G Model

EURR-7021; No. of Pages 11 ARTICLE IN PRESS

European Journal of Radiology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

European Journal of Radiology

journal homepage: www.elsevier.com/locate/ejrad

Spinal pain
R. Izzo a,∗ , T. Popolizio b , P. D’Aprile c , M. Muto d
a
Neuroradiology Department, A. Cardarelli Hospital, Naples, Italy
b
Radiology Department, Casa Sollievo della Sofferenza Hospital, San Giovanni Rotondo (Fg), Italy
c
Neuroradiology Department, San Paolo Hospital, Bari, Italy
d
Neuroradiology Department, A. Cardarelli Hospital, Napoli, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The spinal pain, and expecially the low back pain (LBP), represents the second cause for a medical consul-
Received 26 July 2014 tation in primary care setting and a leading cause of disability worldwide [1]. LBP is more often idiopathic.
Received in revised form 20 January 2015 It has as most frequent cause the internal disc disruption (IDD) and is referred to as discogenic pain. IDD
Accepted 22 January 2015
refers to annular fissures, disc collapse and mechanical failure, with no significant modification of exter-
nal disc shape, with or without endplates changes. IDD is described as a separate clinical entity in respect
Keywords:
to disc herniation, segmental instability and degenerative disc desease (DDD). The radicular pain has as
Spinal pain
most frequent causes a disc herniation and a canal stenosis. Both discogenic and radicular pain also have
Spine degeneration
CT
either a mechanical and an inflammatory genesis. For to be richly innervated, facet joints can be a direct
MR source of pain, while for their degenerative changes cause compression of nerve roots in lateral recesses
and in the neural foramina. Degenerative instability is a common and often misdiagnosed cause of axial
and radicular pain, being also a frequent indication for surgery. Acute pain tends to extinguish along with
its cause, but the setting of complex processes of peripheral and central sensitization may influence its
evolution in chronic pain, much more difficult to treat.
The clinical assessment of pain source can be a challenge because of the complex anatomy and function
of the spine; the advanced imaging methods are often not sufficient for a definitive diagnosis because
similar findings could be present in either asymptomatic and symptomatic subjects: a clinical correlation
is always mandatory and the therapy cannot rely uniquely upon any imaging abnormalities. Purpose of
this review is to address the current concepts on the pathophysiology of discogenic, radicular, facet and
dysfunctional pain, focusing on the role of the imaging in the diagnostic setting, to potentially address a
correct approach also to minimally invasive interventional techniques. Special attention will be done to
the discogenic pain, actually considered as the most frequent cause of chronic low back pain.
© 2015 Published by Elsevier Ireland Ltd.

1. Introduction incidence is disproportionally raising with respect to the popula-

tion growth.
The low back pain (LBP) represents the second leading cause Diagnostic assessment of the spinal pain can be a challenge espe-
of disability worldwide with an estimated 1% of the world popu- cially in the case of discogenic and dysfunctional pain, due to the
lation involved [1] and represents a major welfare and economic complex anatomy and function of the spine. Actually, the disco-
problem in western countries [2]. The traditional notion that acute genic pain is considered a main cause of chronic LBP and disability,
nonspecific back pain is usually a benign, transient and self-limiting as a consequence of internal disc disruption (IDD).
condition has been recently reconsidered with reports of 1-year MR imaging has a limited specificity in the assessment of a
recurrence rate of 20–44% and a lifetime recurrence of up to 72% painful spine, being able to identify any disc and endplate changes
[3]. The actual, estimated lifetime prevalence rates of LBP and with limited diagnostic value in differentiating between patholog-
neck pain in adults are 91% and 66.7% respectively [1,4], but the ical painful abnormalities and simple ageing modifications.
Complex processes of peripheral and central sensitization may
influence the evolution of acute to chronic pain: pain persistence
∗ Corresponding author. Tel.: +39 0817473836. should lead to modify any therapeutic strategy and treatment if
E-mail addresses: [email protected] (R. Izzo), [email protected] focusing only on pathological changes involving the peripheral ner-
(T. Popolizio), [email protected] (P. D’Aprile), [email protected] (M. Muto). vous system. In case of failure of conservative treatment, a more

http://dx.doi.org/10.1016/j.ejrad.2015.01.018
0720-048X/© 2015 Published by Elsevier Ireland Ltd.

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
2 R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx

aggressive strategy has to be employed using the new minimally

invasive techniques before surgery.

