Dorsal Thoracic Pain

in Manual Medicine

Jean-Yves Maigne, MD, Physical Medicine, Hotel-Dieu Hospital, 75181 Paris Cedex 04, France

Written 2002. This article has also been published (text only) in the Journal of Orthopaedic Medicine (last issue, 2002) with co-author Richard Ellis.

Content of the page:
Nerve entrapment
Muscular sprain
Myofascial pain
Pain referred from the lower cervical spine
Thoracic facet syndrome
Discogenic pain
Psychogenic pain

Summary: Dorsal thoracic pain is very common, but there is a paucity of literature on this topic. We present a review of the available literature on the causes of this syndrome. Cyriax attributed thoracic pain to a disc protrusion. The mechanism of pain was primarily dural. The clinical features were the reproduction of the pain under neck flexion contrasting with the freedom of other neck movements, a limitation in thoracic flexion and an increase of pain with coughing.

Experimentally, thoracic discography has partially substanciated this hypothesis. Travell and Simons describe trigger points in thoracic muscles such as longissimus dorsalis, iliocostalis, semispinalis, rhomboideus, levator scapula, etc... The causes for trigger points are many, including trauma, poor postures and repetitive movements.

Dorsal thoracic pain is very common, but there is a paucity of literature on this topic. Different causes have been evoked, such as nerve entrapment, muscular sprain, trigger points, pain referred from the lower cervical spine, thoracic facet syndrome, discogenic pain or psychogenic disease. Our aim is to review theses different sources of pain and to discuss them in light of our own studies and experience in thoracic pain.



Astvatsaturov was the first to describe, in 1934, what he called “Notalgia paresthetica” (notalgia, a Greek name for “pain in the back”). Patients typically report chronic pain and sensory symptoms that are frequently described as intense itching in an area 4 to 10 cm in diameter over the thoracic paraspinal muscles at the inferomedial scapula. There is a diminished pinprick sensitivity corresponding to an area from T2 to T6 thoracic dermatomes. The injury mechanism is essentially unknown. Notalgia paresthetica is thought to be caused by a lesion of a thoracic dorsal primary rami T2 trough T6 (entrapment neuropathy) which run a long course up through the thick paraspinal muscles. The nerves appear to be vulnerable to damage as they pursue a right angle course close to the tip of the spinous process, thus predisposing them to harm from otherwise innocuous insults of a varied nature. Investigations are not helpful in confirming the pathogenesis.

Three Author's dissections. Left: a view of the superficial exit of the cutaneous dorsal ramus of T4, through the superficial layer of the rhomboideus (the fibers of which having being moved away) and, deeper, through the tendon of the splenius cervicis, which attaches on the lateral aspect of the spinous process of T4 and on the adjacent structures. Middle: an entrapped cutaneous thoracic dorsal ramus (black arrows) after having removed the thigh orifice. Left: another cutaneous dorsal ramus with a mark of entrapment.

This condition seems to be very rare, in contrast with an anatomic study showing that entrapment is a common feature of the thoracic nerves (Maigne et al).


Muscular sprain

Muscular sprain is a very common diagnosis, but there is no evidence to support it.


Myofascial pain

Travell and Simons have developed the concept of myofascial pain and trigger points. A trigger point is a very tender and hypersensitive spot within a muscle (or in an aponeurosis, a tendon or even in the subcutaneous tissues), the pressure of which reproduces the actual pain and its radiations. At the thoracic level, these authors describe trigger points in muscles such as longissimus dorsalis, iliocostalis, semispinalis, rhomboideus, levator scapula, etc... One of these points is located in the multifidus, projecting over the vertebral body of T5 and seems to us of special interest (see below). The causes for trigger points are many, including trauma, poor postures, repetitive movements, etc.

The treatment includes local injections of an anesthetic or a cold spray over the muscles to treat, followed by stretching.

Some trigger points according to Travell and Simons. On the left, iliocostalis, pars thoracis. On the right, multifidus. Due to its same location, close to the spinous process of T4, this trigger point ressembles the interscapular point described by R. Maigne and anatomically studied by the Author.


