Journal of Small Animal Practice • Vol 60 • June 2019 • © 2019 British Small Animal Veterinary Association 367

PAPER

Journal of Small Animal Practice (2019) 60, 367–373DOI: 10.1111/jsap.12979

Accepted: 21 November 2018; Published online: 31 January 2019

Low-field magnetic resonance changes in the paravertebral musculature of dogs with acute intervertebral disc extrusionA. R. R. Furtado1, G. B. Cherubini and O. Taeymans

Dick White Referrals, Six Mile Bottom, Cambridge CB80UH, UK

1Corresponding author email: ritarfurtado@gmail.com

Objectives: To describe the MRI features and prevalence of paravertebral muscle signal intensity

changes in dogs with acute intervertebral disc extrusion and to search for associations between the

signal changes and clinical history, signalment, neurological examination, serum creatine kinase

activity and MRI characteristics of the disc herniation.

Materials and MethOds: Medical records and MRI examinations from 688 dogs with surgically confirmed

acute intervertebral disc extrusion were reviewed retrospectively. T2-weighted and STIR MRI

sequences were available for 276 cases and were examined for paravertebral muscle signal intensity

changes. When present, extension, lateralisation and signal characteristics of these changes were

recorded. Exclusion criteria were muscle injections 24 hours before MRI scan, trauma and previous

spinal surgery.

results: Nineteen dogs met the inclusion criteria. There were signal changes in the multifidus muscle,

mostly in the thoracolumbar region and often extending caudally from the level of the intervertebral

disc herniation. Two cases had paravertebral muscle signal intensity changes in the cervical region.

MRI signal changes were seen more frequently in the muscles of non-ambulatory dogs. Clinical history

and neuro-examination did not allow differentiation between dogs with and without paravertebral

muscle signal intensity changes.

clinical significance: Paravertebral muscle signal intensity changes were observed infrequently in the

epaxial musculature of 6.9% dogs with acute intervertebral disc extrusion in both the thoracolumbar

and cervical regions. The pathophysiological processes responsible for these MRI changes

remain unknown.

INTRODUCTION

Hansen type I intervertebral disc extrusion (IVDE) can occur very acutely and lead to variable degrees of spinal cord compres-sion, depending on the volume of extruded disc material (Brisson 2010). Concurrent haemorrhage, oedema, vascular and neuronal injury are also associated with this condition (McKee 2000).

MRI is considered the gold standard for detection of IVDE disease in dogs (Robertson & Thrall 2011) and is a sensitive

modality for detecting different types of muscular changes (Hen-derson et al. 2015, Lerer et al. 2015, Kaiser et al. 2016). Paraver-tebral muscles can be subdivided into hypaxial muscles (longus colli, rectus capitis ventralis, longus capitis, scalenus, quadratus lum-borum, psoas minor and iliopsoas) which are flexors of the neck and tail; and epaxial muscles (iliocostalis, longissimus, transverso-spinalis and multifidus) that are located dorsal to the transverse processes of the vertebrae and are responsible for extension of the vertebral column.

http

://w

ww

.bsa

va

.co

m/

A. R. R. Furtado et al.

368 Journal of Small Animal Practice • Vol 60 • June 2019 • © 2019 British Small Animal Veterinary Association

Paravertebral muscle changes in dogs have previously been reported to be associated with fat infiltration (Boström et al. 2014, Lerer et al. 2015), infection (Holloway et al. 2009), inflam-mation (Platt et al. 2006) and ischaemia secondary to fibrocarti-laginous embolism (Sehl et al. 2017).

The objectives of this retrospective study were: (1) to describe the MRI features of paravertebral muscle abnormalities in dogs with acute IVDE and (2) to search for associations between the muscle signal changes and clinical history, signalment, neurologi-cal examination, serum creatine kinase (CK) activity and MRI characteristics of the IVDE.

