⁷ Limitations in these techniques are that the ranges of normality have not been well established and the accuracy in measurement is low.^68,69 Additionally, variations in the exact technique used, large variability in the motion characteristics of individuals without LBP, and high false-positive rates using the established criteria have resulted in questions about the usefulness of flexion–extension radiographs.^53,69 Because of the intra- and interobserver errors in assessing instability radiographically, Spratt et al.⁷⁰ recommended that a minimum of 4 mm of forward displacement at L3–4 and L4–5 and displacement of more than 5 mm at L5–S1 are necessary to accurately measure instability.

Fig. 101.5 Measurement of sagittal plane translation or displacement. If translation or displacement is 4.5 mm or 15% of the sagittal diameter of the adjacent vertebra, it is considered abnormal.

(Adapted from White AA III, Panjabi MM. Clinical biomechanics of the spine, 2nd edn. Philadelphia: JB Lippincott; 1990.)

Fig. 101.6 Measurement of relative sagittal plane angulation of L3–4 functional spine unit (FSU) on a sagittal plane angulation or more than 22 degrees is abnormal and potentially unstable in the lumbar spine.

(Adapted from White AA III, Panjabi MM. Clinical biomechanics of the spine, 2nd edn. Philadelphia: JB Lippincott; 1990.)

Fig. 101.7 Measurement of sagittal plane rotation of L4–5 functional spinal unit on dynamic (flexion–extension) lateral radiographs.

(Adapted from White AA III, Panjabi MM. Clinical biomechanics of the spine, 2nd edn. Philadelphia: JB Lippincott; 1990.)

Many other techniques using lateral bending films, bipolar radiography, multidirectional reconstruction, computer digitization, and stereophotogrammetry have been investigated in measuring spinal motion and determining instability.^59,71 These methods, however, have not achieved practical clinical use and have not established specific criteria for spinal instability. Perhaps most indicative of chronic instability is the radiographic progression of malalignment or deformity over time, the best two examples of which are degenerative spondylolisthesis and degenerative scoliosis. Newman and Stone⁷² determined that in degenerative spondylolisthesis progression averaged 2 mm every 4 years. Similar observations have been made about adult degenerative scoliosis⁷³ and, to a lesser degree, retrolithitic deformities. Of all the radiographic observations and measurements mentioned, however, none has statistically proved to be specifically associated with significant clinical symptoms.^{57–59,69,71,74} In order to establish a diagnosis of spinal instability, therefore, the clinician must correlate the biomechanical environment and radiographic observations with the patient’s clinical history and physical examination.

To date, no diagnostic gold standard for segmental instability has been identified. Historically, patients complain of chronic and recurrent pain localized to the low back or radiating into the lower extremities and associated with high levels of functional disability.⁵⁸ Because the condition is mechanically based, the pain typically worsens with greater loads on the spine such as general activity, lifting, standing, sitting, and motions such as bending or twisting. Kirkaldy-Willis and Farfan suggested that segmental instability is a condition in which ‘minor perturbations produce acute pain.’¹⁴ Conversely, the pain is relieved by lying supine, the position that places the least mechanical load on the lumbar spine.⁷⁵ Patients most commonly describe their back pain as recurrent, constant, catching, locking, giving way, or associated with a feeling of instability.⁷⁶ Several researchers have reported similar physical examination findings that are consistent with a movement–control problem within the neutral zone:^14,56,77 (1) active spinal movement with good ranges of spinal mobility but with ‘through-range’ pain or a painful arc rather than end-of-range limitation of motion, and (2) the inability to return to erect standing from forward bending without the use of the hands to assist in this motion. Segmental shifts or hinging were frequently associated with the painful movement. Deep abdominal muscle activation during the provocative movement often reduced or eliminated pain. Neurological examination and neural tissue provocation tests were generally normal.⁷⁸

