Muscular Support of the Spine

Published on 27/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1513 times

Chapter 6 Muscular Support of the Spine

Spinal muscles have long been ignored contributors to spine stability. Paraspinal muscles have been inadequately appreciated and regarded (“lift with your legs, not with your back”), incised with impunity, suffered injury from retraction, neglected through rest/traction, and denervated capriciously. Even in the early 21st century, when bone and collagenous tissue structures (ligaments, tendons, intervertebral discs) are targets of intense basic scientific study, information on spinal muscles often remains empirical, subjective, nonreproducible—and, at worst, harmful. Intended as a compendium of available information and inferences of function, this chapter should challenge the reader to preserve, restore, train, and study the spine.

Muscle, in general, is a highly plastic, adaptable organ. Muscles are classified into three broad categories—striated, smooth, and cardiac muscle—categories that have as much to do with their neurologic control as with their histologic appearance. For the most part, our understanding of muscle form and function is derived from an extensive study of muscles of the extremities. Spine muscles, on the other hand, have changed with each phylogenetic selection that produced vertebrates, mammals, and eventually Homo sapiens. Histologically, muscle form and function are preserved from other species, but spine muscles have importance in the human beyond that seen in other animals.

The muscles that act as the dynamic control mechanism of the spine constitute the largest collective, coordinated group of muscles in the human body. Human spinal muscle makes us unique among species, allowing us to walk upright exclusively—hence the further ability to carry items to a safe place and thus “accumulate excess.” The ability to acquire surplus allows for specialization within a societal structure, impelling the dominance of our species. Without fear of overstatement, our evolutionary success as a species owes everything to the unique structure and function of spinal muscle.

Anatomy

Muscle serves a defined (and seemingly paradoxical) purpose to simultaneously vitalize with movement and protect with strength the central neural communication link between the brain and periphery. In the past, mischaracterizations of the spine as an overrated “electrical conduit” protecting vital message transduction to the limbs have predominated. This disingenuous oversimplification ignores the fact that interactions with the environment, including force generation, dynamic control, proprioception, and balance, benefit from and often require the intricate coordination of spine function.

The dynamic structure of the spine relies upon muscle to animate with motion its series of paired joints and hydraulically pressurized discs. This dynamic control protects the neural elements while maximizing freedom of mobility. A muscle’s function is enhanced by the stability, afferent feedback, and proprioceptive information provided by its associated ligaments, tendons, and joint capsules.¹ This unique combination of motion and stability is demonstrated by the human skill of manipulating objects near the ground from a bipedal stance to lift and carry them to another location. Arguably, the ability to accomplish this “everyday task” has been a key to our success as a species, because the ability to bend efficiently from the waist in combination with squatting and hunkering allows accrual of excess to ensure against environmental pressures like famine and drought. A stable “biomechanical chain” that transfers force efficiently from hands to arms, shoulder girdle, spine, pelvis, and legs and to a stable foot base is essential. In this functional example, the spine musculature acts as a dynamic stabilizer of the biomechanical chain in several ways. It is here that one should recall Panjabi’s description of the interplay of the three subsystems of spinal control: muscle control, passive-restraint control, and neural control.²^–⁴ The cotensioning of abdominal muscles at a distance (force multiplied by a lever arm to generate moment or torque) combined with the collective dorsal force of the erector spinae muscles during flexion-extension allows maintenance of a “balance point.”⁵^–⁷ At each individual motion segment, the interspinalis, multifidus, and, possibly, intertransverse muscles also provide stability through compressive force spanning only one motion segment.^5,^6,^8,⁹ Coupled ventral and dorsal forces have a net compressive force that, in turn, balances motion at the instantaneous axis of rotation for each motion segment, thereby maintaining compression at the disc and minimizing angular change. This net muscle force serves to offset other forces, thereby maintaining the force vector perpendicular to the disc’s plane like the guy wires supporting a tent or flagpole or radio tower.^10,¹¹

In sum, muscles supply dynamic, as well as static, axial compression force to allow for maximal load bearing (for bipedal carrying/lifting) capacity while maintaining function with economy of energy output. Energy is economized via the following mechanical adaptations unique to human musculature: maintaining a balanced plumb line (not working against gravity to maintain posture) with three offsetting curves, sharing tension bands to distribute loads, coupling forces when motion is required, and distributing load/work among the other osteoligamentous static structures.¹²

Form Follows Function

As the father of American architecture, Louis Sullivan stated in 1896, “form ever follows function,” and this is true for the human form as for the American skyline. With this in mind, understanding spinal muscle cannot be complete without acknowledging the role of spine muscles in evolutionary change. Evolutionarily, it appears that the common ancestors of land-dwelling vertebrates (including mammals) were ocean-dwelling animals similar to today’s fish. Living in water, these animals had to contend with very different forces from creatures that live on land. Large paravertebral muscles provide lateral flexion-extension (crossing the sagittal plane) to propel their bodies through water—demonstrated by the lateral tail motion of the fish. This form of locomotion is evolutionarily conserved in amphibians and land reptiles (even with addition of limbs) as demonstrated by the lateral locomotion of species in the order Crocodylia, including the modern alligator or crocodile. Currently extant reptiles propel themselves forward using lateral spinal motion (side-winding). Propulsion is achieved by alternating contraction of spine muscles that in turn creates alternating sagittal convexities of the spine. This repetitive spine motion allows the ipsilateral foreleg to move forward while the contralateral hindfoot (on the concave side) is brought closer to the contralateral foreleg in preparation for the next reciprocal lateral movement that repeats the motion-event contralaterally.

