Causes and Sources of Chronic Thoracic Pain

From WikiMSK

Written by: Dr Jeremy Steinberg – created: 25 April 2025; last modified: 25 April 2025

This page is probably complete!
It is awaiting peer review


Chronic thoracic spine pain is an often perplexing clinical problem. While less common than neck or low back pain, mid-back pain is still significant – about 15% of people report thoracic spine pain at any given time.[1] Importantly, thoracic pain is sometimes associated with serious underlying pathology more often than neck or low back pain.[2] Understanding the sources versus the causes of thoracic pain is critical for an anatomical diagnosis (a “reductionist” approach advocated by Bogduk.[1] In this context, “source” refers to the specific anatomical structure generating nociceptive signals, whereas “cause” refers to the pathophysiological process affecting that structure (e.g. degeneration, inflammation, fracture). For example, a thoracic zygapophysial (facet) joint may be the pain source, while osteoarthritic degeneration of that joint is the cause.

Overview and Postulates

Bogduk has outlined four key postulates (criteria) that must be satisfied for any structure to be credibly deemed a source of spinal pain[1]

  • Innervation: The structure must have a nerve supply capable of transmitting pain.
  • Provocation in Normal Volunteers: Stimulating the structure (e.g. via injection or distension) in asymptomatic individuals should reproduce pain similar to that seen in patients.
  • Pathological Susceptibility: The structure should be susceptible to injuries or diseases known to cause pain (e.g. degeneration, inflammation, trauma).
  • Diagnostic Confirmation in Patients: There must be evidence that in patients with pain, the structure can be identified as the source (for instance, by relief of pain with targeted diagnostic blocks or provocation tests)

Using these postulates, researchers (notably Bogduk and colleagues) have investigated various thoracic structures as possible pain generators. The thoracic spine’s complex anatomy – including the rib articulations – and the potential for referred pain from visceral organs or the cervical region make pinpointing a pain source challenging[3][4]. Nevertheless, several anatomical structures have been identified as credible sources of chronic thoracic pain. In the sections below, we review each of these in turn – detailing their nerve supply, evidence as pain generators, common pathological causes of pain, and any epidemiological insights. We also discuss important red flag causes of thoracic pain (fractures, infections, malignancies) that, while less common, must be recognized due to their seriousness.

The breadth of evidence for everything in this article is much lower than for the cervical and lumbar spine.

Thoracic Intervertebral Discs

✅Innervated: The intervertebral discs of the thoracic spine are well-innervated in their outer annulus fibrosus by the sinuvertebral nerves (recurrent branches of the spinal nerves). Sensory nerve fibers penetrate only the outer layers of the disc under normal conditions,[5] which is sufficient to mediate pain if the outer annulus is irritated or damaged. This satisfies the innervation criterion: thoracic discs have the neural supply to be pain sources.

❓Experimental Pain and Volunteer Studies: It is challenging to ethically provoke disc pain in normal volunteers. However, by analogy to lumbar discs, we know that pressurizing healthy discs can elicit pain. For instance, asymptomatic volunteers subjected to high-pressure discography in the lumbar spine report mild back pain when the disc is distended.[6] By extension, a thoracic disc with a stimulated or irritated outer annulus could generate mid-back pain in a similar fashion. Clinically, provocation discography has been attempted in the thoracic region to identify discogenic pain, albeit with caution and only weak recommendations due to limited research.[7]

✅Pathology Known to be Painful: Thoracic discs are susceptible to degenerative changes (intervertebral disc degeneration, annular fissures) and less commonly to herniation. Disc degeneration or internal disruption can sensitize nerve endings in the annulus, leading to chronic discogenic pain. Thoracic disc herniations are relatively rare – only about 0.25–1% of all disc herniations occur in the thoracic spine[8] – but they can cause pain locally and sometimes neurologic symptoms. Degenerative changes in thoracic discs (such as desiccation or osteophyte formation at endplates) are frequently seen on imaging of older adults, though often asymptomatic. When such changes become painful, the cause is the degenerative process or disc injury, and the source is the disc itself. Notably, pathologic innervation can increase with degeneration: nerve fibers may grow deeper into a degenerated disc, including the nucleus, in disease states[5], potentially amplifying pain.

The disc-endplate complex and the adjacent vertebral body itself be a source of pain in the lumbar spine (called Vertebrogenic Pain) from Modic changes. The same process seems to be far less common in the thoracic spine. On the other hand, "dry" endplate pathology as seen in Scheuermann’s disease is associated with mid-back pain in teenagers.