MATERIAL AND METHODS

MRI images of all dogs with surgically confirmed IVDE between January 2015 and July 2017 at our institution were reviewed. Inclusion criteria for the study were at least one plane of an MRI fat suppression sequence and available clinical data, whereas exclusion criteria for the study were: intramuscular injections less than 24 hours before MRI, history of previous spinal surgery or recent external trauma.

Information retrieved from the clinical records included age, gender, breed, bodyweight, onset of clinical signs, duration of clinical signs (less than 24 hours or more than 24 hours), pres-ence and site of spinal pain (cervical, thoracolumbar or lumbar), neurological dysfunction (ambulatory paresis, non-ambulatory paresis or plegia), neuro-localisation and serum CK activity (low reference interval boundary: 190 IU/L).

A 0.4-Tesla magnet (Aperto Lucent; Hitachi Medical Corpo-ration) was used for all dogs. All studies were performed with the dogs under general anaesthesia. Dogs were premedicated according to the anaesthetists’ preferences, induced with propo-fol (Propoflo Plus; Zoetis) to effect and general anaesthesia was maintained with isoflurane (IsoFlo; Abbott) in oxygen. All the cases were placed in dorsal recumbency with the area of interest inside a human knee or open-head coil depending on the size of the covered body area.

T2-weighted (T2W) fast spin-echo images in sagittal plane and transverse plane for the affected intervertebral disc space(s) were obtained in all dogs. In addition, as an inclusion crite-rion, all dogs had to have short tau inversion recovery (STIR) sequences available for review, in sagittal or transverse plane (for four cases) at the level of the disc and muscle hyperintensity.

In two dogs, additional sequences included T1-weighted (T1W) spin-echo sequences before and after manual intravenous bolus administration of paramagnetic contrast agent (1 mmol/mL, Gadobutrol, Gadovist, Bayer, Bayer House) at a dose of 0.1 mL/kg of bodyweight. Slice thickness ranged from 3 to 4 mm in all planes.

MRI images were reviewed by two authors and were evalu-ated for muscle signal intensity changes (MSIC) within the paravertebral musculature. These changes were characterised by ill-defined, hyperintense areas, in T2W and STIR sequences.

Location and extension of the MSIC when compared to the IVDE (caudal, cranial or at the IVDE site), side of MSIC (left,

Table 1. Summary of results in dogs with and without muscle changes

Comparator MSIC

Clinical signsNeurological

dysfunctionAmbulatory paresis 15 (78.9%) 8 (42.1%)Non-ambulatory paresis 3 (15.8%) 7 (36.8%)Plegia 1 (5.3%) 4 (21.1%)

Duration of the clinical signs

≤ 24 hours 9 (47.9%) 13 (68.4%)>24 hours 10 (52.1%) 6 (31.6%)

Neuro-localisation C1 to C5 2 (10.5%) 2 (10.5%)C6 to T2 0 0T3 to L3 13 (68.4%) 15 (78.9%)L4 to S3 4 (21.1%) 2 (10.5%)

Spinal pain Present 11 (64.7%) 10 (71.4%)Absent 6 (35.3%) 4 (28.6%)

Blood testsCK Normal 4 (28.6%) 3 (37.5%)

Abnormal 10 (71.4%) 5 (62.5%)MRI findingsIVDE lateralisation Right 7 (36.8%) 16 (84.2%)

Left 12 (63.2%) 3 (15.8%)Degree of

spinal cord compression

Mild 5 (26.3%) 2 (10.5%)Moderate 10 (52.6%) 13 (68.4%)Severe 4 (21.1%) 4 (21.1%)

Cord contusion/oedema

Present 3 (31.6%) 3 (31.6%)Absent 13 (68.4%) 13 (68.4%)

MSIC muscle signal intensity changes, CK creatine kinase, IVDE intervertebral disc extrusion

right or both) and group of muscles affected (epaxial, hypaxial or both) were documented. Other recorded MRI findings included severity of spinal cord compression (mild <30%, moderate 30 to 60% or severe >60%), lateralisation of the IVDE (right or left) and intramedullary spinal cord signal changes.