Adaptation (the results of which are not observed in any reptiles alive today) resulted in a 90-degree transformation of spine muscle motion. This characteristic, seen almost universally in mammals (the platypus and echidna being exceptions), presumably provides an advantage during land locomotion, allowing for explosive growth of the class Mammalia. The transformation allows for flexion-extension (crossing the coronal plane) that enables a greater distance per stride (as seen with the horse or cheetah at full run). Interestingly, land mammals that subsequently repopulated the oceans (whales, seals, manatees) maintained their motion orientation through the coronal plane. This form of spinal locomotion resulted in the vertical orientation of the mammalian tail fluke (a 90-degree transformation compared with fish), even though adaptive pressure has resulted in changes to the other extremities that appear similar to fish.

From an evolutionary standpoint, motion is a balancing act. First, form is intended to maximize the functions of swiftly arriving at a food source or a potential mate while evading a predator. On the other hand, the demand for speed must be balanced with metabolic efficiency that allows the species to survive perturbations in the environment. As stated previously, controlling spine motion (like flexion) with muscles alone is inefficient. Moreover, the space required for the abdominal/thoracic contents further limits the potential size of spine muscles.¹³ The evolutionary solution is twofold: (1) strong, elastic dorsal spinal structures (midline ligaments, joint capsules, and lumbodorsal fascia) produce (a) passive restraint, particularly to lumbar spine flexion/extension, and allow (b) static “hanging on the ligaments” subject only to slight, plastic “creep,” but without muscular effort; and (2) a lever arm advantage from quadrupeds to use the dorsal pelvic muscles as simultaneous motors and stabilizers of lumbar extension and lower extremity abduction.¹⁴ Specifically, the gluteus and psoas muscles drive the legs more efficiently when the lumbar spine laterally flexes to provide a passive return of energy expended via reciprocal motion.¹⁵ This is of special importance with respect to lumbar and cervical lordosis during surgical procedures as well as in considering the length of construct: excessive fusion length and/or other violations of biomechanical principles lead to decreased efficiency and a painful, less functional patient. Finally, the interplay of the muscles with static structures for metabolic efficiency may have important though insufficiently studied implications for research into so-called motion-preserving technologies.

In summary, the combination of a dorsal ligamentous complex and powerful muscles of the buttocks and dorsal thighs (along with the psoas muscle contributing to controlling the degree of lordosis—discussed later) permits the spine to function like a crane. The boom is the ligament-stabilized flexed spine, the fulcrum is the hips, and the counterweight is the buttocks (maintaining pelvic position with respect to the femurs). Finally, the structure is vitalized by the pelvic extensor musculature that is analogous to the crane’s engine.^16,¹⁷ This combination of passive and active restraint allows for metabolic parsimony. An important, experimentally observable economy of effort is the tendency of the spine to “hang off its ligaments.” This action is a position of comfort and a metabolic conservation frequently observed in stooped laborers and observable in most normal subjects tested. In other words, normal subjects monitored with surface electromyography preferentially flex forward to end range with the lumbar spine (to the point of myoelectrical silence) before adding the component of hip flexion during the initial act of lifting.¹⁸^–²⁰ Another efficiency created by muscle is the curvilinear structure of the spine. The combination of cervical lordosis, thoracic kyphosis, and lumbar lordosis creates a balance (and though not myolectrically “silent”) requiring minimal muscle output by utilizing the static structure of the thoracolumbar fascia during standing.¹²

The curvilinear structure of the spine that optimizes efficiency is also a prerequisite for human bipedal ambulation and stance. The lumbar lordotic curve converts lateral flexion to torque through the pelvis to the femurs. As noted earlier, this action economizes effort, with upright propulsion leading to a balanced human gait that would be difficult without lumbar lordosis. Conversely, ambulation without lumbar lordosis leads to the shuffling strides of the upright apes whose gait is clearly dissimilar to that of healthy humans (but similar to that of flat back surgical failures). Moreover, the curvilinear structure of the spine permits a greater load to be lifted and carried (so important in human evolution). Spine biomechanical research suggests that cocontraction of spinal and abdominal muscles is the primary generator of the curvilinear structure of the spine that enables greater load bearing than straight-spine models (1200 N vs. 100 N). Furthermore, instantaneous, axial-rotational forces between segments in straight-spine models may lead to rapid failure when the spine is progressively loaded.²¹