Clinical syndromes of microinstability

The directional nature of instability based upon the mechanism of injury, resultant site of tissue damage, and clinical presentation is well understood in the knee and shoulder but poorly understood in the lumbar spine. Based on experimental and radiological data, Dupuis et al.⁶⁴ concluded that the location of the dominant lesion in the motion segment determines the pattern of instability manifested. Because the motion within the lumbar spine is three-dimensional and involves coupled movements, tissue damage is likely to cause movement dysfunction in more than one direction.⁷⁶

The descriptions of microinstability syndromes have been developed from clinical observation and are based on the mechanism of spinal injury, the resultant tissue damage, the reported and observed aggravating activities, and movement problems relating to a specific movement quadrant or quadrants. These descriptions provide a basis by which patients can be evaluated and movement dysfunction analyzed in a segment-specific and individual manner.

Common to all the microinstability syndromes are the patient’s subjective feeling of instability and an objectively observed lack of movement control and related symptoms within the neutral zone, all of which are associated with an inability to initiate co-contraction of the local muscle system within this zone. The local muscle system consists of muscles that attach directly to the lumbar vertebrae and are responsible for providing segmental stability and directly controlling the lumbar segments. The lumbar multifidus, psoas major, quadratus lumborum, the lumbar parts of the lumbar iliocostalis and longissimus, transverses abdominis, the diaphragm, and posterior fibers of the oblique abdominis internus form part of this local muscle system. Patients may develop compensatory movement strategies that ‘stabilize’ the motion segment out of the neutral zone and towards an end-range position (such as flexion, lateral shift, or extension). This is achieved by the recruitment of global system muscles and by generating high levels of intra-abdominal pressure during low-load tasks. The global muscle system consists of large torque-producing muscles that act on the trunk and spine without directly attaching to it. These muscles include the rectus abdominus, obliquus abdominus externus, and the thoracic part of the lumbar iliocostals; they provide general trunk stabilization but are not capable of exerting a direct segmental influence on the spine.^76,78

Flexion pattern of microinstability

Patients with flexion pattern microinstability, the most common type, complain of central back pain and relate their injury to either a single flexion–rotation injury or to repetitive strains relating to flexion–rotation activities. Symptoms and ‘vulnerability’ are aggravated by flexion–rotation movements and patients are unable to sustain semiflexed postures (Fig. 101.8). A loss of segmental lumbar lordosis at the level of the ‘unstable motion segment’ often is noticeable when the patient is standing and is accentuated in sitting postures because patients tend to hold the pelvis in a degree of posterior pelvic tilt. Lower lumbar segmental lordosis is further decreased in flexed postures and usually is associated with increased tone in the upper lumbar and lower thoracic erector spinae muscles, with an associated increase in lordosis in this region. During forward bending, patients have a tendency to flex more at the symptomatic level than at the adjacent levels, usually producing an arc of pain and an inability to return from flexion to neutral without use of the hands to assist the movement. During backward bending, extension often is less in the affected segment than in the more proximal segments. Specific movement testing reveals an inability to differentiate anterior pelvic tilt and low lumbar spine extension independent of upper lumbar and thoracic spine extension. Movement tests such as squatting, sitting with knee extension or hip flexion, ‘sit to stand,’ and forward-loaded postures reveal an inability to control a neutral segmental lordosis, with a tendency to segmentally flex at the unstable motion segment, posteriorly tilt the pelvis, and extend the upper lumbar and thoracic spine.

Fig. 101.8 Unstable movement zone: flexion pattern.

(Adapted from O’Sullivan PB. Lumbar segmental instability: clinical presentation and specific stabilizing exercise management. Manual Therapy 2000; 5:2–12.)

Specific muscle tests reveal an inability to activate the lumbar multifidus in co-contraction with the deep abdominal muscles at the ‘unstable’ muscle segment within a neutral lordosis. Many patients are unable to assume a neutral lordotic lumbar spine posture, particularly in four-point kneeling and sitting. Attempts to activate these muscles commonly are associated with bracing of the abdominal muscles with a loss of breathing control and excessive coactivation of the thoracolumbar erector spinae muscles and external oblique. This is associated with a further flattening of the segmental lordosis at the unstable motion segment, often resulting in pain. Palpation reveals a segmental increase in flexion and rotation mobility at the symptomatic motion segment.