2. Pathways of spinal pain

The sensitive innervation of the spine includes both somatic

fibres present inside the spinal nerves and autonomic fibres origi-
nating from the paravertebral sympathetic ganglia and chains [5].
Discs and vertebral bodies are innervated all around their cir-
cumference by three microscopic plexuses (anterior, lateral, and
posterior).
All along the plexuses, there are short periosteal and annu-
lus fibres as well as long penetrating branches, which transverse
the vertebral bodies and supply the endosteal surface of cartilagi-
nous endplates, completing the disc innervation. The sinuvertebral
nerves are the main contributors of the posterior plexus. The sin-
uvertebral (Luschka) nerve is a recurrent branch formed by union
of the grey ramus communicans (GRC) with a little branch com-
ing from the proximal end of anterior primary ramus of the spinal
nerve [5]. It is a polisegmentary mixed nerve which immediately
reenters the spinal canal and sends ascending and descending anas- Fig. 1. The convergence of peripheral sensory inputs upon 2nd and 3rd order
tomosing branches containing both somatic and autonomic fibres neurons is the physiological basis of referred pain. In the absence of any further
for the posterolateral annulus, the posterior vertebral body and the information the CNS can be unable to establish the exact source of a stimulus.
periostium, ventral meninges, anterior epidural vessels [5].
Sympathetic trunks and ganglia directly innervate the anterior The ascending diversion of impulses also contributes to a not
longitudinal ligament (ALL), the anterior periostium and vertebral defined localization of sympathetic painful inputs. For the anatom-
body, the paravertebral muscles and fascias as well as the antero- ical absence of WRC below L2 level, any autonomic sensory input
lateral disc [6]. has to ascend into sympathetic chains at least up L2 level before
Autonomic nervous afferences may enter the white ramus com- reaching the spinal cord. For this reason, the areas of referred pain
municans (WRC) and reach the spinal nerve and the dorsal root from all levels also tend to concentrate within the somatomes of
ganglion, or they can reach the sinuvertebral nerve through the the upper lumbar nerves [7,8]. The local anaesthetic blocks of L2
grey ramus communicans. Through the sinuvertebral nerves, auto- nerve root can relieve a discogenic LBP.
nomic fibres reach all the structure they supply and complete the
innervation of the anterior spine [6]. 3. Pathophysiology of spinal pain
In this way autonomic fibres directly or indirectly supply all the
anterior spine. Painful sensations start in peripheral receptors which can be
The facets joints also receive a double innervation, either either specific, being only responsive to noxious inputs, or poly-
somatic and autonomic. modal, activated from several types of stimuli, and they are
The zygapophysial joints are innervated by the media and inter- transferred along a nociresponsive neuron which can also be spe-
mediate branches of primary dorsal branches of the spinal nerves. cific orwide dynamic range (WDR) type, functioning as noxious or
Each medial branch supplies the apophysial joints at its own level no-noxious transmitter. In the case of an acute pain, the signals
and at the level below. Each joint receives in this way a double sen- reaching the dorsal horns of spinal cord retain the same quality
sory innervation. Somatic fibres directly reaching the spinal cord at and intensity of those travelling in spinothalamic tracts and can be
every level mediate a well-localized local pain, whereas the auto- efficaciously controlled and faded by supraspinal centres [9].
nomic afferences are responsible of referred pain. In case of persistent noxious stimulation, processes of peripheral
Referred pain is perceived in areas innervated by different and central sensitization can be settled facilitating the conversion
nerves from those that supply the site of real transduction: the from acute to chronic pain. Sensitization may either consist in a
nerves supplying the region of referred pain are not engaged nor decrease of neuronal activation threshold, as well as in increased
do they convey any input. Both radicular and referred pains are response to over-threshold stimuli and even in spontaneous neu-
false representations of sensory events by the central nervous sys- ronal activity. While the acute pain tends to extinguish along
tem, but while radicular pain is well-discriminated and distributes with the underlying cause, a persistent stimulation of noxious
within an expected dermatome, referred pain is diffuse and ill local- peripheral nerve endings can lead the neural centres to a hyper-
ized into deep tissues of a somatome. A somatome includes all excitability state where specific neurons activate with increased
tissues having the same embryological origin that also share com- frequency and even spontaneously, while WDR-neurons, normally
mon neural circuits and can also share pathways of sensory referral non-transmitting pain, could be involved in a noxious activ-
[7]. ity creating the basis of allodynia and hyperalgesia phenomena
From a physiological point of view, while referred pain is trig- (“wind-up” response) (Fig. 2) [9].
gered by stimulation of the nociceptive terminals, the radicular pain The wind-up pain is based on an altered and unbalanced action
is evoked into an inflamed dorsal root or ganglion [8]. The basis of either inhibitory and excitatory neurotransmitters. The persis-
of referred pain is considered to be the convergence of neurons tent stimulation of primary-order neurons promotes an excessive
subtending different peripheral sites onto common 2nd/3rd-order release of P-substance (SP) and glutamate into synaptic spaces
neurons within the dorsal horns of spinal cord and in the thalamus of the dorsal horns of the spinal cord. The interaction of the
sending inputs to higher centres. In the absence of further informa- glutamate with N-methyl-d-aspartate (NMDA) receptors of post-
tion, the brain can be incapable of distinguishing the exact source synaptic membranes promotes the cellular influx of Ca2+ ions with
of a sensory input arriving onto the common neuron (Fig. 1). activation of the nitric oxide synthase (NOS). Nitric oxide stimulates

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx 3

To avoid the chronic evolution of an acute pain with related

problems for treatment, it is mandatory to consider both the symp-
toms and the underlying causes, in order to start a therapy as soon
as possible; in case of failure of the symptoms improvement, a
more aggressive treatment could be planned [9], such as the actual
minimally invasive techniques.
Owing to CNS influences, the chronic spinal pain is now consid-
ered as a bio-psycho-social syndrome, a final stage of a process that
from the acute condition due to a biological or physical injury could
be transformed by psychosocial risk factors in a chronic pathology
[13].
Psychosocial and socioeconomic problems, including job dissat-
isfaction, illness behaviour, disability benefits, history of back pain
and previous treatments represent conditions that are all involved
in the maintenance of pain itself, so they are actually considered to
be more significant predictors of the pain and disability than other
biological factors.

4. Dysfunction instability and pain

The CNS influences on the back pain progression from acuteness

to chronicity promoted several researches on the spinal dysfunc-
tions and on the motor control deficits in LBP patients. While the
onset of pain has some biological causes, its persistence seems to
be due to several psychosocial and physical factors (dysfunction,
Fig. 2. The wind-up phenomenon is one of the causes of chronic pain. It has a deconditioning) [13].
biochemical and electrophysiological basis resting on the action of excitatory and
The spinal motor control deficits in patients affected by chronic
inhibitory neurotransmitters. In case of persistent depolarization of post-synaptic
membranes the activation of NMDA-receptors to glutamate primes cellular inflow of LBP can consist in impaired motion, spinal repositioning errors,
Ca2+ ions finally leading to synthesis of new membrane receptors as well as further lack of peripheral feedbacks, postural imbalance, generation of
release of glutamate and substance P through a self-maintaining vicious circle. increased loads, early muscle fatigue, and the absence of muscle
relaxation at full lumbar flexion. In addition, these patients can
have a low reaction time consisting in a delayed or absent muscle
the further presynaptic release of SP and glutamate by diffusing response [14].
across the synaptic spaces, with creation of a self-maintaining circle Patients, in order to avoid pain, can assume a fear/avoidance
[10]. behaviour by which they tend to protect themselves by restricting
At same time, activation of the NMDA receptors induces a pro- more and more the engagement of the spine and finally reduce
duction of new membrane receptors. Also in case of chronic painful the spinal motion, coordination, and strength [15]. Patients with
stimulation supraspinal centres send descending signals trying to whiplash-associated disorders could also have an altered neck
modulate any persistent incoming input, but they are rendered motion.
ineffective by wind-up and by closed and reverberating neuronal Finally, there are evidences that patients having psychological
circuits. Once wind-up is set, it is maintained by a series of posi- distress with no previous history of pain tend to develop an abnor-
tive feedback which persist with minimal or even with no further mal spinal motion control and loading that can elicit a progressive
peripheral inputs [9]. Neuroplastic mutations and wind-up can chronic pain in absence of any direct injury to the spine.
occur at both spinal and supraspinal levels, in 2nd-, 3rd-, and 4th- The degenerative instability is a common cause of axial and
order neurons. In addition, during chronic painful states it can radicular pain and disability. Stability is the first condition required
develop neuroimmune and neuroinflammatory changes which also for a correct function of the spine: it is preserved by a strict correla-
contribute to perpetuation and amplification of pain. The intracel- tion of three subsystems including the spine (passive subsystem),
lular calcium inflow, prompted by NMDA receptors activation, also the muscles (active subsystem) and the CNS (acting as a central
promotes the breakdown of prostanoids to arachidonic acid result- control unit).
ing in activation of central nervous system (CNS) inflammation The spinal components have either a structural and transducer
[11]. function, the latter one due to mechanical receptors located within
The activated glial cells and neurons produce some pro- discs, ligaments and joint capsules which convey to CNS proprio-
inflammatory cytokines (IL-1, IL-6, TNF) which expand the ceptive inputs on the spatial position, the loading status and the
activation of the neurons themselves and promote leucocyte movement of each motion segment [16].
migration and infiltration from the periphery into the CNS, with A damage to any spinal structures (especially the ligaments)
further induction of inflammation mediators (such as glutamate, and receptors can generate some altered inputs for the CNS that
prostaglandins) [12]. can promote an inappropriate muscle action and elicit an abnormal
Very similar pathological reactions and nervous centres acti- peripheral feedback response. Over time, a vicious circle can lead to
vation can also occur in some chronic affective disorders, such as muscle and joint stress, chronic dysfunction and pain. A mismatch
depression, anxiety, panic crises and post-traumatic stress. While between normal and altered peripheral signals can cause reaction
the acute pain tends to solve over time, resolution of a chronic LBP of the CNS with an impaired balance and postural control, as well as
can occur in less than 5–10% of cases. Once electrophysiological re-positioning errors often observed in chronic LBP patients [17].
biochemical and inflammatory changes have been originated, there In a MS with an injury of a restraint controlling a given phase
are little chances for therapy-responses or to spontaneous fading, of motion, the sudden and unexpected change in motion occur too
making a chronic pain much more difficult to be managed [9]. fast to be corrected through a proprioceptive reflex, so the CNS