✅Identified in Patients: Identifying a thoracic disc as the pain source in patients is difficult but has been reported. Patients with chronic thoracic axial pain (without radiculopathy) sometimes undergo thoracic discography. Pain reproduction with disc provocation, especially if concordant with the patient’s typical pain and seen at a degenerated level, suggests a discogenic pain source.[9]

Thoracic Zygapophysial Joints (Facet Joints)

✅Innervated: The thoracic zygapophysial joints (often simply called facet joints) are true synovial joints between articular processes of adjacent vertebrae. They are richly innervated by the medial branches of the dorsal rami of the segmental spinal nerves. Each thoracic facet joint receives a dual innervation: one medial branch from the same level and one from the level above.[10] This overlapping innervation means a single facet joint can refer pain over a somewhat diffuse area. The joint capsules contain free nerve endings, including nociceptors, as demonstrated histologically (thoracic facet blocks-bent needle technique.indd). Thus, thoracic facets clearly fulfill the innervation requirement for pain sourcing.

Thoracic ZA Joint pain patterns in normal volunteers. From Dreyfuss 1994 [11]

✅Experimental Pain in Volunteers: Facet joints have been studied by using induced pain in normal volunteers. In the thoracic region, Dreyfuss et al. performed provocative intra-articular injections of contrast medium in asymptomatic volunteers; about 72% of stimulated thoracic facets produced pain (beyond just needle prick), often with a distinctive referred pain pattern.[11] Thoracic facet joint pain is felt in the parasagittal region and doesn't cross the midline, nor does it go to the chest. There is significant overlap between the referral zones of adjacent segmental levels; most areas fell within the potential referral zones of 3 to 5 different facet joints. Conclusion from volunteer study: The thoracic zygapophysial joints are capable of causing both local and referred pain, confirming their candidacy as pain generators. This mirrors findings in the cervical and lumbar spine, where facet stimulation produces neck or low back pain with characteristic referral to head, shoulder, or buttock regions. This experiment satisfies Bogduk’s second postulate by reproducing clinically relevant pain from facet joints.

❓Pathology Known to be Painful: Thoracic facet joints are prone to osteoarthritic degeneration (facet arthrosis) as part of the normal aging process. They sustain mechanical load, especially in extension and rotation of the thoracic spine. Over time, cartilage wear, osteophyte formation, and capsular laxity can occur. Facet joint pain may also arise after trauma (e.g. a thoracic spine strain causing capsular injury) or in conjunction with adjacent segment degeneration (as in long-standing spinal deformity where facets bear abnormal stress). Cadaveric studies and imaging in older individuals commonly show facet osteoarthritis in the thoracic spine; while often asymptomatic, in some patients it correlates with their pain distribution. Curiously despite all this, in the cervical and lumbar spine which has received a lot more study than the thoracic spine, the evidence does not support osteoarthritis as a cause of facet joint pain.[12]

✅Identified in Patients: Using controlled diagnostic blocks (medial branch nerve blocks to anesthetize the facet joint nerves), investigators have shown that a significant fraction of chronic thoracic pain cases are facet-mediated. In a large clinical study, Manchikanti et al. found that approximately 42% of patients with chronic thoracic spine pain obtained relief with controlled facet joint blocks, indicating facet origin.[1] However the accuracy of diagnostic medial branch blocks in the thoracic spine is poor due to variable nerve course.

Costotransverse Joints

✅Innervated: The costotransverse joints are the articulations between the tubercle of each rib and the transverse process of the corresponding thoracic vertebra. These are synovial plane joints present from T1–T10 (ribs 11 and 12 lack costotransverse joints). They lie just lateral to the facet joints and are supported by the costotransverse ligaments. The innervation of costotransverse joints is shared with the costovertebral joints – through the lateral branches of the dorsal rami at the corresponding levels. Thus, like facets and costovertebral joints, the costotransverse joints have a segmental nerve supply rich in nociceptive fibers. Indeed, Young et al. note that the dorsal ramus passes between the costotransverse joint and rib neck, providing articular branches en route to the joint capsule..

Composite diagram of costotransverse joint pain patterns in normal volunteers.[13]

✅Pain in Volunteers: To directly evaluate costotransverse joint pain referral, Young et al. (2008) conducted a study injecting contrast and irritant into asymptomatic individuals’ costotransverse joints[13]. They performed 21 intra-articular injections across various levels (T2 through T7) in 8 volunteers. The results showed that about 14 out of 21 injections (67%) produced a distinct pain sensation beyond the prick of needle placement. The evoked pain was typically described as a deep, dull ache or pressure around the area of the joint. Notably, the pain remained fairly local to the joint in most cases – patterns were “superficial to the injected joint,” meaning the volunteer felt pain mainly at the back near that rib’s level. Referred pain was minimal: only injections at the T2 costotransverse joints produced any spread beyond the immediate area, referring a dull ache a couple of segments above and below (likely due to overlap in innervation in the upper thoracic region). The pain was fairly focal, there were no reports of chest wall or arm pain from these injections.