This model corroborates observational data of the dynamic contribution of spine muscles to the creation of a compressive-stabilizing force. The cumulative compressive forces applied by the action of muscle, tendon, ligament, and fascia to bony and disc structures enable the spine to withstand greater physiologic forces in sagittal motion as well. This model is analogous to taut guy wires allowing flimsy tent material to withstand 100-mph winds. Tension provided by intrinsic muscle tone and ligamentous passive tension is hypothesized (by the “follower-load” theory) to provide a stabilizing force (in at least the sagittal and coronal planes of motion when standing). This tension directs the force vector to achieve pure compression of the motion segment (which withstands this force largely via the hydraulic force resistance of the disc). The compressive force vector minimizes shear forces implicated as a leading cause of disc degeneration.¹⁰

Finally, the individual contributions of spine muscles can, alternatively, be seen in the context of function dictating form. Instead of viewing spinal musculature in isolation, one may develop an appreciation of the spinal musculature as an efficiently evolved functional unit, improved upon from earlier iterations, and linking all skeletal muscles to act as one functional unit. The cervicothoracic, shoulder girdle, and upper extremity units are linked by paravertebral, abdominal, buttock, pelvic floor, and hamstring muscles to exert specific force vectors that combine with gravity and the constraint of the passive structures to allow carrying and manipulation while simultaneously maintaining bipedal stance or ambulating. The spinal musculature is the crucial link in a complete biomechanical chain that allows lifting and carrying (of greater than one’s own body weight) by the upper extremities while maintaining stable ground contact to haul items out of harm’s way or to a safe location. This ability to carry and hoard excess in turn provides maximal evolutionary advantage in an environmental context. The fine balancing act of performance and metabolic economy can tip over into dysfunction when subtle extrinsic (trauma) and/or intrinsic (fear-avoidance) disruptions evolve into a feed-forward system of dysfunction. This concept is ably demonstrated by Panjabi’s hypothesis of chronic back pain:²²

Sub-failure injuries of the ligaments and embedded mechanoreceptors . . . generate corrupted transducer signals, which lead to corrupted muscle response patterns produced by the neuromuscular control unit. Muscle coordination and individual muscle force characteristics, i.e., onset, magnitude, and shut-off, are disrupted. This results [sic] in abnormal stresses and strains in the ligaments, mechanoreceptors and muscles, and excessive loading of the facet joints . . . inherently poor healing of spinal ligaments, accelerate degeneration of disc and facet . . . over time, may lead to chronic back pain.

Physiology and Microanatomy

The relative resistance of muscles due to redundancy and overengineering belies a complex microstructure. Because muscle functions as the dynamic control mechanism of the skeletal system, its structural complexity allows the tissue to respond to environmental cues to be faster, stronger, or more metabolically parsimonious. This very adaptability has very likely made muscle an overlooked and underappreciated structure.

Muscle incorporates many long, overlapping cells specifically adapted for shortening. Voluntary, or skeletal, muscle is by far the most abundant (by volume) muscle type in humans. Muscles controlling spinal movement, in turn, constitute the largest assemblage of skeletal muscles in the body. Of the various muscle-specific organelles and matrix proteins, the most common constituents are actin and myosin isoforms, which represent approximately 25% to 30% of the total body protein synthesis.^15,²³ This net metabolic consumption underscores muscles’ complexity and versatility, which originates not in its chemistry but in its structure. Sarcomeres, the basic structural units (individual cells) of muscles, are attached end to end to form a muscle filament. Muscle filaments are grouped together in tight formation, with their respective nuclei and organelles pushed to the periphery to form myofibrils.²⁴ Bathing the myofibrils, nuclei, and organelles is a fluid called sacroplasm whose fluctuating electrolyte concentration is controlled by the external, semipermiable lipid bilayer known as the sarcolemma. The myofibril architecture is highly organized, aligning longitudinally within the sarcolemma, which is indented by a motor axon at its myoneural junction. Myofibrils are bundled to form muscle fibers that are, in turn, covered and connected to other muscle fibers by an endomysium. The axial muscle fibers may be only a few millimeters in diameter but can be 5 cm or more in length. Many fibers are bound together by perimesium collagen to form organized fascicles that are bundled together to form what we call muscle.^24,²⁵ Muscle attaches to bone via a collagenous tissue called tendon. The function of this superstructure depends upon the two-way communication (between the alpha motor neuron and muscle and the Golgi-tendon complex and spinal reflex arc) and the variable neural innervation by one motor neuron that may coordinate contraction for anywhere from 15 to 5000 muscle fibers.

The rigidly organized substructure of the myofilament appears as light and dark striations under a light microscope—hence the designation striated muscle for skeletal muscle. Under normal circumstances, contraction of striated muscle does not occur without neural stimulus, whereas contraction of cardiac and most smooth muscle fibers autopropagates, triggering adjacent fibers to contract without neural stimulation. The cellular mechanics of contractions are relatively simple: actin filaments (occupying the light-colored I band at rest) slide over the myosin filaments (found in the A band and interdigitating with the I band at rest) until, with complete contraction, they completely overlap and eliminate the light H band under microscopic visualization. The biochemical reactions are more complex. Contraction initiates with the release of acetylcholine at the myoneural junction, depolarizing the sarcolemma by changing its permeability to sodium and potassium ions. The sarcolemma-induced ion cascade stimulates release of calcium ions, sequestered in the sarcoplasmic reticulum. These calcium ions bind the troponin complex (C, T, and I), inducing a conformational change that uncovers a “sticky” portion of the actin filament. Myosin, fueled by adenosine phosphate molecules (ATP and ADP), binds and unbinds actin to induce the ratcheting of the myosin along the length of the actin filament. The acetylcholine is rapidly hydrolyzed by acetylcholine esterase, and the calcium is rapidly sequestered back into the sarcoplasmic reticulum so that each nerve firing in skeletal muscle is a discrete, pulsed event rather than a sustained spasm. In this way, multiple stimulations of billions of sarcomeres induce the movements we see that form the basis of dynamic control.