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
4 R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx

Fig. 3. 57-Year-old man with long history of axial lumbar pain radiating to buttocks. (a) Sagittal midline FSE T2 image showing a apparently normal alignment of lumbar
vertebrae. (b) FSE axial image at L5-S1 depicting degenerative remodelling of facets and bilateral joint effusion, prevalent on the right, suggesting eventual abnormal joint
excursion. (c) Flexion-dynamic radiograph confirms an occult intermittent degenerative spondylolisthesis appearing only during flexion movement.

replies with an abnormal muscle contraction creating a jerk or catch Several imaging findings can be found in the spine instability,
sensation. The segmental instability may consist in a pure motion including: endplates, peduncles and isthmic oedema [19], traction
syndrome: some movements can generate several symptoms in the spurs, extended discal vacuum, facets gapping with joint effusion
absence of any significant osseous lesions (microinstability), due or vacuum [20], synovial cysts, annular tears, spondylolysthesis,
to an impaired restraint and muscle control. The microinstability retrolysthesis.
syndromes are defined on the basis of the mechanism of injury, Traction spurs are consequence of some increased tensile
the location of tissue damage, the aggravating activities and the stresses exerted by the ALL and Sharpey’s fibres provoked by
impaired movements. increased abnormal movements. Patients with a degenerative
In a degenerated spine instability can evolve in three main spondylolisthesis present significantly larger facet joint effusions,
phases, the so-called “degenerative cascade”: dysfunction, instabil- especially in case of a mobile, intermittent, low-grade anterolisthe-
ity, restabilization [18]. During the dysfunction phase an intermittent sis [20].
nonspecific low back pain can appear at the initial changes in So, even in the presence of a normal vertebral alignment on
the discs and facet joints. In the instability phase the facet car- recumbent MRI, any effusion over 1 mm could be considered as
tilage degeneration, the disc space narrowing and the ligaments an indication for dynamic radiographs or MR in order to diagnose
impairment can lead to some abnormal vertebral movement and an intermittent occult spondylolisthesis (Fig. 3) [20]. The acquired
alignment, up to anterolisthesis or retrolisthesis: at this stage the isthmic spondylolisthesis can also occur in a degenerative spine,
pain can become more persistent. The movement disorders usu- due to the opposite colliding action of the vertically slipped facets
ally starting from an interbody joint, first extend to joints at the upon the isthmus until a stress fracture [19].
same motion segment (three-articular complex), and then extend Type-I Modic changes can be associated to painful instability,
to adjacent segments transforming a segmental dysfuncion in a tend to convert into more stable types II and III changes after fusion,
regional pathology [19]. but may persist or be recurrent in case of pseudarthrosis. Patients

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx 5

Fig. 4. FSE T2-weighted axial image at L5-S1 showing large grade 4 facet osteo-
phytes associated to bone sclerosis and joint spaces collapse. Wrap around bumper
osteophytes develop along the insertions of facet capsule to enlarge the joint sur-
faces and limit abnormal movements.

often complain chronic or recurrent pain in the low back or radiat-

ing to the lower extremities, relieved by lying in supine position.
Degenerative spondylolysthesis may generate through different
mechanisms mechanical pain, radicular pain or neurogenic clau-
dicatio. In the final phase of restabilization fibrosis of the joint
capsules, formation of claw osteophytes, marked discal collapse Fig. 5. Sagittal midline reformatted CT image showing a L4 isthmic grade 2 spondy-
and expansive remodelling of vertebral bodies can lead to an overall lolisthesis in extension engendering neoarthrosis between spinous processes of L3
reduction of mobility and to increased stiffness [18]. and L4 with geodes, bone sclerosis and cortical erosion. Spinous processes con-
tact creates a new support column opposing the further translation of the slipping
The restabilization phase is generally associated with signifi-
vertebra.
cant disc collapse, radial remodelling of vertebral bodies [19], claw
osteophytes, “wrap-around bumper” osteophytes around the facets
[21] (Fig. 4), facet and endplate sclerosis and arthrosis between the instability; according to Mulholland, instability is often an overem-
spinous processes (Fig. 5) [22]. phasized cause of pain, a kind of “myth” [26].
Disc collapse, osteophytes and interspinous contacts may In many cases the spinal pain is not due to dysfunctional move-
block the progression of vertebral slippage, often with secondary ment, but to abnormal irregular distribution of loads between joint
improvement of pain. The stabilizing effect of the disc space col- surfaces; as in other load-bearing joints, an altered distribution of
lapse, in particular, is such that surgery of spondylolisthesis is loads can generate pain in nociceptive endings [26]. Out of the poor
recommended only when the preoperative disc height is greater correlation between clinics and imaging, the pain may persist even
than 2 mm [23]. after technically successful fixations or may unexpectedly solve in
Conventional static imaging cannot be reliable in assessing cases of pseudarthrosis. While most rigid fixation did not improve
spinal instability. the pain control, flexible stabilization techniques are commonly
Open MR systems allow positional-dynamic studies in either successful [26].
standing or seated positions to detect increased and abnormal
intersegmental movements [24]. 5. Discogenic pain
Axial-loaded CT (AL-CT) and MR (AL-MR) examinations can sim-
ulate the weight-bearing upright position but postural changes Only in a limited percentage of cases back pain is due to a disc
related to muscle tone and physiological loads that increase in the herniation.
caudal direction cannot be detected [22]. Actually it is estimated that only fewer than 30%, and even just
Despite the advantages of modern technology studies, dynamic as few as 1% of cases of LBP are due to nerve root compression [27].
radiographs remain the most commonly used reference before The most frequent cause of nonspecific lumbar axial spinal pain
surgery because of the feasibility and lower cost; however with is believed to be the discogenic pain due to internal disruption of
this imaging method any movements not occurring in the plane of the disc (IDD), with eventual endplate changes [28].
the study cannot be assessed; conventional radiographs have also Internal disc disruption consists in annular tears, degradation
poor accuracy and sensitivity, lack of any standardization in exe- of nucleus matrix, disc collapse with eventual endplates failure, in
cution techniques and measurement methods, and show a large a the absence of significant modifications of the disc external contour
large overlap of motion patterns between normal and symptomatic and with no compression of nervous elements.
individuals [25]. Crock reported first this condition [29] that actually is known
Anyway, any abnormal movements detectable in a degenerated as a distinct clinical entity with respect to other painful processes
spine not necessarily cause pain. The spinal pain may be elicited such as segmental instability, disc prolapse and degenerative disc
from a MS because it is damaged, but it is independent from the disease (DDD).