❓Pathology Known to be Painful: The costotransverse joints bear significant mechanical stress during thoracic movements (especially rotation and lateral bending) as they couple rib motion with spinal motion. Like any synovial joint, they can develop osteoarthritis. Degenerative changes in costotransverse joints have been observed, particularly in conjunction with costovertebral joint arthritis in older patients (as part of generalized spondylosis). In certain inflammatory arthritides (e.g. ankylosing spondylitis), costotransverse joints may also fuse or show erosion, though the costovertebral joints are more frequently noted. A traumatic cause of costotransverse joint pain can be direct or indirect – for example, a severe rib fracture near the tubercle could disrupt the joint, or heavy lifting might strain the costotransverse ligaments (analogous to a sprain). There is also a clinical concept of costotransverse joint dysfunction, sometimes discussed in manual therapy literature, where a subtle malalignment or hypomobility of the rib-transverse process articulation leads to local pain and tenderness (often respond to mobilization or exercise). Despite all of this, it isn't known whether any of these pathological processes are associated with pain.

❓Identified in Patients: There is less research on costotransverse joints than facet joints, limited to isolated case reports, none with comparative controlled diagnostic blocks.[14] They are frequently included in the differential diagnosis of chronic thoracic spine pain but whether this is based in truth is unknown.

Costovertebral Joints

✅Innervated: The costovertebral joints (also called costocorporeal joints) are the articulations between each rib head and the thoracic vertebral bodies (and intervertebral disc) at the same level. Ribs 2–9 articulate with two adjacent vertebrae and the intervening disc, whereas ribs 1, 10, 11, 12 have a single-vertebra articulation. These joints are stabilized by strong ligaments (radiate and intra-articular ligaments). Costovertebral joints are innervated by the lateral branches of the dorsal rami of thoracic spinal nerves (approximately C8 through T11 levels supply the upper to lower joints). This innervation is segmental, providing a neural pathway for rib joint pain to be perceived in the spine or chest. There is a nerve supply to costovertebral joints, including mechanoreceptors and nociceptors, implying these joints can sense and transmit pain.[15]

❓Pain Generation (Volunteer or Clinical Evidence): Direct experimental induction of pain in costovertebral joints is more difficult (they are deep and adjacent to vital structures), and there are no well-known volunteer injection studies specifically isolating the costovertebral joints. However, given the shared innervation with the costotransverse and facet joints, it is plausible that stimulating a costovertebral joint would produce pain in a similar distribution (local thoracic pain with possible referred pain along the rib or to the back), but is yet to be tested.

❓Pathology Known to be Painful: The costovertebral joints are synovial joints and can undergo degenerative changes (osteoarthritis) similar to other spinal joints. In the elderly, costovertebral osteoarthritis (OA) is common on imaging – one paper notes it is a common degenerative finding in geriatric patients.[16] Such degeneration can cause joint space narrowing, subchondral sclerosis, and osteophytes at the rib head–vertebra interface. These changes may or may not be painful, but in some individuals they clearly produce chronic thoracic pain. In spondyloarthropathies (like ankylosing spondylitis), costovertebral joints are frequently inflamed and can eventually fuse, contributing to chest wall pain and restricted expansion. Approximately 13% of patients with axial spondyloarthritis show costovertebral inflammation on MRI[17], and advanced cases can develop ankylosis of these joints. But this pathological process is not necessarily relevant to most cases of chronic thoracic pain.

❓Diagnostic Confirmation: There is no widely practiced equivalent of the facet block or discogram for these joints, partly due to technical difficulty and overlap with adjacent structures. There are case reports demonstrating confirmation[18] The author of this article had a case of man with a painful 12th costovertebral joint that was hot on SPECT and CT guided injection provided 100% relief from the anaesthetic, and two months relief from the steroid. The diagnostic and therapeutic effect was reproducible on repeat of the procedure. He had focal pain and tenderness. Other case reports make the diagnosis clinically not through diagnostic block.[19]

Somatic Referred Pain from the Cervical Spine or Shoulder Girdle

Not all pain felt in the thoracic region originates from thoracic structures. It is well known that pain can be referred from other musculoskeletal sources via neural convergence mechanisms. In the context of thoracic spine pain, two common regions to consider are the lower cervical spine and the shoulder girdle. In these cases, the cause of pain lies in the cervical or shoulder tissues, but the perceived pain (the symptom) is felt in the upper or mid back. By strict definition, the source of nociception is not actually in the thoracic spine at all, but patients and clinicians may initially localize the pain to the thoracic area, which is why these are important to include.

Cervical Spine Referred Pain: The lower cervical intervertebral discs and facet joints frequently refer pain to the upper thoracic region. Bogduk and colleagues have documented this in detailed mapping studies. Stimulation of the C5/6 and C6/7 facet joints in normal volunteers can cause pain down into the upper thoracic spine also over the shoulder and into the upper scapular area.[20] Referral to the upper thoracic spine has been confirmed in symptomatic patients.[21][22] Similarly, cervical disc and/or nerve root pathology can refer pain to the upper thoracic spine.[23][24] In clinical practice it is very common to identify the lower cervical spine as the source of upper thoracic spine pain.

The physiological basis is that cervical and upper thoracic spinal nerves converge on some of the same second-order neurons in the spinal cord (especially around the C7–T1–T4 segments), so the brain can misinterpret cervical-origin pain as coming from the back or scapula.