When broken down to its constituent biomechanical parts, it appears that the rate-limiting steps to muscle function are myoneural junction integrity, ionic stability, filament cohesion, and energy. On a more holistic scale, series elasticity, motivation, training, endurance, and energy supply become the rate-determining steps of muscle function. Discrete, independent control of muscle fibers (e.g., only a few motor units contracting or muscle control by multiple motor neurons) permits a gradation of contraction that enables—depending on circumstance—voluntary vacillation between refined control or rapid, maximal contraction. The strength of a single contraction, or “twitch,” depends on the number of fibers that contract. The ability to sustain the contraction (endurance) depends on the ability to recruit more muscle fibers with increasingly repeated firing frequency so that just enough fibers are recruited to do the minimum necessary to complete a task (muscle efficiency). Other factors, muscle fiber type or fuel source, may independently affect endurance (ability to sustain a contraction), but recruitment adapts through training and neurocognitive, motivational factors.

The fuel for muscle contraction (as well as for most bodily functions) is phosphate from the disassociation of adenosine triphosphate (ATP) to adenosine diphosphate (ADP). How the ATP is derived and the cost associated with fuel manufacture pays the salary of many professionals (and continues to propel an illicit subculture of pseudoscientists in medicine and nutrition). For the scope of this chapter, the major consideration is whether ATP is produced via hydrolysis of glucose into water and carbon dioxide or via the citric acid cycle (Krebs cycle). The implication of emerging research is that exercise may stimulate a more favorable milieu for local and distant cells through a paracrine effect. Lactic acid, when it accrues in “anaerobic” metabolism, is one of the implicated protein-signaling molecules.

As we go to press, the implications of new basic scientific research in muscle metabolism have not seen wide application in clinical care. Research based on the experimental work of George Brooks, termed lactate shuttle theory, suggests that higher concentrations of lactic acid produced in the skeletal muscles have beneficial local and possibly distant paracrine effects.²⁶^–²⁸ This growing body of research implies that instead of being a “dead-end metabolite” or mediator of muscle fatigue (as was widely published in the 1960s through the 1980s), lactate may be the mediator of beneficial effects seen empirically in training and exercise. Some research even refutes the implication of lactic acid in fatigue and notes that pH effects of hydrogen ion excess are the primary agents of diminished contractile power.²⁷ In total, the lactate ion may serve multiple beneficial roles in stimulating change in body milieu in the presence of muscle exertion to maintain constant energy (via conversion of lactate to glycogen); to recruit new energy sources (gluconeogenesis); to stimulate new vascularity (angiogenesis); and to promote a local cascade of healing, plasticity, and hyperplasia.^26,²⁹ Although it is beyond the scope of this chapter, there is an urgent need for research into this area because the raison d’être of spine physicians is based on activity, strength, and maintaining function after the patient leaves our office.

There are several ways to infer how the microstructure we have described influences the function of a healthy spine. Some analysis has focused on structural composition, enumerating the relative contribution of fiber length, fiber size, and fiber directional orientation to classify muscle. This modeling of physiologic cross-sectional area is combined with geometric calculations from the fulcrum (moment arm) to model idealized function and classify muscle type. Alternatively, the ATPase work of Engel in 1962 initiated a body of research demonstrating distinctly different motor units within skeletal muscle.³⁰ Myotype classification schemes have proliferated based on histology, morphology, or function. Briefly stated, the interaction between the type of myosin heavy chain (ATP-binding site) and actin within individual sarcomeres determines functional differences based on this classification. Furthermore, the rate at which the myosin heavy chains can repetitively bind ATP and release ADP under conditions of physiologic stress defines the function of the sarcomere into one of three broad functional categories.³¹ Type I fibers have a slower twitch response (rate or frequency of a single contraction), with good fatigue resistance and lower tension development (power). Type II muscle displays a fast twitch, with broader recruitment for more forceful tension development, but relatively poor endurance as compared with type I muscle fibers. Type II fibers are often subdivided into type IIA (that still show a fast twitch response, but have a fatigue threshold between type I and type IIB) and type IIB, showing the fastest speed, the greatest recruitment force (power), and the most precipitous onset of fatigue.^32,³³ Though researchers continue to further subclassify fiber types, type I, type IIA, and type IIB muscles remain the basis of the broadest functional class of voluntary skeletal muscles. Structurally, type I fibers have rich capillary beds with high concentrations of mitochondrial enzymes and relatively low concentrations of glycogen and myosin adenosine triphosphatase—making them appear ideally suited for resisting fatigue associated with aerobic activity. The milieu of type II muscles is very different, with high concentrations of glycogen and a ready supply of ATP for fast, strong contractions in a fixed time period. It should be remembered that in gross structure, each muscle is a heterogeneous, woven tapestry consisting of all of the above fiber subtypes. Relative predominance of one particular fiber type is largely based on genetics and anatomic location of a particular muscle. However, one cannot forget the plasticity inherent in muscle and the mutability based on environmental factors of muscle, age of the individual, nutrition, training, demand, and type of exercise.^34,³⁵