Pain referred from the lower cervical spine

1- Interscapular pain of cervical origin

Robert Maigne described what he called “interscapular pain of cervical origin”, considered by him as a very common cause of  thoracic pain (Maigne, 1967). The syndrome is made of thoracic and cervical signs.

Thoracic signs. The pain is felt between the scapulae, and can be reproduced by a firm pressure on a tender point projecting over the body of T5, located close to the midline. This tender point was labelled “interscapular point”. It was viewed by R. Maigne as the superficial exit of the T2 nerve (in accordance to certain anatomic plates). It is accompanied by an area of cellulalgia (skin and subcutaneous tissues hypersensitivity to pinch and roll test) located on the same side and distributed in the territory of the upper thoracic cutaneous rami (mainly the T2 nerve).

Interscapular pain of cervical origin, as desribed by R. Maigne. The pain is experienced in the interscapular area and is in fact refered from the lower cervical spine. The familiar pain is reproduced by a pressure on the "interscapular point", left or right to the projection of T5. In R. Maigne 's view, this syndrome is the most common explanation for dorsal thoracic pain.

Cervical signs. At the cervical level (mainly at C5-6), a segmental dysfunction is present on the same side. When this dysfunction is not felt (or perceived) by the patient, there is no cervical pain and the only complaint is thoracic pain. But in many cases of common neck pain, there is also a radiation to the interscapular area, by the same mechanism.

In the author’s view, the dorsal pain is a referral from the cervical level. His hypothesis for explaining this link between the lower cervical levels and the interscapular point is that a segmental dysfunction at the lower cervical levels can induce hypersensitivity in the medulla, spreading to the adjacent inferior levels, as far as the T2 metamere. As there is no cutaneous dorsal ramus from C6 to T1, the closest adjacent level to C6 is the T2 dorsal dermatome. Maigne stressed that this link also appeared in case of acute cervical radiculopathy, which was often preceded by interscapular pain.

As a consequence, the treatment (mostly manual, but also by facet injection) is directed to the cervical spine, and its efficacy (if so…) is considered as the definite proof of the cervical origin of the pain.

2- The “interscapular pain of cervical origin” revisited

Ten years ago, we underwent a clinical and anatomic study on this topic. Fifty three consecutive patients with thoracic pain were examined. Among them, 40 (75.5%) had their familiar pain reproduced by the pressure on the interscapular point. It appeared that the most acute pain was elicited by an oblique pressure directed toward the lateral aspect of the tip of the spinous process of  T4 (and not a PA pressure toward the facet). A metallic marker was sticked on the skin and these patients were X-rayed. The marks projected on the body of T5 in the majority of the cases (and sometimes T6), which did correspond to the spinous process of T4 (or T5). Thus appeared a relationship between this anatomic spot and the interscapular point.

Left: palpation of the interscapular point. The spine must be flexed forward, to stretch the dorsal structures. The most teder spot is easily located by an oblique pressure (toward the midline) on the lateral aspect of a spinous process. A metallic mark shows this spinous process projecting over the body of T5, thus belonging to T4 (right).

In the same time, we dissected on 17 cadavers. It appeared obvious that the only anatomic structure corresponding to this interscapular point was the lower attachment of the muscle splenius cervicis. In the classic textbooks of anatomy, this muscle is portrayed with a lower attachment on the spinous processes of T2, T3, T4 and T5. We did not find this disposition. Rather, the muscle inserted mainly on the spinous process of T4 and on the adjacent interspinous ligaments of T3-4 and T4-5 for its upper and lower fibres. It was the only structure actually corresponding to the clinical data. Furthermore, we found that each cutaneous dorsal nerve ran from the intervertebral foramen to the spinous process, parallel to the lamina, and exited superficially close to the tip of the spinous process. In other words, the T4 nerve (and not the T2 nerve) exited at the level of the T4 spinous process. Interestingly, many of these nerves exited through tendons and appeared pinched by them. Thus, the prerequisite for an entrapment (or at least, an irritation) neuropathy were present.


Samples of our dissections (JYM). Superior view, left: superficial layer of the upper back (trapezius removed) showing the attachements of the splenius cervicis on the spinous process of T4 (and, at a lesser extent, on the adjacent interspinous ligaments). Inferior view, left: closer view of a left side (SC: splenius cervicis) with two nerves (the cutaneous dorsal rami of T3 and T4) exiting through the tendon. Superior view, right: another specimen, with dorsal rami appearing through the tendon. Inferior view, right: right side, a nerve exits through the tendon. See the anatomic section for more details.