The clinical signs and MRI features of the patients with MSIC were then compared with 19 patients (comparator group), that were selected from a population of dogs affected by acute IVDE but without MSIC on T2W and STIR. These cases were matched with the MSIC group by applying a systematic random sampling method.

Statistical analysis of the data was performed using SPSS® 17.0 (SPSS Inc.). The main results and associations are summarised in Table 1.

RESULTS

Between January 2015 and July 2017, 688 dogs underwent MRI and surgery for IVDEs. Of those, the most common breeds were 138 dachshunds (20%), 109 cocker spaniels (15.9%), 64 French bulldogs (9.3%), 25 Jack Russell terriers (3.6%) and 23 Lhasa apso dogs (3.3%). All dogs had T2W sequences available but only 276 cases had STIR images available for review.

SignalmentNineteen dogs (6.9%) showed MSIC, of which 16 (84.2%) were purebred including three French bulldogs (15.8%), three cocker spaniels (15.8%), two miniature schnauzers (10.5%), two Staf-fordshire bull terriers (10.5%) and one dachshund (5.2%). The ages ranged from 2.6 to 11.0 years [mean 6.5 years ± standard

Paravertebral muscle changes in dogs with IVDE

Journal of Small Animal Practice • Vol 60 • June 2019 • © 2019 British Small Animal Veterinary Association 369

deviation (sd) 2.5 years]. There were 16 male dogs (84.2%; 14 neutered and two entire) and three female neutered (15.8%) dogs. Bodyweight ranged from 6.8 to 32 kg (mean 14.3 ± 6.2 kg).

Nineteen dogs were included as a comparator group, of which 16 (84.2%) were purebred including three Jack Russell terri-ers (15.8%), three dachshund (15.8%), two French bulldogs (10.5%), one cocker spaniel (5.3%) and one miniature schnauzer (5.3%). The ages of this group ranged from 1.2 years to 9.5 years (mean 5.3 years ± sd 2.8 years). There were 13 male dogs (68.4%; 10 neutered and three entire) and six female neutered (31.6%). Bodyweight ranged from 6.7 to 33.8 kg (mean 12.1 ± 6.9 kg).

Clinical signsIn the MSIC group, eight patients (42.1%) presented with ambulatory paresis, seven patients (36.8%) with non-ambula-tory paresis and four (21.1%) with plegia on initial neurologi-cal examination. In the comparator group 15 (78.9%) presented ambulatory and four were non-ambulatory (21.1%).

All dogs had acute onset of neurological signs. Thirteen cases (68.4%) with MSIC presented to the hospital within 24 hours of onset, whereas 10 (52.1%) cases without muscle changes pre-sented more than 24 hours after onset.

Presence or absence of pain was documented in 14 dogs from the MSIC group and 17 from the comparator group. Ten dogs (71.4%) with MSIC and 11 (64.7%) comparator cases exhibited spinal pain. The most common site of pain was the thoracolum-bar region for both groups (four comparator cases and five MSIC cases).

CK activity was measured before MRI in eight dogs with MSIC and in 14 comparator cases. Five (62.5%) with MSIC and 10 (71.4%) comparator cases had increased CK activity.

Fifteen out of 19 (78.9%) cases with MSIC had T3 to L3 lesions and the most commonly affected discs were T12 to T13 (four cases, 7.6%) and T13 to L1 (four cases, 7.6%). Similar results were seen for the comparator group in which 13 cases (68.4%) had T3 to L3 lesions (Fig.1). The most common disc space affected was T12 to T13 (five cases, 26.3%), followed by L2 to L3 (four cases, 21%). The only two cases (10.5%) with C1 to C5 lesions and MSIC were cocker spaniels with C3 to C4 IVDE.

FIG 1. Patient distribution of MSIC group and comparator group according to their neuro-localisation

MRI findingsTable 2 describes the individual MRI changes for the affected dogs. Ten out of 19 comparator cases (52.6%) had moderate spi-nal cord compression, five (26.3%) had mild compression and four (21.1%) had severe compression. Similar results were seen for the MSIC group, in which the spinal cord compression was moderate in 13 cases (68.4%), severe in four (21.1%) and mild in two (10.5%). All dogs with severe spinal cord compression had signs of pain on admission.