In addition to muscle substructural form, there is also the distance from the joint’s axis of rotation. In a simple model, this distance is termed the lever arm or, more correctly, the moment arm. In the case of only one muscle acting on a joint, the moment arm can be represented by the distance of a muscle’s action in relation to the joint’s axis of rotation. In other words, the amount of muscle shortening causes joint excursion through an arc. From this basic knowledge, it is easy to appreciate that even if a muscle is predominantly type II muscle and built for speed and power, it might not translate to rapid joint angular velocity if there is a large moment arm. Instead, in this scenario, the muscle’s activity would be generating high torque at lower angular velocity. The architectural superstructure adds another layer of complexity with multiple intrinsic and extrinsic muscles exerting force to maximize strength and minimize shear, while economizing metabolic expenditure.

Musculature of the Spine Functional Unit

This section offers a brief overview of a compendium of work by McIntosh, Bogduk, Delp, Kamibayshi and Richmond. We refer the reader to these and other sources for more detailed description of the morphometry of individual muscle groups.

Intrinsic Muscles

Erector Spinae Muscles

This large group of interconnected muscles has robust functionality for movement and restraint. It spans the entire spine from the sacrum to the skull. Although the biomechanics of this muscle are still the subject of study, the intricate redundancy of its neural control manifests the importance of this muscle group.³⁶ The innervation arises from the dorsal rami division of the adjacent nerve root that spreads out to coinnervate up to two levels rostral and caudal (four levels total). The intricately redundant neuromuscular control (as opposed to the single-root control seen in limbs) allows one to infer the importance of this structure. The muscle mass can be divided up into four main divisions whose prefix or suffix (lumborum, thoracis, cervicis, or capitis) denote location—but not necessarily division from the whole. These muscles arise from a robust aponeurosis attaching to the sacrum and pelvis. Medially, the spinalis group attaches to the spinous processes. It may be absent in the cervical spine, where it is replaced by semispinalis capitis and cervicis that attach the transverse processes of cervical and upper thoracic vertebrae to the nuchal lines and cervical spinous processes, respectively.³⁷ Lateral to the spinalis are the longissimus muscles: long, robust sarcomeres probably well adapted to generate great force even when stretched beyond their optimal length.³⁸ The most lateral group is the intercostalis, connecting the lumbar anoneurosis to the ribs and the rib fulcrum to the neck and head.

Multifidus Muscles

This group of muscles is deep, short, and powerful, acting with short moment arms to generate significant force. Multifidus muscles span the entire length of the spine in the form of bridging, short, overlapping segments. An individual multifidus has several bands that illustrate its multidirectional function to alternatively control and resist rotation, abduction (lateral flexion), and extension. The fascicle length of a single muscle varies from two to four segments, connecting the mammillary process to the rostral spinous process over two to four segments proximally. In the upper cervical spine, these important muscles connect to the facet capsules, and in the lower cervical spine, they attach to the transverse processes of the upper thoracic spine.³⁸ Like the erector spinae, these muscles have a redundant, multilevel innervation, allowing function to be maintained even if a proximate dorsal ramus is injured.

Deep Muscles

The interspinalis, rotator, and intertransversarii muscles are the deep muscles of the spine whose function is only elementarily understood. These muscles are paired, deep muscles on either side of the spine, spanning one segment to contribute dynamic force to the strong, elastic interspinous ligaments. In the lumbar spine, the intertransversarii consist of a pair of muscles bilaterally, spanning the transverse processes of adjacent vertebrae. The splenius cervicis, semispinalis cervicis, and capitis are deep muscles unique to the cervical spine, connecting the spinous process to transverse processes in criss-crossing, overlapping patterns from the thoracic spine to the cervical vertebrae. Their contribution controls lateral flexion, extension, and, to a lesser extent, rotation. These deep muscles are often cavalierly excised during surgical approaches, to the untold detriment of the patient (especially in longer dorsal fusions that often have adjacent segment kyphosis or failure later). Finally, the rectus capitis group and obliquus capitis group (major/minor, superior/inferior) span C1-2 to control rotation and restraint of this biomechanically tenuous area of fine engineering.