Shoulder Girdle Referred Pain: Shoulder structures can refer to the suprascapular region, but it is unusual to refer to the thoracic spine proper.[25][26]

Other Anatomical Sources of Thoracic Pain

Beyond the major joint and disc structures discussed, other tissues in the thoracic region can generate pain. These secondary sources are also important to recognize, as they may be the culprit in certain cases or contribute to pain alongside the primary sources.

Interspinous and Supraspinous Ligaments

The interspinous ligaments (connecting adjacent spinous processes deep between them) and the supraspinous ligament (running along the tips of the spinous processes) stabilize the posterior column of the spine. They are innervated by medial branches of the dorsal rami (the same nerves that supply the facet joints). In fact, the facet joints and these midline ligaments share innervation, which means they can both be pain generators and may refer pain to similar areas.[10] Early experimental work by Kellgren in 1939 showed that injecting hypertonic saline into the thoracic interspinous ligaments produced a characteristic pattern of referred pain in subjects. The induced pain was diffuse and could spread segmentally, indicating that these deep ligaments, when irritated, can cause pain radiating away from the spine.[27]

Whether these structures are implicated clinically is unknown. Because of the shared nerve supply, pain from these ligaments might be relieved by a medial branch block that also numbs the facet joint – thus a patient with apparent facet pain relief might also have had interspinous ligament pain that responded. (See also Interspinous Oedema)

Ribs and Related Tissues

The ribs are an integral part of thoracic anatomy, and pathology of the ribs or their junctions can cause pain that patients perceive as “thoracic spine” pain. One must distinguish true spine pain from rib pain, but there is overlap. Some notable rib-related pain sources include:

  • Rib Fractures: A rib fracture, especially if it doesn’t heal properly (nonunion or fibrous union), can lead to persistent pain along the rib and at the fracture site. Posterior rib fractures (near the costovertebral junction) can be mistaken for spine pain. Patients will have tenderness over the rib and pain exacerbated by thoracic motion or deep breathing. In the subacute phase, fractures cause sharp pain, but chronically they can cause an aching pain at the fracture edges. Multiple rib fractures (as in old trauma) can alter mechanics of the chest wall and cause chronic pain with movement. While simple rib fractures usually heal in 6–8 weeks, a subset of patients may develop chronic pain at the fracture site. Bogduk’s criteria: ribs are innervated by intercostal nerves; injuring a rib obviously causes pain; ribs are susceptible to fractures; and one can confirm by, say, an intercostal nerve block that numbs the rib area and relieves pain, confirming rib origin.
  • Costochondral and Sternocostal Junctions: Although anterior (at the sternum) rather than in the back, costochondritis can cause pain that radiates around to the back. Typically, costochondritis presents with anterior chest wall pain at the costal cartilages, but patients sometimes feel a band of pain along that rib to the spine. This is a referred pattern along the intercostal nerve. While the source here is the costosternal junction (and thus outside the spine proper), it’s worth noting as a differential for thoracic pain, especially when the patient reports pain that wraps around from front to back.
  • Slipping Rib Syndrome: This is a condition where the cartilaginous end of a lower rib (8–10) subluxes or hyper-mobilizes, irritating the intercostal nerve. It often causes lower chest or upper abdominal pain, but it can also manifest as pain at the thoracolumbar junction or lower thoracic spine due to the shared innervation and the patient’s difficulty in pinpointing the pain. The cause is a connective tissue laxity or tear at the costal margin. Again, not a thoracic spine source per se, but in the differential for thoracic region pain.
  • Scapulocostal musculature and bursae: As mentioned in shoulder referred pain, structures like the scapulothoracic bursa (between the scapula and ribs) can become inflamed (scapulothoracic bursitis or “snapping scapula”). This causes a focal pain at the medial scapula, which some patients think is “in the spine” near that level. The source is the inflamed bursa or irritated scapular periosteum. Likewise, trigger points in the rhomboid muscles can cause pain along the medial scapula border, simulating a thoracic spinal ache.

Red Flag Conditions Causing Thoracic Spine Pain

“Red flag” conditions are serious pathologies that can present with back pain and warrant urgent attention. In the thoracic spine, the key red flags include vertebral fractures, spinal infections, and malignancies. These are not common causes of chronic thoracic pain in the general population, but they are vitally important to recognize because the pain from these causes signals significant underlying disease requiring specific management. Unlike the mechanical sources discussed above, red flag causes typically involve structural failure or systemic illness affecting the spine. Below, we outline each and their relationship to thoracic pain.

Vertebral Fractures

Vertebral compression fracture pain is typically midline, deep, and aching – consistent with basivertebral nerve distribution. It is common for patients to have chronic pain following apparent healing of the fractured vertebra. Some speculate this is from malalignment (see Sagittal Balance of the Spine). Some have found that the pain in chronic cases may arise from the facet joints (either from associated injury or related to differential loading)

A vertebral fracture in the thoracic spine – most often a compression fracture of a vertebral body – is a significant cause of acute pain, and in many cases can lead to chronic pain. Such fractures can occur due to major trauma (e.g. a fall or car accident) or, very commonly, due to osteoporosis with only minor stress (a fragility fracture). These fractures often involve the mid-thoracic vertebrae (T7–T8) or the thoracolumbar region (T12–L1), as these areas undergo significant stress.