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
6 R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx

Fig. 6. A normal disc receives a double somatic and sympathetic innervation with
proprioceptive and nociceptive terminals limited to outermost 2–3 mm where ten- Fig. 8. The histological hallmark of painful disc is the radial tear containing inflam-
sile stresses overcome compressive loads. Both nerves and capillaries could not matory vascularized and densely innervated granulation tissue. From a radial tear it
support the high pressures within a well pressurized young disc. can start an inflammatory process which can involve all the disc matrix. In addition,
through a full thickness radial tear, inflammatory mediators expressed by the disc
can directly reach the nerve roots and the dural sac or the normal nociceptors of the
external annulus eliciting pain.
Within healthy discs free and complex nerve endings are nor-
mally limited to outermost few millimetres or 2–3 most external
lamellae of the annulus fibrosus and they are nociceptive and, to a Freemont [30] compared specimens of degenerated lumbar
lesser extent, proprioceptive (Fig. 6). discs painful at provocative discography (PD) and normal control
While nerve endings and their capillaries could not withstand discs using immunohistochemical techniques, founding a strong
high hydrostatic pressures present within inner normal discs, they association between presence and extension of nociceptive endings
can penetrate into a degenerated disc for the internal depressuriza- and positive responses at disc stimulation. The neoinnervation cre-
tion [30]. Inside a very degenerated disc nervous fibres may even ates throughout the disc a close association between sympathetic
reach the nucleus (Fig. 7) [30]. Neural ingrowth into the disc is an postganglionic efferent and autonomic afferent endings, with the
important factor contributing to discogenic pain. first exterting a neuroregolatory function, with a similar pattern
Neoinnervation is significantly greater in painful discs than in to that of enteric organs suggesting that discogenic pain is a kind
degenerated asymptomatic discs. a visceral pain, unique example in the muscular–skeletal system
[31].
Main condition to successfully manage a back pain is to make an
accurate pathological diagnosis: the composition and structure of
painful disc differ from those of non-painful degenerated disc [32].
IDD cannot be considered as a disc degeneration, being an indepen-
dent condition starting as a focal disorder of the annulus fibrosus
that appears to be normal in the remaining segments at least at the
beginning of the process [33]. The histological hallmarks of painful
disc are the annulus radial tears. (Fig. 8) [32]. Radial fissures occur
at an earlier age than degenerative changes and their prevalence
does not increase over time. Within and along the radial fissures,
starting from the external annulus, it develops a densely vascular-
ized and innervated granulation tissue as a reaction and tentative
of repair and healing after an injury [32].
Inflammatory cells produce a series of growth factors (GF) that
regulate all the process of tears repair and healing. The mass expres-
sion of GFs such as basic fibroblast growth factor (b-FGF), nerve
growth factor (NGF) and connective tissue growth factor (CTGF) is
the other marker –biological– of painful disc [32]. GFs contribute
to expand the focal damage and inflammation of annulus leading
to a progressive inflammatory degradation of all disc matrix. IDD is
often preceded and flanked by endplate derangements.
Vertebral endplates are subject to fatigue failures under
repeated loads during normal daily work activities. Focal failure can
Fig. 7. In a very degenerated disc, favoured by the fall of internal pressure, nervous
occur at loads much less than ultimate endplate strength. The end-
fibres can penetrate until the nucleus Neural disc ingrowth is an important factor
contibuting to discogenic pain. Neoinnervation is significantly greater in painful plate fracture can initiate the biophysical and biochemical changes
discs than in degenerated but asymptomatic discs. of IDD.

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx 7

Discogenic axial pain is mechanical, elicited from the pos-

itions and activities which raise intradiscal pressure and stress
the annulus, while relieves during recumbency. The frequent mis-
registration of discogenic pain to sites distant from real origin
contributes to rendering the precise localization of the painful
source clinically difficult.
There is no way to definitely clinically diagnose IDD: the diag-
nosis can be based on reproduction of typical patient’s pain by
discography, the detection of an annular tear by postdiscogra-
phy CT or by MR, and/or Modic lesions in vertebral endplates
by MR.
The role of discography in the evaluation of discogenic pain
remains controversial: proponents consider it to be the most sen-
sitive imaging tool for the detection of disc abnormalities, whereas
opponents argue that discography it not specific because LBP can
be elicited also in asymptomatic subjects and morphological abnor-
Fig. 9. Small endplate failure fractures can occur under repeated loads during daily malities are not always correlated with painful responses.
normal works. The endplate fracture can initiate the biophysical and biochemical According to discography, Peng [32] classified the discogenic
changes of IDD. It primes a sudden depressurization of nucleus reproducing the LBP into two types: annular disruption- and endplate disruption-
same stress profile as IDD, but it can also engender the inflammatory degradation
induced LBP. For both ones the disruptive changes were classified
of disc matrix.
into four grades: Grade I, II, and III radial fissures reach the inner,
It primes a sudden depressurization of nucleus reproducing the middle, and outer third of the annulus, respectively. Grade IV fis-
same stress profile as IDD, but it can also engender the inflamma- sures include the circumferential spread of contrast medium [34].
tory degradation of disc matrix. Several hypothesized mechanisms Some 70% of grade III fissures are associated with pain and 70% of
have been proposed as causing this process, including an autoim- all painful discs show at least a grade 3 fissure. Discs with no fissures
mune response elicited by the exposition of antigenic nuclear or having only grade I or II fissures, are uncommonly symptomatic
content to the circulation of vertebral spongiosa [33], or simply a [35].
metabolic change of nucleus which engenders an unbalanced activ- Across clinical studies the prevalence of IDD in patients with
ity of disc proteinases, in the absence of any inflammatory process chronic back pain was 26–42% [36].
(Fig. 9). Whatever the cause of IDD could be, the disc becomes and The endplates have similar density of innervations to annu-
remains painful for mechanical and chemical sensitization mecha- lus and are another important source of lumbar discogenic pain.
nisms. Endplate lesions are also graded from 0 (no disruption) up to 4
The discs affected by IDD show altered stress profiles with a fall (extensive dispersion in the cancellous bone of contrast medium
of internal pressure, irregular distribution of the inner stresses and [32].
concentration of compressive loads upon the posterior annulus [2]. Peng found all discs with concordant pain responses having end-
The concentration and the irregular stress distribution may plate lesions more severe than grade III; endplate-related LBP is
generate pain, overcoming the threshold of transduction of noci- considered to account for 16.7% of chronic discogenic LBP [32].
ceptors, even under physiological stresses. Either an overcharged On MR imaging the annulus radial fissures correspond to high
posterior annulus or a few residual lamellae around an incomplete intensity zones (HIZ) and endplate fractures are reflected by Modic
radial tear can become a source of mechanical pain [33], having to changes.
bear a much greater and concentrated stress. However more than HIZs are bright spots inside posterior annulus whose hyper-
from mechanical factors, a pain sensitization mainly depends from intensity on T2-weighted MR images reflects the presence of
inflammatory mechanisms. inflammatory tissue. They have to be distinguished by the low
In response to mechanical oveloads or nutritional problems, intensity zones corresponding to asymptomatic, inactive disc fis-
chondocytes and fibroblasts activate and upregulate mem- sures. The clinical significance of HIZs has been widely debated.
brane metalloproteinases (MMPs), and promote infiltration of Despite a frequency of 24% in asymptomatic subjects, Saifuddin
macrophages which produce multiple pro-inflammatory media- considered HIZs specific (96%) and predictive (likelihood ratio 6.8)
tors: bradykinin, IL-I, NO, TNF, phospolipase A2, GFs. of a concordant pain response during discography [37].
Inflammatory mediators activate and sensitize the nocicep- Similarly, Schell has found 87 out 100 HIZ-discs on MR having
tors (bradykinin, serotonin, excitatory aminoacids), recruiting new concordant response at provocative discography, whereas all of 67
receptors with enlargement of the painful perception area. no HIZ-discs giving negative or non-concordant responses [38].
The inflammation and the disaggregation of a nuclear matrix However, no matter how much specific, the low sensitivity of
lead to a progressive dehydration and depressurization of the just 26% limits the relevance of the HIZ in selecting patients for
nucleus, overcharge and buckling of the annulus, with a final surgery [37].
mechanical disc impairment. The presence of an HIZ strongly increases the chances that the
A full-thickness radial tear also directly exposes nociceptors of involved disc is the real source of pain: with a prevalence of IDD of
external annulus to chemicals agents and antigens coming from 46% and a likelihood ratio of an HIZ of 3.8 there is a confidence of 73%
nucleus [33]. of the affected disc will be painful at stimulation [36]. The clinical
Chemical and mechanical sensitization can coexist. The inflam- relevance of Modic changes has also been long debated. Modic type
mation creates a background of constant, dull inflammatory pain 1 and type 2 changes occur with significantly greater frequency in
that is aggravated by mechanical stimulation of the annulus during patients with chronic pain than in asymptomatic subjects.
movements. Inflammatory processes amplify the response of nerve Type 1 and type 2 Modic changes correlated with positive
endings to mechanical stress. response during provocative discography with an average likeli-
Patients complaining discogenic pain experience unilateral or hood ratio of 3.4 (95% confidence limits of 2.8 and 4.1) in several
bilateral axial pain with frequent referral to lower limbs, but history studies, one of which based on a retrospective analysis of 2457
of symptoms often reveals not specific findings. patients [39]. However, other studies did not find any significant