Extension pattern of microinstability

Patients with extension pattern microinstability relate their injury to an extension–rotation incident or repetitive trauma usually associated with sporting activities involving extension–rotation. Their symptoms are aggravated by extension and extension–rotation movements and activities such as standing, fast walking, running, swimming, and overhead activities such as throwing (Fig. 101.9). When the patient is standing, segmental lordosis at the unstable motion segment usually is increased, sometimes with an increased level of segmental muscle activity at this level and anterior pelvic tilt. Extension activities reveal segmental hinging at the affected segment with a loss of segmental lordosis above this level and associated postural ‘sway.’ With the patient prone, hip extension and knee flexion movement tests reveal a loss of co-contraction of the deep abdominal muscles and dominant patterns of activation of the lumbar erector spinae that cause excessive segmental extension–rotation at the unstable level. During forward bending, these patients tend to hold the lumbar spine in lordosis (particularly at the level of the unstable motion segment), with a sudden loss of lordosis at the midrange of flexion that usually is associated with an arc of pain. During return-to-neutral, they tend to hyperlordose the spine segmentally before the upright position is achieved, with pain on returning to the erect posture and the necessity to assist the movement with the use of the hands. Specific movement tests reveal an inability to initiate posterior pelvic tilt independent of hip flexion and activation of the gluteals, rectus abdominis, and external obliques.

Fig. 101.9 Unstable movement zone: extension pattern.

(Adapted from O’Sullivan PB. Lumbar segmental instability: clinical presentation and specific stabilizing exercise management. Manual Therapy 2000; 5:2–12.)

Specific muscles tests also reveal an inability to co-contract the segmental lumbar multifidus with the deep abdominal muscles in a neutral lumbar posture, with a tendency to ‘lock’ the lumbar spine into extension and brace the abdominal muscles. Attempts to isolate deep abdominal muscle activation commonly are associated with excessive activation of the lumbar erector spinae, external oblique, and rectus abdominis and an inability to control diaphragmatic breathing. Palpation reveals a segmental increase in extension and rotation mobility at the symptomatic motion segment.

Lateral shift pattern of microinstability

The recurrent lateral shift pattern of microinstability usually is unidirectional and is associated with unilateral low back pain, usually reported to occur during reaching or rotating in one direction with the spine flexed (Fig. 101.10). This is the same movement direction that patients report as ‘injuring’ their back.

Fig. 101.10 Unstable movement zone: lateral shift pattern.

(Adapted from O’Sullivan PB. Lumbar segmental instability: clinical presentation and specific stabilizing exercise management. Manual Therapy 2000; 5:2–12.)

Their standing posture is similar to that of patients with flexion pattern microinstability, with a loss of lumbar segmental lordosis at the affected level, but with an associated lateral shift at the same level. Palpation of the lumbar multifidus muscles with the patient standing commonly reveals resting muscle tone on the side of the shift, and atrophy and low tone on the contralateral side. The lateral shift is accentuated when the patient stands on the foot ipsilateral to the shift. When walking, the patient has a tendency to transfer weight through the trunk and upper body rather than through the pelvis. Sagittal spinal movements cause a further lateral shift at the midrange of flexion, usually associated with an arc of pain. With the patient supine, rotary and lateral trunk control is lost in the direction of the shift, with asymmetrical leg loading and unilateral bridging, and during four-point kneeling trunk control is lost when one arm is flexed. Sitting-to-standing and squatting movements usually reveal a tendency towards lateral trunk shift with increased weight-bearing on the side of the shift.

Specific muscle testing reveals an inability to bilaterally activate segmental lumbar multifidus in co-contraction with the deep abdominal muscles, with dominance of activation of the quadratus lumborum, lumbar erector spinae, and superficial lumbar multifidus on the side ipsilateral to the shift and an inability to activate the segmental lumbar multifidus on the contralateral side.