When a thoracic vertebra fractures (usually collapsing anteriorly in a wedge shape), the immediate result is acute pain and kyphotic deformity. Over time, the acute pain may subside, but a substantial number of patients are left with chronic mid-back pain at the fracture level or adjacent levels. The source of pain in vertebral fractures is multifactorial: the damaged bone and periosteum (richly innervated by the basivertebral nerve) could potentially transmit pain, and there could be sagittal malalignment leading to facet joint strain above and below, or even simply associated injury to the facet joints. Also, the healing process (if incomplete) can leave pseudoarthrosis or nonunion in the bone that continues to hurt. Patients with multiple old fractures may develop a hyperkyphosis (“dowager’s hump”).

It’s worth noting that in chronic thoracic pain patients without a history of trauma, occult compression fractures should be considered – especially in older individuals or those on long-term corticosteroids. Sometimes an old fracture is only found on imaging done for persistent pain. From an epidemiological standpoint, thoracic back pain in an elderly patient has a higher likelihood of being due to a fracture (or malignancy) than in a young patient. Also, vertebral fractures are a red flag because they can lead to neurologic compromise if severe (spinal cord compression, though thoracic cord compression from an osteoporotic fracture is not common unless it’s a burst fracture).

Infections (Spondylodiscitis)

Spinal infections involving the thoracic region – termed spondylodiscitis when both vertebra and disc are involved, or vertebral osteomyelitis – are another red flag cause of thoracic pain. These infections are relatively rare but very serious. The thoracic spine is the second most common region affected (after the lumbar spine)[28] Approximately 30% of hematogenous spinal infections occur in the thoracic spine, which is a substantial proportion. A classical example is Pott’s disease (tuberculous spondylitis) which has a predilection for the lower thoracic spine. Bacterial (pyogenic) spondylodiscitis (often due to Staphylococcus aureus) can occur at any level, including thoracic, especially in patients with risk factors like IV drug use, endocarditis, diabetes, or immune suppression.

Pain from infection: The hallmark of spinal infection is persistent, unrelenting pain often accompanied by systemic signs (fever, chills, weight loss), though in up to one-third of cases fever may be absent. The pain is typically a deep aching pain in the spine that is present even at rest and often worse at night (due to inflammatory cytokine activity) (Evaluation and treatment of spinal metastases: an overview). Movement tends to aggravate it, and patients often have localized tenderness over the affected segment. In thoracic spondylodiscitis, patients might present with mid-back pain that gradually worsens over weeks, sometimes with radicular pain wrapping around the chest if abscess or inflammation irritates the intercostal nerves.

Spinal infection is a red flag because it can lead to catastrophic outcomes if missed (progressive vertebral destruction, spinal deformity, or epidural abscess causing spinal cord compression). In the thoracic region, an epidural abscess can compress the spinal cord, leading to myelopathy or paralysis.

Clinically, red flags that raise suspicion of spondylodiscitis in a thoracic pain patient include: unexplained fever, history of bacteremia or recent infection (e.g. urinary, skin), IV drug use, immune compromise, or pain that is constant, severe, worse at night, and not relieved by rest (distinct from mechanical pain that often improves with rest). Laboratory markers (elevated ESR/CRP) are usually present. But it’s the pain out of proportion and unremitting nature of the pain that often prompts imaging. MRI of the thoracic spine will typically show disc space infection with vertebral marrow edema and possible abscess.

Malignancy (Primary or Metastatic)

Malignancy involving the spine is another major red flag. Metastatic cancer to the spine is far more common than primary spine tumors. The thoracic spine is the most frequent site for skeletal metastases in the spine – about 70% of metastatic spinal tumors are found in the thoracic region.[29] This is because many common cancers (breast, lung, prostate, kidney) have a propensity to spread to the axial skeleton via Batson’s venous plexus, and the mid-spine receives a high volume of this venous flow. Spinal metastases are actually the most common tumors of the spine (accounting for 70–90% of all spinal tumors).[30] In autopsy studies, up to 30–40% of patients with solid tumors have evidence of vertebral metastasis. Clinically, about 10% of cancer patients will develop symptomatic spinal metastases.

When metastases (or a primary bone tumor like myeloma) involve the thoracic spine, pain is a predominant symptom. Cancer-related spine pain often has certain characteristics: it is typically a constant, dull pain that can be worse at night or worse with rest (the classic “night pain” of cancer)[31], and not clearly linked to motion or positioning (though movement can aggravate it as well). This contrasts with mechanical back pain which often improves with rest. The pain may escalate over time as the tumor enlarges. Often, patients describe it as deep-seated. Because metastases frequently affect the vertebral body (and pedicles), the pain arises from the bone (basivertebral nerve) and possibly from encroachment on other structures (e.g. a tumor expanding to press on a nerve root will cause radicular pain).