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
8 R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx

correlation. According to Weishaupt, moderate to severe endplate 6. Radicular pain

Modic-type 1 and type-2 Modic changes predict a concordant
response at disc stimulation in even 100% of cases [40]. Radicular pain is generated by ectopic discharges within an irri-
Most of studies confirm the high specificity of Modic changes tated spinal nerve or in its dorsal root or ganglion [33].
which thus are unlikely to be false positive. With a likelihood ratio Radicular pain is a pain radiating from the spine into the
of 3.4 an investigator has a confidence of 69% that the affected disc dermatome of a nerve root and can be associated or not with radicu-
will reveal as painful [41]. lopathy. Radiculopathy consists in a block of conduction in sensory
Current evidence considers MRI as a valuable diagnostic tool for axons, causing numbness, or in motor axons, causing weakness: it
assessing disc morphology that avoids the expense and invasive- does not provoke pain.
ness of discography; it can be considered as the study of choice for Radicular pain has as far most frequent causes the disc herni-
the evaluation of patients with LBP. ation and canal stenosis. The physiopathology of radicular pain
Both MR and discography show high sensitivity in depicting remains incompletely understood. In animal studies mechani-
morphological abnormalities. cal traction or compression of normal nerve roots elicits only a
Controversial issues still remain as to whether discography can momentary discharge. A sustained painful response can be evoked
add further diagnostic information to those offered by MR. only by stressing an inflamed or previously damaged dorsal root
Several studies have compared the sensitivity, specificity and [48]. Mechanical compression alone becomes painful only when it
predictive value of MR imaging and discography in detecting disc involves a dorsal root ganglion.
abnormalities and reported high concordance rates between them. In both cases, discharges occur in A␦, C fibres, and, in a het-
Bernard reported a PPV of an abnormal MR for a morphologically erospecific way, A␤ fibres which render radicular pain not a purely
abnormal discography of 92% and a NPV of a normal MR for a normal nociceptive experience [48].
discography of 88% [42]. Disc herniation is the single most common cause of radicular
Even though discography can seldom depict abnormalities in pain. Even though the compression of lumbar nerve roots provokes
discs appearing as normal with MR, the clinical relevance of these oedema and raised intraneural pressure, inflammatory mecha-
findings has been not clarified. nisms prevail and explicate the frequent discrepancy between disc
Several studies have found a frequent discrepancy between herniation largeness and symptoms. Inflammation can be either
morphological abnormalities and pain response at stimulation dur- based on an autoimmune cell response elicited when the disc, being
ing discography with only 37% of patients with disc abnormalities no more segregated by annulus, is recognized as non-self by the
experiencing a painful stimulation [43] and 17–37% of asymp- immunological system [49], and/or a biochemical cascade of medi-
tomatic subjects having abnormal discograms [44]. ators expressed by disc itself directly, such as phospholipase A2,
In comparison to conventional discography, post-discography PGE2, IL-6, MMPs [50].
CT offers a much more detailed demonstration of the internal Disc material harvested from patients with disc herniation
anatomy of the disc and the changes of its substructures through and radiculopathy contains high levels of phospholipase A2 [50].
the axial depiction of the radial tears. Phospholipase A2 has either direct neurotoxic effects and is
Two studies addressed the ability of MR and discography in pre- involved in the production of arachidonic acid, leukotrienes and
dicting the outcome of surgical lumbar fusion: Colhoun reported prostaglandins. Not only the disc components prime the produc-
an 89% of good result by fusion in patients with either anatomical tion of IgM and IgG antibodies to the glycolsphingolipid of the
changes and positive disc stimulation, but only 52% of success in nerve roots, but also the exposition of nerve roots to nuclear mate-
those having abnormal disc morphology alone [45]. Gill and Blu- rial induces increased vascular permeability, endoneural oedema
menthal reported higher success rates (75%) in patients having and spontaneous firing. Inflammation finally leads to degenera-
both concordant pain on discography and anatomical changes at tion of axons, decrease in conduction velocity, up to conduction
MR than in those showing discordant results for having normal MR block [51].
and abnormal discograms (50%) [46]. Herniated disc attracts macrophages, lymphocytes and fibro-
Both anatomical and functional concordant results are needed blasts which produce a variety of chemical mediators including
for a discographic indication to surgical intervention. By virtue of phospholipase A2, prostaglandins, nitric oxide, interleukins, TNF-␣.
the high NPV (94–100%) of a normal MR study for a positive disc Inflammatory cells and mediators are in general more abundant
stimulation, discography is not indicated in case of negative MR in noncontained herniations than in contained herniations and in
findings. degenerative discs while they are lacking in normal discs.
Discography remains a presurgical invasive technique reserved Mechanical compressive and inflammatory effects can be
to patients with persistent presumed discogenic pain, unresponsive strictly related.
to conservative management and for whom a percutaneous or open In an animal model of lumbar radiculopathy, a positive correla-
treatment has been planned. This method could be considered an tion was found between pain behaviour, mRNA levels to produce
extension of the clinical examination and an useful additional tool cytokines in the spinal cord, and the magnitude of root compression
in case of equivocal MR results. [52] and, concordantly, in another study was the graded microglial
There is no alternative or superior means of stating if a partic- activation occurring in the spinal cord in response to various defor-
ular disc is really the source of patient’s symptoms, but PD is not mation degrees of lumbar nerve roots [53].
a screening procedure, although if it could be considered as a con- Inflammation can abut on nerve root or ganglion ischaemia.
firmatory test, reliable in guiding surgery and to predicting surgical Like in others spinal pathologies, also in case of disc hernia-
outcomes with specificity of 80–100% [47]. tion the clinical relevance of the abnormalities detected by MR has
Advanced noninvasive imaging of the spine in patients with pre- been questioned by several epidemiological studies showing the
sumed discogenic pain often does not provide results valuable for frequency of the same findings in asymptomatic individuals.
surgical decision making. Jensen et al. in MR studies of lumbar spine of 98 asymptomatic
Generally, imaging studies may only play a complementary role subjects noted a disc bulge at least at one level in 52% of individuals,
to the clinical evaluation: a clinical correlation is always required to a disc protrusion in 27%, a disc extrusion in 1%, while only 36% had
determine the real significance of any abnormal findings observed normal discs at all levels [54].
on CT or MRI. Therapy cannot be relied uniquely upon any imaging Despite the finding by imaging of an extruded or migrated disc
abnormalities in the absence of a clinical concordance. herniation with root compression in a patient complaining an acute

