The facet joints are a pair of joints in the posterior aspect of the spine. Although these joints are most commonly called the facet joints, they are more properly termed the zygapophyseal joints (abbreviated as Z-joints; also commonly spelled as "zygapophysial joints"), a term derived from the Greek roots zygos, meaning yoke or bridge, and physis, meaning outgrowth. This “bridging of outgrowths” is most easily seen from a lateral view, where the Z-joint bridges adjoin the vertebrae. The term facet joint is a misnomer because the joint occurs between adjoining zygapophyseal processes, rather than facets, which are the articular cartilage lining small joints in the body (eg, phalanges, costotransverse and costovertebral joints). This joint is also sometimes referred to as the apophyseal joint or the posterior intervertebral joint.
As is true of any synovial joint, the Z-joint is a potential source of pain. In fact, the Z-joint is one of the most common sources of low back pain (LBP). The first discussion of the Z-joint as a source of LBP was by Goldwaith in 1911.  In 1927, Putti illustrated osteoarthritic changes of Z-joints in 75 cadavers of persons older than 40 years.  In 1933, Ghormley coined the term facet syndrome, suggesting that hypertrophic changes secondary to osteoarthritis of the zygapophyseal processes led to lumbar nerve root entrapment, which caused LBP.  In the 1950s, Harris and Mcnab  and McRae  determined that the etiology of Z-joint degeneration was secondary to intervertebral disc degeneration.
Hirsch et al were later able to reproduce LBP with injections of hypertonic saline solution into the Z-joints, thus affirming the role of the Z-joints as a source of LBP.  Mooney and Robertson also performed provocative hypertonic saline Z-joint injections and recorded pain referral maps with radiation mainly to the buttocks and posterior thigh. 
Thus, the history and presence of Z-joint pain has been well published. However, despite all of these studies, the diagnosis of Z-joint–mediated pain remains a challenge because no history findings or examination maneuver has been found to be unique or specific to this entity. [8, 9] Schwarzer et al and other authors have reported up to a 45% false-positive diagnostic rate when the physical examination findings are correlated to diagnostic medial branch blocks of the posterior rami. [10, 11, 12, 13, 14]
Authors have concluded that in most cases, Z-joints are not the single or primary cause of LBP. In many cases, Z-joint pain is mistaken for discogenic pain. Thus, many clinicians agree that correlating historical or physical examination findings with pain emanating from the Z-joint is a challenge. This review may help broaden the clinician's knowledge of this entity and may assist in making the diagnosis of lumbosacral facet joint syndrome.
Related Medscape Reference topics include the following:
LBP is the most common musculoskeletal disorder of industrialized society and the most common cause of disability in persons younger than 45 years. Given that 90% of adults experience LBP sometime in their lives, the fact that it is the second leading cause for visits to primary care physicians and the most frequent reason for visits to orthopedic surgeons or neurosurgeons is not surprising. As the primary cause of work-related injuries, LBP is the most costly of all medical diagnoses when time off from work, long-term disability, and medical and legal expenses are taken into account.
The lumbosacral Z-joint is reported to be the source of pain in 15-40% of patients with chronic LBP. Ray believed that Z-joint–mediated pain is the etiology for most cases of mechanical LBP,  whereas other authors have argued that it may contribute to nearly 80% of cases. Thus, the diagnosis and treatment of this entity may help alleviate LBP in a significant number of patients.
International data on lumbosacral facet syndrome have not been clearly established.
The spine is composed of a series of functional units. Each unit consists of an anterior segment, which is made up of 2 adjacent vertebral bodies and the intervertebral disc between them, and the posterior segment, which consists of the laminae and their processes. One joint is formed between the 2 vertebral bodies, whereas the other 2 joints, known as the Z-joints, are formed by the articulation of the superior articular processes of one vertebra with the inferior articular processes of the vertebra above. Thus, the Z-joints are part of an interdependent functional spinal unit consisting of the disc-vertebral body joint and the 2 Z-joints, with the Z-joints paired along the entire posterolateral vertebral column.
In the lumbar spine, the superior articular processes face anterolaterally, whereas the inferior articular processes face posteromedially. The superior articular process has a concave orientation in order to accommodate the more convex orientation of the inferior articular process. The upper lumbar Z-joints are oriented in a sagittal plane, whereas the lower lumbar Z-joints approach a more frontal orientation. Thus, as the lumbosacral Z-joints maintain a progressive coronal orientation, greatest at the S1 level, they are functionally able to resist rotation in the upper lumbar region as well as resist forward displacement in the lower lumbosacral region.