Cord contusion/oedema, defined as increased T2W signal intensity of the cord parenchyma, was absent in the majority of comparator and MSIC cases.

Muscle intensity changesAll dogs included in the MSIC group had notable regional epaxial musculature hyperintense signal intensity, compared to surrounding muscles, on T2W and STIR images (Fig. 2). This most specifically affected the multifidus muscle (Fig. 3); only two cases (10.5%) had concurrent hypaxial musculature changes. MSIC were noted most frequently (94.7%) at the level of the affected IVD, and extended caudally (52.6%) more commonly than cranially (15.8%).

Fourteen cases (73.7%) had unilateral MSIC, with half of the cases on the right side and the other half on the left side. The majority of these changes did not occur on the same side as the disc extrusion and six cases with right-sided IVDE had muscle signal changes in the left epaxial paravertebral musculature and four had bilateral muscle signal changes.

HistopathologyOne of the cases with MSIC had muscle biopsy samples taken during hemilaminectomy. Histopathology examination was indicative of severe muscle degeneration and necrosis associ-ated with mild interstitial lymphoplasmacytic and neutrophilic inflammation (Fig. 4).

DISCUSSION

Paravertebral muscles constitute one of the largest muscle masses of the body and play an important role in stabilisation and mobil-ity of the vertebral column (Bierry et al. 2008). Degenerative spinal disease consists of disc desiccation and facet joint osteo-arthritis, which lead to loss of discal height and loss of articular facet joint congruity, subsequently leading to spinal instability (Fujiwara et al. 2000, Jinkins 2004). This instability is repeatedly reported in humans as a cause for acute and subacute damage of the paravertebral muscles, either directly (myogenic atrophy due to myonecrosis) or indirectly by traumatic denervation (neu-rogenic atrophy) (Jinkins 2004, Bierry et al. 2008, Freeman et al. 2010). These acute/subacute changes can be demonstrated by signal hyperintensity on fat-suppressed T2W images and have been associated with contrast enhancement on T1W images (Osamura et al. 2000, Kim & Kim 2011). Fat infiltration occurs in the later stages of muscle degeneration, and chronic paraverte-bral muscle atrophy therefore appears suppressed on fat-nulling sequences (Kjaer et al. 2007, Freeman et al. 2010).

A. R. R. Furtado et al.

370 Journal of Small Animal Practice • Vol 60 • June 2019 • © 2019 British Small Animal Veterinary Association

Tabl

e 2

. Li

st o

f al

l do

gs w

ith

mu

scle

sig

nal

chan

ges

wit

h th

e d

ura

tio

n o

f cl

inic

al s

igns

, pre

senc

e o

f sp

inal

pai

n, n

euro

logi

cal dy

sfunc

tion,

neuro

-loca

lisa

tion,

CK

le

vels

, ext

ens

ion

of

MS

IC w

hen

com

pare

d to

the

IV

DE

, sid

e o

f M

SIC

, lev

el o

f sp

inal

co

rd c

om

pre

ssio

n an

d la

tera

lisa

tion

of

the I

VD

E

Bre

edA

geSex

Dur

atio

n of

cl

inic

al s

igns

Spi

nal

pain

Neu

rolo

gica

l dy

sfun

ctio

nN

euro

-lo

calis

atio

nC

K le

vels

Exte

nsio

n of

the

mus

cle

sign

al in

tens

ity

chan

ges

rela

tive

to

the

IVD

E si

te

Sid

e of

mus

cle

sign

al in

tens

ity

chan

ges

Sev

erit

y of

sp

inal

cor

d co

mpr

essi

on (

%)