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx 9

radiculitis may be a plausible cause of pain, due to the prevalence Z-joints are the second most important spinal pain generators
of disc herniations in 20–28% of asymptomatic subjects, a detected and in case of negative imaging for disc pathology, they have to be
herniation may not be the real cause of pain nor can predict its taken into account.
appearance with time. In studies using diagnostic controlled blocks the origin of pain
On the other hand, a purely chemical radiculopathy may be trig- from the facets has had a prevalence of 15–52% among subjects
gered by an annulus incompetent which, despite a normal external complaining chronic lumbar pain [57].
contour, spills out inflammatory mediators into the epidural Owing to the rich innervation of capsules, synovial membranes
space. and subchondral bone, facet joints can be an important direct
As stressed by Bogduk, discrepancies between clinics and imag- source of acute, recurrent and chronic pain.
ing also result from mistaking referred pain for radicular pain [41]. The origin of facet pain may be osteo-chondral, but the most
Axial spinal pain and referred pain are common, but radicular common initial event is thought a synovial reaction provoked from
pain is not. Although overestimated in the past, the real prevalence a trauma of the joint associated to capsular stretching or tearing
of radicular pain is only 12% or less of cases [41]. or capsular fluid distension. The syndrome of acute locked back is
While in case of radicular pain imaging can often demonstrate thought to be due to entrapment of fibroadipose meniscoid or their
the responsible lesion, in case of somatic referred pain this is often fragment within the facet joint. All these mechanisms provoke the
not possible. mechanical or chemical stimulation of joint nociceptors and type-C
Since inappropriate treatment can ensue from mistaking pain fibres of the densely innervated capsules.
referred pain for radicular pain, an accurate clinical evaluation is
mandatory. Radicular pain is lancinating, shooting down along a
narrow well localized band, whereas the somatic referred pain
starts in the back and spreads into the buttock and thigh within
a larger and ill-localized, deep area, where it is usually less intense
than the axial component [41].
Radicular and referred pain can coexist. A patient after disc her-
niation surgery can continue to complain axial discogenic pain after
removal of radicular pain. The role of imaging in predicting the
outcome also remains debated. In MR imaging, the degree and
thickness of herniation-encircling rim enhancement, expressing
neovascularized granulation tissue, are proportional to the ten-
dency to spontaneous resorption [55].
However, while the extruded discs show more improvement
in patients with acute LBP or sciatica the type, size, and location
of herniation at presentation, and the changes in herniation size
and type over time at MR do not correlate with outcome and MR
imaging does not appear to be valuable in predicting outcome and
planning conservative care [56].
Once again, any excessive confidence in imaging without con-
cordant clinical findings may lead to wrong and not indicated
treatments.
Selective nerve root blocks consent to establish the diagnosis of
radicular pain in atypical presentations such a mismatch between
imaging, electromyography and clinics.
The other main cause of radicular pain is the central and lateral
canal stenosis and foramenal stenosis.
The pathophysiology of pain and neurogenic claudication sec-
ondary to spinal stenosis is not completely elucidated.
Vascular and CSF flow impairment do not account for spinal
stenosis in asymptomatic subjects neither for the utility of injec-
tions of corticosteroids in symptomatic subjects, which once again,
suggests a role of inflammatory mechanisms also in this pathology.
In the presence of cocausative inflammatory mechanisms and
in case of dynamic canal stenosis, conventional static imaging is
expected to show poor concordance with clinics.
Axial loaded MRI can detect changes in the dural sac cross-
sectional area which conventional studies cannot demonstrate.
Upright MRI allows patients to reproduce the positions that elicit
their symptoms and also may uncover MRI findings that were not
visible with routine supine imaging.
The role played by inflammatory mechanisms in radicular pain
has supported the large diffusion of epidural or perineural injec-
tions as minimally invasive procedures.

7. Facet pain
Fig. 10. MR FSE T2 fat-sat axial (a) and FSE T1 fat-sat (b) axial images of L4-L5 facet
joints. During the evolution of facet degeneration it can overlap repeated arthritic
A possible cause of back and referred pain is pain emanating phases. The contrast medium administration consent to distinguish between a sim-
from the zygapophysial joints. ple joint effusion and a synovitis.