Lateral Control Arms

This group comprises several muscles lateral to the spine with large moment arms. The quadratus lumborum originates on the iliac crest and iliolumbar ligament and obliquely inserts on the lowest rib, connecting to transverse processes of the upper four lumbar vertebrae. The innervation is from the ventral rami of T12-L1-L2-L3 roots. The psoas major muscle attaches to the transverse processes and vertebral bodies of all the lumbar segments and combines with the iliacus (arising from ilium) to form the iliopsoas muscle.³⁶ Though generally thought of as a primary hip extensor (and therefore extrinsic to the spine), the iliopsoas is a primary generator of force ventral to the coronal balance point. Paradoxically, iliopsoas is an intersegmental extensor in the midlumbar spine, even as it produces flexion at the lumbosacral junction in the process of increasing the lumbar lordosis. This action increases lordosis and, like the tent guy-wire model, creates spinal stability during sitting and standing through compressive force.³⁹ Additionally, the iliopsoas muscle doubles the flexion strength and triples flexion dynamic power compared with that of the abdominals alone.⁴⁰ Finally, the contribution of the iliopsoas to lateral flexion is likely responsible for a reciprocal economy of motion in normal gait and restraining shear while sitting.⁴¹ In the authors’ (possibly controversial) opinion, these contributions to spinal stability and control make the iliopsoas an intrinsic spine muscle. The psoas major and iliacus muscles are innervated by the femoral nerve (L2 and L3 root segments innervation with minor contribution of L4) and lie in close proximity to the lumbosacral plexus. Proximal weakness resulting in hip and/or back pain is a consequence of poorly conceived surgical approaches that denervate or devascularize via aggressive retraction. The analogue lateral control muscles in the cervical spine are the sternocleidomastoid and the trapezius muscles. Both of these may be myometrically (categorization of muscles by movement and orientation) divided into three sections, each of which provides flexion, lateral motion, contralateral rotation, and extension based on the direction and length of the fascicles. Like the lumbar lateral intrinsics (iliopsoas and quadratus lumborum), sternocleidomastoid, and trapezius have long moment arms and allow motion while providing a high magnitude of downward force to resist motion. The innervation and control of these two muscles remain debated, with a large motor contribution from the cranial nerve XI (spinal accessory) but proprioceptive, sensory, and possibly motor contributions from the upper cervical root segments. Analogous to lumbar spine motion, the intrinsic control of these lateral control muscles (exerting force through longer moment arms) in the cervical spine is vital for both efficient motion and resistance to shear. More focused investigations of individual muscles and their respective roles are available in other sources.

Extrinsic Spine Muscles

As we have demonstrated, the majority of voluntary muscle in the human skeleton exerts some influence over spine function. Extrinsic to the muscles already discussed are the four layers of abdominal muscles (transversus abdominis, internal obliques, external obliques, and rectus abdominis). These muscles have been the subject of much discussion and popular press—mostly centered on the laudable, but poorly defined notion of core strength. Despite the popularity of core exercises, there is surprisingly scant research to support the focus on abdominal strength for improving spine function or resistance to injury. To be fair, no information credibly demonstrates harm resulting from exercises focused on abdominal core strength either. Additionally, the large muscles of the buttocks that act as hip extensors or abductors (gluteus maximus, medius, minimus) also have intricate contributions to the spine. In a rudimentary analogy, they act as the “engine” and “counterweight” when the spine approximates a “crane” during lifting/bending activity as well as providing the reciprocal coupled motion across the lumbar spine during ambulation. Thigh muscles (biceps femoris, semimembranosus, semitendinosus) dorsally and (rectus femoris) ventrally restrain pelvic translation—thus preventing wasted energy during ambulation as well as providing resistance force when the pelvis serves as a fulcrum during stooping/bending labor. Similarly, the rhomboid major and minor and levator scapulae connect the scapula to the thoracic spine and provide vital scapular stability for lifting or ballistic arm motion. Ventrally, the cervical vertebrae are connected to the ribs by the poorly studied and poorly understood scalenus muscles. This muscle group may serve a purpose similar to one of the iliopsoas’ functions, but its role remains a subject of folklore rather than hard data. In the thoracic spine, the latissimus dorsi and serratus anterior are generally associated with arm movement, but with arms or ribs fixed and stabilized, they may actively contribute to trunk mobility. Structurally speaking, there have been good studies of isolated spine muscle contributions, but in terms of integrated motion, adequate data of either normal motion or the sequence of failure that leads to dysfunction, pain, deconditioning, and, occasionally, disability are lacking.