Primary malignancies like multiple myeloma (plasma cell cancer) commonly involve multiple thoracic vertebrae, causing diffuse mid-back pain. Other primary tumors (chordoma, osteosarcoma, etc.) in the thoracic spine are rare but can also present with pain and often neurological symptoms if large.

Epidemiology and clues: In an older adult with a history of cancer (especially breast, lung, prostate, thyroid, kidney – the common metastasis sources), new thoracic back pain should be assumed metastatic until proven otherwise. Even without a known cancer, features like unexplained weight loss, failure of pain to improve with conservative therapy, age > 50 with new pain, or pain that’s worse supine (some patients with metastases prefer to sit up because lying down increases internal vertebral pressure) all raise suspicion. Sometimes the only symptom of an otherwise asymptomatic cancer is back pain – for example, an occult multiple myeloma or a metastatic renal cell carcinoma can initially present as mid-back pain from a pathological fracture.

Imaging (X-ray may show lytic or blastic changes, MRI is more sensitive) helps identify suspicious lesions. A telling sign on plain film or CT is a pedicle erosion (“winking owl” sign) in the thoracic spine, often from metastatic cancer. Pain from malignancy often progressively worsens despite rest, whereas benign mechanical pain might plateau or fluctuate.

Spinal cord risk: A tumor in the thoracic spine can also lead to cord compression – thoracic spine is a common level for metastatic epidural spinal cord compression. So any signs of myelopathy (leg weakness, gait disturbance, bowel/bladder changes) combined with thoracic pain are an emergency and very suggestive of an epidural tumor or fracture from metastasis.

From a pain source perspective, what exactly is hurting in malignancy? Typically: the vertebral body (and possibly posterior elements if involved). The tumor can also induce a local inflammatory reaction and destabilize the spine, causing secondary facet pain or muscular spasm. But principally, it’s bone pain – often described as gnawing or constant.

Malignancy-related thoracic pain is a must-not-miss cause. Although only a minority of thoracic pain patients have cancer, the high prevalence of spine metastasis among cancer patients means every clinician stays alert for this. Studies have shown that about 20% of patients with back pain red flags actually have a serious pathology, and malignancy is one of the top ones (others being fracture).[32] The thoracic location of pain, especially if atypical, is sometimes a clue; guidelines often state that isolated thoracic back pain (without obvious cause) in an older patient is a red flag prompting imaging, precisely due to the higher chance of bad pathology like cancer.

Note: Besides metastases and myeloma, primary tumors like spinal cord tumors (meningiomas, schwannomas) can cause thoracic pain, but those usually present with neurological signs and the pain is often radicular (due to nerve root involvement) or described as intrathecal. We won’t delve into intradural tumors. But one could consider them as part of the broad differential.

Conclusion

Chronic thoracic spine pain can originate from a variety of sources. Applying Bogduk’s rigorous postulates helps distinguish which anatomical structures are credible pain generators: the thoracic discs, facet joints, costovertebral and costotransverse joints, as well as less appreciated elements like ligaments, vertebral bodies, and ribs, all have evidence supporting their role as pain sources. Epidemiologically, facet joints may account for a large share of non-specific thoracic pain, with discs and other structures contributing smaller portions. Somatic referred pain from the cervical spine or shoulder reminds us that not all pain located in the mid-back is due to local pathology; a careful evaluation must include these possibilities.

Finally, vigilance for red flag causes – fractures, infection, malignancy – is essential, since thoracic pain disproportionate in intensity or accompanied by systemic signs may herald a serious condition. Each of these red flag entities involves a clear pathological cause that, when present, makes the affected structure (be it bone or disc) a true source of pain.