The Z-joint is considered a motion-restricting joint, able to resist stress and withstand both axial and shearing forces. In back extension, the Z-joints, along with the intervertebral discs, absorb a compressive load. In addition, the transmission of the Z-joint load occurs through contact of the tip of the inferior articular process with the pars of the vertebra below. The overloaded Z-joint then causes posterior rotation of the inferior articular process, resulting in stretching of the joint capsule.
If one considers the disc and each of the adjacent Z-joints as an interdependent functional spinal unit, degenerative changes within this 3-joint complex can influence each of the segments. Thus, degeneration of the discs can lead to loss of disc height, resulting in a relative increase in Z-joint load that is found in compression and extension maneuvers. One theory is that these excessive Z-joint loads cause the inferior articular process to pivot about the pars and stretch the joint capsule, in addition to causing rostrocaudal subluxation (ie, Z-joint malalignment). Thus, some authors postulate that Z-joints undergo osteoarthritic changes in response to disc degeneration secondary to changes in loading. 80% of the loading weight is carried through the vertebral bodies and intervertebral discs and 20% by the Z-joints. However, thinning of the intervertebral discs from wear and tear causes more weight bearing through the Z-joints. This in turn results in arthropathy and slight angulation deformity, or a more horizontal orientation, which allows for degenerative spondylolisthesis to occur. Most commonly, this occurs at L4-5 and to a lesser degree at the superior levels.
The Z-joint is a common pain generator in the lower back. The 2 common mechanisms for this generation of pain are either (1) direct, from an arthritic process within the joint itself, or (2) indirect, in which overgrowth of the joint (eg, Z-joint hypertrophy or a synovial cyst) impinges on nearby structures.
The Z-joints are diarthrodial joints with a synovial lining, the surfaces of which are covered with hyaline cartilage, which is susceptible to arthritic changes and arthropathies. Repetitive stress and osteoarthritic changes to the Z-joint can lead to zygapophyseal hypertrophy. Like any synovial joint, degeneration, inflammation, and injury can lead to pain with joint motion, causing restriction of motion secondary to pain and, thus, deconditioning. In addition, Z-joint arthrosis, particularly trophic changes of the superior articular process, can progress to narrowing of the neural foramen. In addition, as is the case for any synovial joint, the synovial membrane can form an outpouching and, thus, a cyst. Z-joint cysts are most commonly seen at the L4-L5 level (65%), but they are also seen at the L5-S1 (31%) and L3-L4 (4%) levels. These synovial cysts can be clinically significant, particularly if they impinge on nearby structures (eg, the existing nerve root).
The neural foramen is bordered by the superior articular process, pars interarticularis, and posterior portion of the vertebral body. Z-joint hypertrophy or a synovial cyst can contribute to lateral and central lumbar stenosis, which can lead to impingement on the exiting nerve root. Thus, Z-joint pain can occasionally produce a pain referral pattern that is indistinguishable from disc herniation.
To understand the pattern of pain generation from the Z-joint, knowledge of the innervation pattern is essential. This pattern is frequently misunderstood even by experienced practitioners. Each Z-joint is innervated by branches of the dorsal ramus, termed the medial branch. The medial branch is 1 of 3 branches of the dorsal ramus, with the other 2 being the lateral branch (which does not exist for the L5 dorsal ramus) and the intermediate branch. The lateral branch innervates the iliocostalis muscle, and the intermediate branch innervates the longissimus muscle. The medial branch innervates many structures, including the Z-joint, but it also innervates the multifidus, interspinales, and intertransversarii mediales muscles, the interspinous ligament, and, possibly, the ligamentum flavum (see image below).
After the medial branch splits off from the dorsal ramus, it courses caudally around the base of the superior articular process of the level below toward that level’s Z-joint (eg, the L2 medial branch wraps around the L3 superior articular process to approach the L2-L3 Z-joint). The medial branch then continues in a groove between the superior articular process and transverse process (or, in the case of the L5 medial branch, between the superior articular process of S1 and the sacral ala of S1, which is the homologous structure to the transverse processes of the lumbar vertebrae). As it makes this course, the medial branch is held in place by a ligament joining the superior articular process and the transverse process, termed the mamillo-accessory ligament (MAL) (see image below).
The MAL is so named because it adjoins the mamillary process of the superior articular process to the accessory process of the transverse process (see image below). The MAL is clinically important because it allows precise location of the medial branch of the dorsal ramus using only bony landmarks, which is essential for fluoroscopically guided procedures.