The splenius cervicis is a thin superficial neck muscle, sharing its upper insertions with the levator scapula on the transverse processes of the upper cervical vertebrae. The two muscles belong to the same muscular layer. They run close together along the spine, dividing to attach on the scapula laterally and on the thoracic spine medially. The splenius syndrome resembles the levator scapula syndrome. The causes could be a fatigue of the muscle, related to prolonged neck postures, or to a dysfunction of the cervical spine.


Thoracic facet syndrome

A facet origin has been advocated by Dreyfuss et al. By injecting a contrast medium in the zygapophyseal joints of normal volunteers, they were able to establish that these joints were a potential source of local and referred pain under capsular distension. Evoked referral patterns were consistent in all subjects. Significant overlap occurred in the referral patterns, with most thoracic regions sharing 3-5 different joint referral zones.

R. Maigne also described pain referred from the thoracic facet joints. In his view, the pain followed the cutaneous dorsal ramus and was accompanied by a stripe of cellulalgia within the dermatome.

Reprinted from R. Maigne: distribution of pain arising from the thoracic facets. These areas also correspond to areas of cellulalgia.


Discogenic pain

Cyriax was the first to attribute thoracic pain to a slipped disc. For him, the pain was dural for a part, elicited by the pressure of the protruded disc against the thoracic dura mater. Some of the clinical features were a limitation in thoracic flexion and the reproduction of the pain under neck flexion, coughing or sneezing. He used to treat his patients by combining manipulation and traction.

Thoracic MRI often reveals degenerated discs, or disc protrusion, but despite the lack of studies addressing this topic, it can be assumed that, as at the lumbar level, these abnormalities are frequent in asymptomatic subjects.

More recently, thoracic discography has been used to investigate patients with thoracic pain. A clinical concordance (reproduction of the familiar pain) was found in approximately 50% of the cases. When the discogram was pathologic (anular tears, intrinsic degeneration, and/or associated vertebral body endplate infractions), this figure raised up to 75%. Control levels were usually painless.


Psychogenic disease

Thoracic pain has also been attributed to psychological factors. Some decades ago, for example, the French Rheumatologic School considered it as exemplary of a genuine psychological pain. The arguments were the lack of objective symptoms (what practitioners of manual medicine, trained to spinal palpation, know to be false), and the large female predominance (an insight into macho medicine!)

The description of fibromyalgia has renewed this conception. The clinical features of fibromyalgia resemble those of “psychogenic pain”. A neurologic mechanism could be involved.


  • Bernard Pleet A, Wayne Massey E. Notalgia paresthetica. Neurology 1978;28:1310-12.

  • Bruckner FE, Greco A, Leung AWL. Benign thoracic pain syndrome : role of magnetic resonance imaging in the detection and localization of thoracic disc disease. Journal of the Royal Society of Medicine 1989;82:81-3.

  • Cloward R. The clinical significance of the sinu-vertebral nerve of the cervical spine in relation to the cervical disk syndrome. J. Neurol. Neurosurg. Psychiat. 1960;23:321.

  • Dreyfuss, Tibiletti and Dreyer. Thoracic zygapophyseal joint pain patterns. A study in normal volunteers. Spine 1994;19:807-11.

  • Maigne JY, Maigne R, Guerin-Surville H. Upper thoracic dorsal rami: anatomic study of their medial cutaneous branches. Surgical Radiolology Anatomy 1991;13:109-12

  • Maigne JY. Muscle splenius du cou et dorsalgies communes. Etude anatomique et clinique. Rhumatologie

  • Maigne R. A propos du mécanisme de la douleur dans les dorsalgies. Le point inter scapulo-vertebral. Revue du Rhumatisme. 1967;34:636-41.

  • Schellhas, Pollei and Dorwart. Thoracic discography. A safe and reliable technique. Spine 1994;19:2103-9.

  • Travell JG, Simons DG: Myofascial Pain and Dysfunction: The Trigger Point Manual, Baltimore, Williams and Wilkins, 1983