References

  1. 1.0 1.1 1.2 1.3 Manchikanti, Laxmaiah; Boswell, Mark V.; Singh, Vijay; Pampati, Vidyasagar; Damron, Kim S.; Beyer, Carla D. (2004-05-28). "Prevalence of facet joint pain in chronic spinal pain of cervical, thoracic, and lumbar regions". BMC Musculoskeletal Disorders. 5 (1): 15. doi:10.1186/1471-2474-5-15. ISSN 1471-2474.
  2. Rex Michael, Antony Louis; Newman, James; Seetharam Rao, Abhay (2010-02). "The assessment of thoracic pain". Orthopaedics and Trauma. 24 (1): 63–73. doi:10.1016/j.mporth.2009.10.005. ISSN 1877-1327. Check date values in: |date= (help)
  3. Young, Brian A; Gill, Howard E; Wainner, Robert S; Flynn, Timothy W (2008-12). "Thoracic costotransverse joint pain patterns: a study in normal volunteers". BMC Musculoskeletal Disorders (in English). 9 (1): 140. doi:10.1186/1471-2474-9-140. ISSN 1471-2474. PMC 2587465. PMID 18922181. Check date values in: |date= (help)CS1 maint: PMC format (link)
  4. Afonso, João Ribeiro; Soares, Diogo; Lopes, Daniel Brás; Matos, Rui Milheiro; Pinto, Rui Peixoto (2021 Mar 31). "Costovertebral Osteoarthrosis: Rare Differential Diagnosis of Back pain in young patients. Case Report". Revista Brasileira de Ortopedia (in English). 57 (2): 345. doi:10.1055/s-0040-1722587. PMID 35652017. Check date values in: |date= (help)
  5. 5.0 5.1 Groh, Adam M. R.; Fournier, Dale E.; Battié, Michele C.; Séguin, Cheryle A. (2021 Feb 17). "Innervation of the Human Intervertebral Disc: A Scoping Review". Pain Medicine: The Official Journal of the American Academy of Pain Medicine (in English). 22 (6): 1281. doi:10.1093/pm/pnab070. PMID 33595648. Check date values in: |date= (help)
  6. Derby, Richard; Lee, Sang-Heon; Kim, Byung-Jo; Chen, Yung; Aprill, Charles; Bogduk, Nikolai (2005-05). "Pressure-Controlled Lumbar Discography in Volunteers Without Low Back Symptoms". Pain Medicine (in English). 6 (3): 213–221. doi:10.1111/j.1526-4637.2005.05034.x. ISSN 1526-2375. Check date values in: |date= (help)
  7. Derby, Richard; Lee, Sang-Heon; Chen, Yung (2005-04). "Discograms: Cervical, thoracic, and lumbar". Techniques in Regional Anesthesia and Pain Management. 9 (2): 97–105. doi:10.1053/j.trap.2005.05.009. ISSN 1084-208X. Check date values in: |date= (help)
  8. Sarsılmaz, Ayşegül (2018-06-01). "The incidence and most common levels of thoracic degenerative disc pathologies" (PDF). Turkish Journal of Physical Medicine and Rehabilitation. 64 (2): 155–161. doi:10.5606/tftrd.2018.1302. PMC 6657757. PMID 31453506.CS1 maint: PMC format (link)
  9. Singh, Vijay; Manchikanti, Laxmaiah; Shah, Rinoo V.; Dunbar, Elmer E.; Glaser, Scott E. (2008). "Systematic review of thoracic discography as a diagnostic test for chronic spinal pain". Pain Physician. 11 (5): 631–642. ISSN 1533-3159. PMID 18850027.
  10. 10.0 10.1 Fortin, Joseph D.; McKee, May Jane (2003-10). "Thoracic facet blocks: bent needle technique". Pain Physician. 6 (4): 513–516. ISSN 1533-3159. PMID 16871307. Check date values in: |date= (help)
  11. 11.0 11.1 Dreyfuss, Paul; Tibiletti, Claire; Dreyer, Susan J. (1994-04). "Thoracic Zygapophyseal Joint Pain Patterns: A Study in Normal Volunteers". Spine (in English). 19 (7): 807–811. doi:10.1097/00007632-199404000-00014. ISSN 0362-2436. Check date values in: |date= (help)
  12. Bogduk, Nikolai; MacVicar, John (2024-05-03). "Osteoarthritis of zygapophysial joints as a cause of back pain and neck pain: a scoping review". Pain Medicine. 25 (9): 541–552. doi:10.1093/pm/pnae036. ISSN 1526-2375.
  13. 13.0 13.1 Young et al.. Thoracic costotransverse joint pain patterns: a study in normal volunteers. BMC musculoskeletal disorders 2008. 9:140. PMID: 18922181. DOI. Full Text.
  14. Patel, Janus; Eaves, Ashley M.; Deschler, Emily; Bowlin, Christian; Galang, Enrique (2024 Jul 6). "Radiofrequency Ablation of the Costotransverse Joint in a Case of Chronic Thoracic Back Pain". Cureus (in English). 16 (7): e63958. doi:10.7759/cureus.63958. PMID 39109099. Check date values in: |date= (help)
  15. Erwin, W.Mark; Jackson, Patrick C.; Homonko, Darlene A. (2000-07). "Innervation of the human costovertebral joint: Implications for clinical back pain syndromes". Journal of Manipulative and Physiological Therapeutics (in English). 23 (6): 395–403. doi:10.1067/mmt.2000.108144. Check date values in: |date= (help)