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
10 R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx

CT is more sensitive and accurate than MR for depicting Z-joint

space thinning and bone sclerosis.
MR overcomes CT in detecting joint effusion and cysts, synovi-
tis (Fig. 10), facet and periapophyseal inflammation and oedema
(Fig. 11), thanks to routine employment of FSE STIR/T2 fat sat and
eventual FSE T1 fat sat sequences after gadolinium administration
[58].
However, degenerative facets changes assessed by CT did not
show any significant correlation with the results of diagnostic intra-
articular injections.
The low reliability of either clinical and imaging findings ren-
ders challenging the choice as to what patient undergo a diagnostic
intra-articular injection or medial branch block.
Controlled diagnostic blocks are the only means of making a
diagnosis of zygapophysial joint pain.
Single diagnostic blocks have been shown to be associated with a
high false positive rate whereby, for the likelihood of false-positive
responses to be reduced, repeated blocks are required.
In patients whose pain is relieved by such blocks radiofrequency
medial branch neurotomy may be considered.
Nath conducted a randomized controlled study of percutaneous
radiofrequency neurotomy in 40 patients with chronic low back
pain (20 active and 20 controls) selected after repeated blocks,
reporting in the active treatment group statistically significant
improvement in back and leg pain as well as in quality of life vari-
ables [59].

8. Conclusions

Diagnostic assessment of the spinal pain still remains a chal-

lenge, despite the widely diffusion, because of the complex
anatomy and function of the spine.
No imaging modality could be considered as a gold standard
to diagnose the pain source, neither a clear correlation between
symptoms and imaging: degenerative changes and abnormal
movements are very often present also in asymptomatic individ-
uals.
The concepts regarding the origin and evolution of spinal pain
are radically changed in the last decades.
In 1970s and 1980s most of clinicians focused only on the
anatomical and biological aspects of acute and chronic spinal pain,
Fig. 11. MR FSE T2 (a) and FSE T1 (b) axial fat-sat images of L5-S1 facet joints shoving considering the severity of the acuteness and the association of
inflammatory changes extending far beyond the joints into periarticular tissues,
Contrast medium administration can better depict the full extension of changes.
radiculopathy as principal predictors of recurrency and chronicity.
The complete assessment of site and extension of changes is essential in guiding Actually, while the onset of spinal pain seems to have an
therapeutic approaches such as steroid and ozone percutaneous infiltrations. unequivocal biologic origin, its evolution is considered to be mainly
conditioned by psychological and social risk factors dealing with
dysfunction and abnormal load distribution in and between motion
The facets joints degeneration is also a frequent cause of radicu- segments.
lar pain by compression of nerve roots in lateral recesses and in the This growing knowledge leads to radically reconsider the
foramina secondary to hypertrophic and ostephytic remodelling of approach to patients complaining chronic pain and disability and to
the facets, sublussation, joint effusion with capsular tension, syno- contribute in justifying the limitations and controversial imaging
vial cysts. results in most patients.
There are no features, on history or clinical examination,
whereby lumbar zygapophysial joint pain can be diagnosed clin- References
ically, nor specific symptoms correlate with the positive responses
to diagnostic joint injections, but a patient over 60 who experiences [1] Kang DJ, Hanks S. Inflammatory basis of spinal pain. In: Slipman CW, Derby
pain during standing and walking, relieved by sitting and worsened R, Simeone FA, et al., editors. Interventional Spine. An algorithmic approach.
Philadelphia: Saunders Elsevier; 2008. p. 17–27.
by ipsilateral extension and palpation is suspect. [2] Adams MA. Biomechanics of back pain. Acupunture in Medicine 2004;22(4):
Axial lumbar pain and referred pain to the extremities may be 178–88.
elicited by injecting saline solution in synovial spaces and into z- [3] Andersson GBJ. Epideliological features of chronic low-back pain. Lancet
1999;354:581–5.
joint capsulae, respectively. Referred lumbar pain generally does
[4] Cote P, Cassidy JD, Carroll L. The Saskatchewan health and back pain survey. The
not extend over the knee. The cervical facetogenic pain is not prevalence of neck pain and related disability in Saskatchewan adults. Spine
referred distally to the elbow. 1998;23(15):1689–98.
Facet arthrosis and arthritis are as many frequent as in patients [5] Pedersen HE, Blunck CFJ, Gardner E. The anatomy of lumbosacral posterior
rami and meningeal branches of spinal nerves (sinu-vertebral nerves). With
with or without lumbar pain, but the most evolved lesions tend to an experimental study of their functions. J Bone Joint Surg [Am] 1956;38-
be more often symptomatic. A:377–91.

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018
G Model
EURR-7021; No. of Pages 11 ARTICLE IN PRESS
R. Izzo et al. / European Journal of Radiology xxx (2015) xxx–xxx 11