Multidirectional pattern of microinstability

The most serious and debilitating of the clinical presentations, multidirectional microinstability frequently is associated with a traumatic injury and high levels of pain and functional disability. Provocative movements are described as multidirectional (Fig. 101.11), all weight-bearing postures are painful, and pain-relieving positions during weight-bearing are difficult to obtain. Locking of the spine commonly is reported after sustained flexion, rotation, and extension postures, and patients may assume a flexed, extended, or laterally shifted spinal posture. Excessive segmental shifting and hinging patterns may be present in all movement directions, with ‘jabbing’ pain and associated back muscle spasm. Assuming neutral lordotic spinal positions is very difficult, and attempts to facilitate lumbar multifidus and transverses abdominis co-contraction (especially during weight bearing) usually are associated with a tendency to flex, extend, or laterally shift the spine segmentally, with associated global muscle substitution, bracing of the abdominal wall, and pain. Palpation reveals multidirectional increased intersegmental motion at the symptomatic level. High levels of irritability and an inability to tolerate compression loading in any position indicate a poor prognosis for conservative exercise management.

Fig. 101.11 Unstable movement zone: multidirectional pattern.

(Adapted from O’Sullivan PB. Lumbar segmental instability: clinical presentation and specific stabilizing exercise management. Manual Therapy 2000; 5:2–12.)

REHABILITATION OF INSTABILITY

A recent focus in the rehabilitation of patients with chronic low back pain has been the training of specific muscles, such as the transverses abdominis, diaphragm, and lumbar multifidus, involved in the dynamic stability and segmental control of the spine based on the identification of specific motor control deficits in these muscles.^79–81 Once the faulty movement pattern or patterns are identified, the individual components of the movement are isolated and the muscles are retrained for functional tasks specific to the patient’s needs. O’Sullivan et al. have reported reductions in pain and functional disability with this exercise training approach in patients with chronic low back pain and a diagnosis of lumbar segmental instability.^76,80,82 In its simplest form, this exercise program represents the process of motor learning; three stages have been described in the learning of a new motor skill (Fig. 101.12).^76,83

Fig. 101.12 Stages of rehabilitation based on a motor learning model (LMS, local muscle system).

(Adapted from O’Sullivan PB. Lumbar segmental instability: clinical presentation and specific stabilizing exercise management. Manual Therapy 2000; 5:2–12.)

SUMMARY

Lumbar segmental instability continues to be a diagnostic challenge. Correlating clinical instability and radiographic instability has been difficult because of the overlap of symptomatic and asymptomatic motion patterns. Additionally, conventional radiography often is insensitive and unreliable in detecting abnormal or excessive intersegmental motion. As the pathomechanics of the lumbar spine have become better understood, however, defining instability in terms of quality of motion throughout the range of motion rather than relying solely on the traditional total range of motion values for diagnosis has been useful in providing a basis by which patients can be evaluated in a segment-specific and individual manner (Fig. 101.13). Lumbar segmental instability can be viewed as a purely movement syndrome: microinstability (without osseous injury) in which directional patterns of motion produce an observed lack of movement control and relative symptoms within the neutral zone or may be associated with other conditions. With increasing understanding of the relative involvement of different muscle groups contributing to lumbar stability, diagnosis-specific rehabilitation protocols are being developed to assist in the assessment and treatment of patients with lumbar instability. Good results have been reported with such an approach using a motor learning model in which the faulty movement pattern or patterns are identified and the components of the movement are isolated and retrained into functional tasks specific to the patient’s individual needs. Scientific trials comparing this approach to other treatment methods are required to validate its efficacy.

Fig. 101.13 Recommended flow diagram for physical examination assessment to determine spinal instability.

(Adapted from O’Sullivan PB. Lumbar segmental instability: clinical presentation and specific stabilizing exercise management. Manual Therapy 2000; 5:2–12.)