  16. Afonso, João Ribeiro; Soares, Diogo; Lopes, Daniel Brás; Matos, Rui Milheiro; Pinto, Rui Peixoto (2021 Mar 31). "Costovertebral Osteoarthrosis: Rare Differential Diagnosis of Back pain in young patients. Case Report". Revista Brasileira de Ortopedia (in English). 57 (2): 345. doi:10.1055/s-0040-1722587. PMID 35652017. Check date values in: |date= (help)
  17. Jung, Seung Min; Kim, Moon-Young; Hong, Yeon Sik; Park, Sung-Hwan; Kang, Kwi Young (2023-07). "Costovertebral joint involvement in patients with axial spondyloarthritis". Joint Bone Spine. 90 (4): 105546. doi:10.1016/j.jbspin.2023.105546. ISSN 1297-319X. Check date values in: |date= (help)
  18. Afonso, João Ribeiro; Soares, Diogo; Lopes, Daniel Brás; Matos, Rui Milheiro; Pinto, Rui Peixoto (2021 Mar 31). "Costovertebral Osteoarthrosis: Rare Differential Diagnosis of Back pain in young patients. Case Report". Revista Brasileira de Ortopedia (in English). 57 (2): 345. doi:10.1055/s-0040-1722587. PMID 35652017. Check date values in: |date= (help)
  19. Arroyo, J F; Jolliet, Ph; Junod, A F (1992-08-01). "Costovertebral joint dysfunction: another misdiagnosed cause of atypical chest pain". Postgraduate Medical Journal. 68 (802): 655–659. doi:10.1136/pgmj.68.802.655. ISSN 0032-5473.
  20. Dwyer, A.; Aprill, C.; Bogduk, N. (1990-06). "Cervical zygapophyseal joint pain patterns. I: A study in normal volunteers". Spine. 15 (6): 453–457. doi:10.1097/00007632-199006000-00004. ISSN 0362-2436. PMID 2402682. Check date values in: |date= (help)
  21. Cooper, Grant; Bailey, Beverly; Bogduk, Nikolai (2007). "Cervical zygapophysial joint pain maps". Pain Medicine (Malden, Mass.). 8 (4): 344–353. doi:10.1111/j.1526-4637.2006.00201.x. ISSN 1526-2375. PMID 17610457.
  22. Fukui, Sei; Ohseto, Kiyoshige; Shiotani, Masahiro; Ohno, Kenji; Karasawa, Hidetake; Naganuma, Yoshikazu; Yuda, Yasumasa (1996-11). "Referred pain distribution of the cervical zygapophyseal joints and cervical dorsal rami". Pain. 68 (1): 79–83. doi:10.1016/S0304-3959(96)03173-9. ISSN 0304-3959. PMID 9252002. Check date values in: |date= (help)
  23. Grubb, S. A.; Kelly, C. K. (2000-06-01). "Cervical discography: clinical implications from 12 years of experience". Spine. 25 (11): 1382–1389. doi:10.1097/00007632-200006010-00010. ISSN 0362-2436. PMID 10828920.
  24. Slipman, Curtis W.; Plastaras, Chris; Patel, Rajeev; Isaac, Zacharia; Chow, David; Garvan, Cynthia; Pauza, Kevin; Furman, Michael (2005). "Provocative cervical discography symptom mapping". The Spine Journal: Official Journal of the North American Spine Society. 5 (4): 381–388. doi:10.1016/j.spinee.2004.11.012. ISSN 1529-9430. PMID 15996607.
  25. Kennedy, David J.; Mattie, Ryan; Nguyen, Quang; Hamilton, Scott; Conrad, Bryan (2015-08). "Glenohumeral Joint Pain Referral Patterns: A Descriptive Study". Pain Medicine. 16 (8): 1603–1609. doi:10.1111/pme.12797. ISSN 1526-2375. Check date values in: |date= (help)
  26. Gerber, Christian; Galantay, Richard V; Hersche, Otmar (1998-07). "The pattern of pain produced by irritation of the acromioclavicular joint and the subacromial space". Journal of Shoulder and Elbow Surgery. 7 (4): 352–355. doi:10.1016/s1058-2746(98)90022-2. ISSN 1058-2746. Check date values in: |date= (help)
  27. Kellgren J. – Deep pain referral from ligament injection (1939)
  28. Gouliouris, T.; Aliyu, S. H.; Brown, N. M. (2010-11-01). "Spondylodiscitis: update on diagnosis and management". Journal of Antimicrobial Chemotherapy (in English). 65 (Supplement 3): iii11–iii24. doi:10.1093/jac/dkq303. ISSN 0305-7453.
  29. Liao, Jen-Chung; Chen, Wen-Jer; Chen, Lih-Hui (2022-04). "Surgery for metastatic epidural spinal cord compression in thoracic spine, anterior or posterior approach?". Biomedical Journal (in English). 45 (2): 370–376. doi:10.1016/j.bj.2021.03.004. PMC 9250068. PMID 35595649. Check date values in: |date= (help)CS1 maint: PMC format (link)
  30. Ziu, Endrit; Viswanathan, Vibhu Krishnan; Mesfin, Fassil B. (2025). "Spinal Metastasis". StatPearls. Treasure Island (FL): StatPearls Publishing. PMID 28722979.
  31. Jacobs, W. Bradley; Perrin, Richard G. (2001-12). "Evaluation and treatment of spinal metastases: an overview". Neurosurgical Focus. 11 (6): 1–11. doi:10.3171/foc.2001.11.6.11. ISSN 1092-0684. Check date values in: |date= (help)
  32. Rex Michael, Antony Louis; Newman, James; Seetharam Rao, Abhay (2010-02). "The assessment of thoracic pain". Orthopaedics and Trauma. 24 (1): 63–73. doi:10.1016/j.mporth.2009.10.005. ISSN 1877-1327. Check date values in: |date= (help)