[6] Bogduk N. Nerves of the lumbar spine. In: Bogduk N, editor. Clinical anatomy of [33] Bogduk N. Low Back Pain. In: Bogduk N, editor. Clinical and Radiological
the lumbar spine and sacrum. London: Churchill Livingstone; 1997. p. 127–44. Anatomy of the Lumbar Spine. London: Churchill Livingstone, Elsevier; 2012.
[7] Jinkins JR, Whittemore AR, Bradley WG. The anatomic basis of vertebrogenic p. 173–96.
pain and the autonomic syndrome associated with lumbar disk extrusion. AJR [34] Sachs BL, Vanharanta H, Spivey MA, et al. Dallas discogram descrription: a new
1989;152:1277–89. classification of CT/discography in low-bach disorders. Spine 1987;12:287–94.
[8] Bogduk N. On the definitions and physiology of back pain, referred pain and [35] Vanharanta H, Sachs BL, Spivey MA, et al. The relationship of pain provocation
radicular pain. Pain 2009;147:17–9. to lumbar disc deterioration as seen by CT/discography. Spine 1987;12:295–8.
[9] Moskowitz MH. Central influences on pain. In: Slipman CW, Derby R, Simeone [36] Bogduk N, Aprill C, Derby R. Lumbar discogenic pain: state-of-the-art review.
FA, et al, editors. Interventional spine. An algorithmic approach. Philadelphia: Pain Med 2013;14(6):813–36.
Saunders Elsevier; 2008. p. 39–52. [37] Saifuddin A, Braithwaite I, White J, et al. The value of lumbar spine magnetic
[10] Coderre C. Excitatory amino acid antagonists: potential analgesics for persis- resonance imaging in the demonstration of anular tears. Spine 1998;23:453–7.
tent pain. In: Sawynok, Cowan, editors. Novel aspects of pain management. [38] Schellhas KP, Pollei SR, Gundry CR, Heithoff KB. Lumbar disc high-intensity
New York: Wiley Liss; 1999. p. 157–78. zone. Corrrelation of magnetic resonance imaging and discography. Spine
[11] De Leo JA, Yezierski RP. The role of neuroinflammation and neuroimmune acti- 1996;21:79–86.
vation in persistent pain. Pain 2001;91:1–6. [39] Thomson KJ, Dagher AP, Eckel TS, Clark M, Reinig JW. Modic changes on MR
[12] Rutkowski MD, Winkelstein BA, Hickey WF, et al. Lumbar nerve root injury images as studied with provocative diskography: clinical relevance—A retro-
induces CNS neuroimmune activation and neuroinflammation in the rat: rela- spective study of 2,457 disks. Radiology 2009;250:849–55.
tionship to painful radiculopathy. Spine 2002;27(15):1604–13. [40] Weishaupt D, Zanetti M, Hodler J, et al. Painful lumbar disk derangement: rel-
[13] Negrini S, Bonaiuti D, Monticone M, et al. Medical causes of low back pain. evance of endplate abnormalities at MR imaging. Radiology 2001;218:420–7.
In: Slipman CW, editor. Interventional Spine. Philadelphia: Saunders; 2008. p. [41] Bogduk N. On the definitions and physiology of back pain, referred pain and
803–11. radicular pain. Pain 2009;147:17–9.
[14] Taimela S, Osterman K, Alaranta H, et al. Long psychomotor reaction time [42] Bernard Jr TN. Lumbar discography followed by computed tomography.
in patients with chronic low-back pain: preliminary report. Arch Phys Med Refining the diagnosis of low-back pain. Spine 1990;15:690–707.
Rehabil 1993;74:1161–4. [43] Milette PC, Melanson D. A reappraisal of lumbar discography. J Can Assoc Radiol
[15] Croft PR, Papageorgiou AC, Ferry S, et al. Psycologic distress and low back 1982;33:176–82.
pain. Evidence from a prospective study in the general population. Spine [44] Carragee EJ, Tanner CM, Khurana S, et al. The rates of falsepositive lumbar
1995;20(24):2731–7. discography in select patients without low back symptoms. Spine 2000;25:
[16] Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunction, 1373–81.
adaptation and enhancement. J Spinal Disord 1992;5:383–9. [45] Colhoun E, McCall IW, Williams L, et al. Provocation discography as a guide to
[17] Panjabi MM. A hypothesis of chronic back pain: ligament subfailure injuries planning operations on the spine. J Bone Joint Surg (Br) 1988;70(2):267–71.
lead to muscle control dysfunction. Eur Spine J 2006;15:668–76. [46] Gill K, Blumenthal SL. Functional results after anterior lumbar fusion at L5-S1
[18] Kirkaldy-Willis WH, Farfan HF. Instability of the lumbar spine. Clin Orthop in patients with normal and abnormal MRI scans. Spine 1992;17:940–2.
1982;165:110–23. [47] Derby R, Howard MW, Grant JM, et al. The ability of pressure-controlled discog-
[19] Jinkins JR. Acquired degenerative changes of the intervertebral segments at and raphy to predict surgical and non-surgical outcomes. Spine 1999;24(4):364–72.
suprajacent to the lombosacral junction. A radioanatomic analysis of the non [48] Howe JF, Loeser JD, Calvin WH. Mechanosensitivity of dorsal root ganglia and
discal structures of the spinal column and perispinal soft tissues. Eur J Radiol chronically injured axons: a physiological basis for the radicular pain of nerve
2004;50(2):134–58. root compression. Pain 1977;3:25–41.
[20] Chaput C, Padon D, Rush J, et al. The significance of increased fluid signal on [49] Gertzbein SD, Tile M, Gross A, et al. Autoimmunity in degenerative disc disease
magnetic resonance imaging in lumbar facets in relationship to degenerative of the lumbar spine. Ortop Clin North Am 1975;6:67–73.
spondylolisthesis. Spine 2007;32(17):1883–7. [50] Saal JS, Franson RC, Dobrow R, et al. High levels of inflammatory phospholipase
[21] Fujiwara A, Lim TH, An HS, et al. The effect of disc degeneration and facet A2 activity in lumbar disc herniations. Spine 1990;15(7):674–8.
joint osteoarthritis on the segmental flexibility of the lumbar spine. Spine [51] Olmaker K, Blomquist I, Stromberg I, et al. Inflammogenic properties of nucleus
2000;25(23):3036–44. pulposus. Spine 1995;20:665–9.
[22] Cartolari R. Functional evaluation of the lumbar spine with axial-loaded com- [52] Winkelstein B, Rutkowski M, Weinstein J, et al. Quantification of neural tis-
puter tomography and cine ALCT. Riv Neuroradiol 1997;10:569–84. sue injury in a rat radiculopathy model: comparison of local deformation,
[23] Matsunaga S, Sakou T, Morizono Y, et al. Natural history of degenerative behavioural outcomes, and spinal cytokines mRNA for two surgeons. J Neurosci
spondylolisthesis: pathogenesis and natural course of the slippage. Spine Methods 2001;111:49–57.
1990;15:1204–10. [53] Hashizume H, DeLeo J, Colburn R, et al. Spinal glial activation and cytokine
[24] Alyas F, Connell D, Saifuddin A. Upright positional MRI of the lumbar spine. Clin expression following lumbar root injury in the rat. Spine 2000;25:1206–17.
Radiol 2008;63:1035–48. [54] Jensen MC, Brant-Zawadzki MN, Obuchowski N, et al. Magnetic resonance
[25] Leone A, Guglielmi G, Cassar-Pullicino VN, et al. Lumbar intervertebral insta- imaging of the lumbar spine in people without back pain. N Engl J Med
bility: a review. Radiology 2007;245:62–77. 1994;331(2):69–73.
[26] Mulholland RC. The myth of lumbar instability: the importance of abnormal [55] Spendiani A, Puglielli E, De Amicis R, et al. Spontaneous resolution of lumbar
loading as a cause of low back pain. Eur Spine J 2008;17:619–25. disck herniation: predictive signs for prognostic evaluation. Neuroradiology
[27] Mooney V. Where is the pain coming from? Spine 1987;12:754–9. 2004;46:916–22.
[28] Schwarzer AC, Aprill CN, Derby R, et al. The prevalence and clinical features [56] Modic MT, Obuchowski NA, Ross JS, et al. Acute low back pain and radiculopa-
of internal disc disruption in patients with chronic low back pain. Spine thy. Radiology 2005;237:597–604.
1995;20:1878–83. [57] Manchikanti L, Singh V, Vidyasagar P, et al. Evaluation of the relative con-
[29] Crock HV. A reappraisal of intervertebral disc lesions. Med J Aust 1970;1: tributions of various structures in chronic low back pain. Pain Physician
983–9. 2001;4(4):308–16.
[30] Freemont AJ, Peacock TE, Goupille P, et al. Nerve ingrowth into diseased inter- [58] D‘Aprile P, Tarantino A, Jinkins R, et al. The value of fat saturation sequences and
vertebral disc in chronic back pain. Lancet 1997;350:178–81. contrast medium administration in MRI of degenerative disease of the poste-
[31] McMahon SB. Mechanisms of sympathetic pain. Br Med Bull 1991;47: rior/perispinal elements of the lumbosacral spine. Eur Radiol 2007;17:523–31.
584–600. [59] Nath S, Nath CA, Pettersson K. Percutaneous lumbar zygapophysial (Facet) joint
[32] Peng BG. Pathophysiology, diagnosis, and treatment of discogenic low back neurotomy using radiofrequency current, in the management of chronic low
pain. World J Orthop 2013;4(2):42–52. back pain: a randomized double-blind trial. Spine 2008;33(12):1291–7.

Please cite this article in press as: Izzo R, et al. Spinal pain. Eur J Radiol (2015), http://dx.doi.org/10.1016/j.ejrad.2015.01.018