Late

ralis

atio

n of

the

IVD

E

1M

ixed

bre

ed6 y

ears

and

11 m

onth

sM

N>24 h

ours

Pres

ent

Ambu

lato

ry p

ares

isT3

to

L3N

orm

alC

auda

lR

ight

40

Left

2Fr

ench

bul

ldog

2 y

ears

and

10 m

onth

sM

N≤24 h

ours

Unk

now

nN

on-a

mbu

lato

ry

pare

sis

T3 t

o L3

Hig

hC

auda

lLe

ft30

Left

3C

aval

ier

Kin

g C

harle

s sp

anie

l6 y

ears

MN

>24 h

ours

Unk

now

nN

on-a

mbu

lato

ry

pare

sis

L4 t

o S

3U

nkno

wn

Cau

dal

Bila

tera

l70

Rig

ht

4B

eagl

e7 y

ears

and

7 m

onth

sFN

>24 h

ours

Abse

ntN

on-a

mbu

lato

ry

pare

sis

T3 t

o L3

Unk

now

nAt

the

leve

l of

the

IVD

ER

ight

50

Rig

ht

5M

inia

ture

sc

hnau

zer

7 y

ears

and

6 m

onth

sM

N≤24 h

ours

Pres

ent

Ambu

lato

ry p

ares

isT3

to

L3U

nkno

wn

Cau

dal

Left

50

Rig

ht

6M

inia

ture

da

chsh

und

5 y

ears

MN

>24 h

ours

Pres

ent

Ambu

lato

ry p

ares

isT3

to

L3U

nkno

wn

Cau

dal

Left

40

Rig

ht

7C

ocke

r sp

anie

l6 y

ears

MN

>24 h

ours

Pres

ent

Ambu

lato

ry p

ares

isL4

to

S3

Nor

mal

At t

he le

vel o

f th

e IV

DE

Rig

ht80

Rig

ht8

Min

iatu

re

schn

auze

r4 y

ears

MN

≤24 h

ours

Abse

ntN

on-a

mbu

lato

ry

pare

sis

T3 t

o L3

Unk

now

nC

rani

alLe

ft20

Rig

ht

9C

ocke

r sp

anie

l11 y

ears

MN

≤24 h

ours

Pres

ent

Ambu

lato

ry p

ares

isC

1 t

o C

5H

igh

At t

he le

vel o

f th

e IV

DE

Bila

tera

l50

Left

10

Bas

set

houn

d9 y

ears

and

2 m

onth

sFN

≤24 h

ours

Unk

now

nAm

bula

tory

pare

sis

T3 t

o L3

Nor

mal

Cau

dal

Left

40

Rig

ht

11

Sta

ffor

dshi

re b

ull

terr

ier

5 y

ears

and

6 m

onth

sFN

≤24 h

ours

Pres

ent

Ambu

lato

ry

para

pare

sis

T3 t

o L3

Unk

now

nAt

the

leve

l of

the

IVD

EB

ilate

ral

50

Rig

ht

12

Mix

ed b

reed

4 y

ears

and

3 m

onth

sM

N>24 h

ours

Pres

ent

Pleg

iaT3

to

L3U

nkno

wn

At t

he le

vel o

f th

e IV

DE

Bila

tera

l80

Rig

ht

13

Mix

ed b

reed

10 y

ears

MN

≤24 h

ours

Pres

ent

Non

-am

bula

tory

pa

resi

sT3

to

L3U

nkno

wn

Cau

dal

Bila

tera

l40

Rig

ht

14

Fren

ch b

ulld

og3 y

ears

and

4 m

onth

sM

E≤24 h

ours

Unk

now

nPl

egia

T3 t

o L3

Hig

hC

rani

alR

ight

60

Rig

ht

15

Fren

ch b

ulld

og2 y

ears

and

8 m

onth

sM

N≤4 h

ours

Abse

ntN

on-a

mbu

lato

ry

pare

sis

T3 t

o L3

Hig

hC

rani

alR

ight

20

Rig

ht

16

Engl

ish

sprin

ger

span

iel

7 y

ears

and

11 m

onth

sM

N≤24 h

ours

Unk

now

nPl

egia

T3 t

o L3

Unk

now

nAt

the

leve

l of

the

IVD

ER

ight

50

Rig

ht

17

Coc

ker

span

iel

9 y

ears

and

9 m

onth

sM

N≤24 h

ours

Pres

ent

Ambu

lato

ry p

ares

isC

1 t

o C

5H

igh

Cau

dal

Left

30

Rig

ht

18

Sta

ffor

dshi

re b

ull

terr

ier

5 y

ears

and

2 m

onth

sM

E≤24 h

ours

Abse

ntN

on-a

mbu

lato

ry

pare

sis

T3 t

o L3

Unk

now

nC

auda

lR

ight

40

Rig

ht

19

Jack

Rus

sell

terr

ier

8 y

ears

and

3 m

onth

sM

N≤24 h

ours

Pres

ent

Pleg

iaT3

to

L3U

nkno

wn

Cau

dal

Left

80

Rig

ht

MN

mal

e ne

uter

ed, M

E, m

ale

entir

e, F

N f

emal

e ne

uter

ed, I

VDE

inte

rver

tebr

al d

isc

extr

usio

n, M

SIC

mus

cle

sign

al in

tens