After passing underneath the MAL, the medial branch of the dorsal ramus gives off 2 branches to the nearby Z-joints. One branch innervates the Z-joint of that level, and the second branch descends caudally to the level below. Therefore, each medial branch of the dorsal ramus innervates 2 joints—that level and the level below (eg, the L3 medial branch innervates the L3-L4 and L4-L5 Z-joints). Similarly, each Z-joint is innervated by the 2 most cephalad medial branches (eg, the L3-L4 Z-joint is innervated by the L2 and L3 medial branches). Some authors have also suggested that the L5-S1 Z-joint has a unique triple innervation; in addition to the expected innervation by the L3 and L4 medial branches, the S1 medial branch emerging from the S1 posterior sacral foramen ascends cranially to also innervate the L5-S1 Z-joint. This has not, however, been consistently reported.
Understanding of this anatomy is crucial for procedures that attempt to obliterate Z-joint–mediated pain by blunting the innervation, whether through anesthesia (eg, a medial branch block) or denervation (eg, medial branch radiofrequency ablation [RFA]).  Practitioners commonly make the mistake of thinking that each Z-joint is innervated by the 2 adjoining medial branches (eg, that the L4-L5 Z-joint is innervated by the L4 and L5 medial branches of the dorsal rami, when it is actually innervated by the L3 and L4 medial branches). Two common reasons are cited for why practitioners make this mistake.
First, in the cervical region, the Z-joints are innervated by the 2 medial branches of the same name (eg, the C3-C4 Z-joint is innervated by the C3 and C4 medial branches), with the transition occurring at the T1-T2 Z-joint, which is innervated by the C8 and T1 medial branches. The second reason practitioners commonly confuse the innervation pattern is because they fail to recognize that the medial branch descends one level to reach the Z-joint. For example, the L2 medial branch courses around the L3 superior articular process, crosses underneath the L3 MAL, and then sends branches to the L2-L3 and L3-L4 Z-joints. Therefore, in a medial branch block, the medial branches closest to the Z-joint are targeted; they simply descended from a higher level.
Moreover, it is important to note that the medial branch of the posterior rami also innervates other posterior back structures. This has several important clinical implications. First, pain relief from anesthetizing the medial branch does not necessarily implicate the Z-joints as the primary pain generator, because one of the other structures innervated by the medial branch may have been the pain generator. Second, denervation of the medial branch by RFA may affect the nerve supply to the multifidus muscle. This is important because lumbosacral radiculopathy is often another consideration in the differential diagnosis of LBP.
One test to confirm the diagnosis of a lumbosacral radiculopathy is electromyography (EMG) of the multifidus muscle. Normally, denervation potentials in the multifidus muscle of a patient with LBP might be interpreted as evidence of a lumbosacral radiculopathy. However, in the context of a patient who has had RFA of the medial branch of the dorsal rami for the treatment of Z-joint pain, an alternative explanation for the denervation potentials in the multifidus would be denervation from the RFA, not from a lumbosacral radiculopathy.
The Z-joints contain nociceptive nerve fibers from nerves of the sympathetic and parasympathetic ganglia, which can be activated by local pressure and capsular stretch. Nociceptive type IV receptors have been identified in the fibrous capsule and represent a plexus of unmyelinated nerve fibers and type I and II corpuscular mechanoreceptors. In addition, encapsulated type I and II nerve endings have been found to be primarily mechanosensitive and likely provide proprioceptive and protective information to the central nervous system.
In addition, the Z-joints have been found to undergo sensitization of neurons by naturally occurring inflammatory mediators such as substance P and phospholipase A2. Peripheral nerve endings release chemical mediators such as bradykinin, serotonin, histamine, and prostaglandins, which are noxious and can cause pain. Substance P has been implicated because of its ability to act directly on nerve endings or indirectly through vasodilation, plasma extravasation, and histamine release. Phospholipase A2 hydrolyzes phospholipids to produce arachidonic acid, causing an inflammatory reaction, edema, and prolonged nociceptive excitation.
In all, many sources of pain can be found at the Z-joint, ranging from degenerative changes to irritated nerve endings (chemical and mechanical) to concomitant nerve root entrapment.
Related Medscape Reference topics include the following:
Athletes involved in nearly any type of sport are susceptible to Z-joint injury. From linemen on a football team, who may sustain repetitive and compressive forces to an extended spine, to baseball players or golfers, who perform repeated spinal rotational maneuvers, lumbosacral facet syndrome can impact athletes in most sports.
Related Medscape Reference topics include the following:
What would you like to print?