Motion and Strength—Putting It All Together

George Bernard Shaw once said, “The only man I know who behaves sensibly is my tailor; he takes my measurements anew each time he sees me. The rest go on with their old measurements and expect me to fit them.” By this definition, those of us involved in spine care in general—and in our regard for muscle in particular—behave NON-sensibly. Even preceding Cady et al.’s work, an empirical understanding of the importance of strength and flexibility to overall health existed.⁴² Unfortunately, our rigor in measuring or tracking these fundamental components of function has bordered on lackadaisical. The tendency for nihilism has overcome our best instincts as scientists to rigorously measure what we do and what we advise.⁴³ Moreover, numerous studies have shown that pain is self-serving, and that following the maxim, “if it hurts . . . don’t do it” further reinforces stiffness, atrophy, and psychological fear—exacerbating the “corrupted response patterns” described by Panjabi.²² This leaves physicians in a conundrum, as they do not wish to advocate activity that leads to injury or to lose patient confidence.⁴⁴^–⁴⁷ Only functional measurement to quantify physical deficits will overcome physician nescience when tailoring rehabilitation to meet patient-specific goals. The current expectation of a “one-size-fits-all” evaluate and treat approach lacks sufficient specificity for the physician to help the patient achieve the desirable gains. In 60 years since DeLorme’s groundbreaking work in 1945, we have learned much about the secondary physical changes accompanying immobilization and disuse in the spine and extremities.^48,⁴⁹ Physician intervention combines with spontaneous healing to produce maximum recovery of disrupted collagenous tissues (soft or osseous) in a relatively short period of time—6 to 12 weeks. Exercise science elucidates the effect of training to increasing strength of contraction by enhancing muscular factors, such as muscle size, fiber type, and fiber number, but also (and perhaps to a greater extent) to neural factors.¹⁵ Muscle plasticity is at its apex under the influence of training, and at its nadir with senescence. Between these extremes, certain anabolic hormones (either endogenous or exogenous) may combine with force production to create the characteristic rapid increase in strength and muscle diameter exemplified during hormonal drive of pubescence. Specific exercises, sequences, and frequency of training in healthy normals remain under study. Lost is the understanding of how bulk appearance of muscle translates to function of muscle. The paradox of healthy individual training is demonstrated by data showing that isometric contraction (in which the contracting muscle is not permitted to shorten) is far more effective in increasing muscle bulk. However, exaggerated isometric muscle bulk elevates injury risk and decreases dynamic function with concomitant poor correlation to strength gains. In fact, most research agrees that isotonic and isokinetic training correlate far better with dynamic strength than any isometric regime.

The simplicity of muscle strength gains in healthy individuals (where production of force to failure increases tolerance through training) does not necessarily result in the recovery of coordination, mobility, and force after injury. Injury, pain, and cognitive factors may create a feed-forward system of deconditioning—thereby establishing a pattern of further degenerative change.⁵⁰ Functional testing, though in its infancy and poorly remunerated, provides the opportunity for longitudinal measurement coinciding with functional improvement. Specifically, several longitudinal studies correlate spinal strength performance to imaging (e.g., CT and MRI) findings as well as occupational gains.^16,⁵¹^–⁵³ Based on DeLorme’s work, the obvious relationship between extremity joints and strength of their contiguous musculature in normal, athletic (supernormal), and pathologic (subnormal: traumatic, arthritic, or deconditioned) situations has led investigators to study similar relationships in the spine.^54,⁵⁵ The study of spine muscle strength has suffered from a lack of a gold standard (contralateral limb) against which to test; as well as disagreement over the functional implications of isometric, isotonic, or isokinetic force approximations to real world kinematics. In the end, development of a normative database to assess not just pathologic states, but to verify that rehabilitation has achieved anything meaningful (other than comfort care) is still only sporadically available.

Though “rehab” may be at the outer edge of interest to surgeon readers of this book, a basic familiarity with such seems prudent. Isometric test models employing strain gauges have been in use for over 60 years; however, like the false appearance of strength in a body-builder, isometrics has a wide gaussian distribution when used to predict function versus appearance. Twenty years of data seem to favor isokinetic measures. Multiple papers demonstrate that measurable deficits in strength, endurance, and neuromuscular coordination correlate with dysfunction and disability. Moreover, patients who decrease isokinetic deficits show ability to return to work.^56,⁵⁷ In short, the crux of assessment requires visual analysis of the area below the curve, plotting force in relation to range of motion. The integral of that curve represents work, and its shape has a relation to effort, injury, and deconditioning. Controversy exists regarding the clinical utility of trunk strength testing, in part because of normal human variability and in part because of unrecognized sources of error related to testing procedures.^58,⁵⁹ Despite some controversy, clear decrements in the pathologic states are seen with selective loss of extensor strength compared with flexors and an inability to maintain strength at high speeds.^16,⁵¹^–⁵³ ^,60 By contrast, supernormal individuals or athletes (e.g., female gymnasts, male soccer players, tennis players, and wrestlers) appear to exceed mean torque/body weight strength ratios for the normal population by 15% to 40%. Furthermore, they show no decrease in torque output at high speeds (termed “high-speed drop-off”), which often is the hallmark of pathology, but may be seen in normals, as well. Additionally, supernormals maintain a very stable ratio of extensor to flexor strength (balanced, efficient use of coupled force).^60,⁶¹ The precise cause of reduced strength in the face of some pain-producing pathology remains a mystery. This mystery is heightened with advances in computerization making curve analysis possible to show precise measurements of work performed, power consumed, and torque exerted that may give insight to assessment of “effort.” Because only maximal muscular effort is truly reproducible, variability of curve shape on test-retest may inform an effort factor. Whereas muscle atrophy undoubtedly occurs with prolonged disuse and deconditioning, pain may inhibit neuromuscular function through a nociceptive reflex feedback mechanism. Similarly, psychosocially induced phenomena (e.g., anxiety, fear of reinjury, or depression changing psychomotor responses) may unconsciously attenuate effort, producing submaximal, variable measurements.^62,⁶³ This unrecognized pathologic feedback loop, in turn, hinders optimal outcomes of spine care, which affects the reputation of our field.