ity c

hang

es, C

K c

reat

ine

kina

se

Paravertebral muscle changes in dogs with IVDE

Journal of Small Animal Practice • Vol 60 • June 2019 • © 2019 British Small Animal Veterinary Association 371

FIG 2. Miniature schnauzer, male neutered, 4 years old. MRI (A, STIR sagittal and B, transverse) showing bilateral paravertebral muscle hyperintensity. B: note that the left side is more affected than the right

FIG 3. English springer spaniel, male neutered, 7 years and 11 months. A, T2W sagittal and B, STIR sagittal. Note the muscle hyperintensity signal changes on the multifidus muscle on both sequences (see Movies S1 and S2, Supporting Information)

Our study reports the prevalence and signal characteristics of muscle changes on MRI in dogs with surgically confirmed IVDE. A recently published study (Trampus et al. 2018) reported that paravertebral muscle changes occur in 36% of dogs with IVDE, which contrasts with the number we report (6.9%). Possible rea-sons for this discrepancy are the larger size of our study popula-tion and the lack of STIR images in some cases in the other study.

Dachshunds have been reported to be one of the most com-mon breeds affected by IVDE. However, only one dachshund

(0.36%) from our study population of 276 had muscle inten-sity changes. Instead, the most commonly included breeds in our MSIC group were French bulldogs and cocker spaniels. We spec-ulate that low number of dachshunds in the MSIC group might be because of the lack of available STIR images in the majority of them. However, dachshunds are predisposed to muscle fat infil-tration (Boström et al. 2014) and not acquiring STIR images may potentially mask other muscle hyperintensities on T2W images.

Non-ambulatory paraparesis was the most common presenta-tion in the MSIC group, whereas the comparator group predom-inantly remained ambulatory. This difference is in agreement with a previous report (Trampus et al. 2018).

Spinal pain associated with IVDE not only results from com-pression or ischaemia of the meninges, spinal cord and/or spinal nerve roots, but has also been reported as a result of local inflam-matory reactions that occur secondarily to the extruded disc (McKee 2000). In our study, the majority of cases with and with-out MSIC were painful in the thoracolumbar region. It would have been interesting to investigate if muscle changes were related to the severity of spinal pain but pain scoring was not available for review. In dogs, lateralisation of pain can be difficult to assess but in people with myogenic atrophy, the muscle signal changes are usually located ipsilateral to the side of pain (Bierry et al. 2008, Kim & Kim 2011). Further prospective studies need to be done to determine if the same occurs in veterinary medicine.

CK is a marker of skeletal muscle damage in dogs (Aktas et al. 1993). In our study, abnormal CK activity was found in the MSIC and control groups, but not all dogs with MSIC had high CK. Possible explanations could derive from the fact that CK has a short half-life of about 2 hours (Aktas et al. 1993), that not all cases with MSIC had CK measured, the muscle lesions might not have been sufficiently severe to increase measured CK activity, or some of the patients in the comparator group could have had injuries in muscles not included in the MRI examination.