Muscle-Sparing Surgery

In the last decade, improvements in technology, visualization, technique, material innovation, imaging, and device performance have led to greater use of lumbar surgical approaches that are defined as “muscle sparing.” These techniques are covered in greater detail in the following chapters, but it is worth examining the claims of muscle sparing. As noted, the lumbar spine is a finely balanced biomechanical wonder that relies on the integration of intervertebral height, joint mobility, proprioception, muscle balance, and osseoligamentous constraint to allow us to function without pain. Though ample redundancy is undoubtedly built into the system, minimizing the disruption of biomechanical integrity hopefully will lead to better functional outcomes for all spine patients.

Lumbar surgery has classically involved extensive dissection of the dorsal muscle, fascia, ligamentous structures, and occasionally joints. The dissection is even more extensive when arthrodesis (with or without instrumentation) is performed. Unintended consequences of denervation, compression, retraction, or vascular injury have led to other biomechanical sequelae.^16,⁶⁴ Iatrogenic muscle injury related to the aforementioned has been documented histologically, histochemically, and electrophysiologically.⁶⁵ Preventive measures of back muscle injury after dorsolumbar spine surgery have been conducted in rats.^16,⁶⁴^–⁶⁶ It has been hypothesized that this iatrogenic injury may result in instability of the spinal motion segment, loss of the previously described biomechanical balance, and straightening of the lumbar lordosis. All of this may result in a dysfunctional motion segment and pain. Decreased morbidity by sparing vulnerable structures may serve to maintain mobility and function, thereby improving outcomes.

The dorsolateral approach first described by Cloward has been controversial through the years.^67,⁶⁸ The extensive morbidity of the traditional muscle stripping approaches led to the development of muscle-sparing approaches to the spine. Most such surgery descriptions are based on the approach originally characterized by Wiltse et al.⁶⁹ When performed judiciously, the paramedian, or Wiltse, approach allows the erector spinae to be split along the aponeurosis, thus “sparing” the muscle, fascia, and some of the ligamentous structures. Until recently, much larger incisions were still required to perform multilevel decompression or arthrodesis.

Foley and Smith receive credit for describing a percutaneous/minimal access endoscopic approach to the lumbar intervertebral disc.⁷⁰ Subsequently, the field exploded, and similar MIS techniques may be applied to virtually every region of the spine. The distinguishing feature of this procedure utilizes the a technique of muscle dilation through a small paramedian incision with multiple dilators until a final tubular retractor is “docked” on the area of surgical interest (e.g., lumbar lamina). The marketing of these methods cite decreased trauma, dissection, blood loss, and pain. However, with increased pressure from tubular retraction denervation/devascularization of the deep muscles is still possible. Additionally, compared with a standard microdiscectomy, only the approach differs. Unfortunately, minimally invasive tubular approaches have not shown improved long-term outcomes over traditional microdiscectomy. Moreover, at least one randomized controlled study shows worse outcome scores for leg or back pain following endoscopic, tubular discectomy compared with standard microdiscectomy.⁷¹ For the most part, the beneficial data associated with the tubular approach are measured in decreased hospitalization time and medication usage.^72,⁷³

Muscle-sparing technology and approaches have been applied to lumbar arthrodesis. Minimally invasive fusion techniques through tubular retractors were popularized with transforaminal lumbar interbody fusion and now are widely applied to multiple approaches for interbody fusion.⁷⁴ Although long-term outcomes with minimally invasive arthrodesis are indistinguishable (and more expensive) than standard fusion, there are data that short-term outcomes with muscle sparing are superior for pain, time to ambulation, hospital length of stay, and medication usage.⁷⁵ The implication, awaiting literature confirmation, is that earlier mobilization because patients hurt less may in fact lead to better outcomes. We still await associations of outcomes with studies of muscle bulk, isokinetic strength, biomechanical integrity, and histologic appearance.

The ventral transperitoneal approach has been described since the 1930s in various iterations.⁷⁶ A spine surgeon can achieve (indirect) decompression, stabilization, fusion (at a variable rate), and now motion preservation though this approach. Often these surgeries lose the benefit of dorsal muscle preservation when surgeons later elect for additional dorsal stabilization. With newer instrumentation and biomechanical wisdom (outlined later in this book) these surgeries are often performed as a stand-alone ventral procedure. Additionally, retraction can damage abdominal and ventral lumbar muscles that play a key role in maintenance of proper spine balance. The ventral approach has been modified to go retroperitoneally with laparoscopic devices that allow the perceived (unconfirmed) benefit of splitting of the abdominal musculature to provide for more rapid postoperative healing and less perturbation of the abdominal viscera, but often at the cost of crossing and possibly injuring the psoas muscle, which may have a very important role in lumbopelvic coordination of functional tasks. As of the writing of this chapter in 2010, the laparoscopic anterior lumbar interbody fusion has largely been abandoned in the United States because of lack of improved outcomes and increased complications. Keeping in mind the advantages of protecting lumbar musculature and maintaining mobility, there is hope that the functional outcomes of surgical spine procedures will continue to improve.