The most commonly extruded IVD spaces, in the MSIC group, were T13 to L1 and T12 to T13, which is consistent with previous studies (McKee 2000, Ferreira et al. 2002, Penning et al. 2006, Trampus et al. 2018). Two cases showed cervical paraver-tebral signal intensity changes, documenting that these changes can be found beyond the thoracolumbar region. It is important to highlight the difference with signal changes previously reported in dogs with meningoencephalomyelitis of unknown aetiology (Eminaga et al. 2013). The latter being more linear along the ventral cortex of the cervical vertebral bodies.

The severity of spinal cord compression was assessed for all patients. In accordance to previous studies (Sukhiani et al. 1996, Besalti et al. 2005, Trampus et al. 2018), the severity of compres-sion was not associated with severity of neurological deficits or spinal pain, and also not associated with paravertebral muscle signal changes. Conversely, all dogs with reported spinal pain had more severe spinal compression.

The side of the IVDE was compared with the side of the para-vertebral muscle changes but no associations were found. Con-sidering the similarity of signal changes in people with acute back injury, a similar mechanism relating to increased range of motion of adjacent spinal segments can therefore be postulated in dogs.

A. R. R. Furtado et al.

372 Journal of Small Animal Practice • Vol 60 • June 2019 • © 2019 British Small Animal Veterinary Association

FIG 4. Histopathology of the paravertebral muscle, showing variably sized myocytes with frequent degeneration and either pale, swollen or vacuolated sarcoplasm. Note the contrast in appearance between normal muscle fibres (N), myofibres with vacuolated sarcoplasm (V) and with hypercontraction bands (H) indicating degeneration and myofibres with fragmented sarcoplasm indicating muscle necrosis

The fact that the multifidus muscle was usually the only affected muscle is interesting because of its exclusive innervation by the medial ramus of the dorsal root of the spinal nerve (Bierry et al. 2008). This could imply a component of focal nerve root compres-sion/neuritis either directly caused by disc material or by extension of meningeal inflammation. These muscle changes could also be the result of muscle spasms but the lack of association with spinal pain and disc lateralisation would make this less likely.

In conclusion, paravertebral muscle hyperintensity changes were seen in 6.9% dogs with acute IVDE. It was most common in French bulldogs and cocker spaniels. Muscle signal changes were seen more frequently in non-ambulatory dogs, occurred both ipsilateral and contralateral to the side of disc extrusion and clinical history and neuro-examination did not allow differentia-tion between dogs with and without MSIC. The pathophysiol-ogy of MSIC remains unknown.

Conflict of interestNone of the authors of this article has a financial or personal relationship with other people or organisations that could inap-propriately influence or bias the content of the paper.

ReferencesAktas, M., Auguste, D., Lefebvre, H. P., et al. (1993) Creatine kinase in the dog: a

review. Veterinary Research Communications 17, 353-369Besalti, O., Ozak, A., Pekcan, Z., et al. (2005) The role of extruded disk material

in thoracolumbar intervertebral disk disease: a retrospective study in 40 dogs. Canadian Veterinary Medical Association 46, 814-820

Bierry, G., Kremer, S., Kellner, F., et al. (2008) Disorders of paravertebral lumbar mus-cles: from pathology to cross-sectional imaging. Skeletal Radiology 37, 967-977

Boström, A. F., Hielm-Björkman, A. K., Chang, Y.-M., et al. (2014) Comparison of cross sectional area and fat infiltration of the epaxial muscles in dogs with and without spinal cord compression. Research in Veterinary Science 97, 646-651

Brisson, B. A. (2010) Intervertebral disc disease in dogs. Veterinary Clinics of North America: Small Animal Practice 40, 829-858. https://doi.org/10.1016/j.cvsm.2010.06.001

Eminaga, S., Cherubini, G. B., Villiers, E., et al. (2013) STIR muscle hyperintensity in the cervical muscles associated with inflammatory spinal cord disease of unknown origin. Journal of Small Animal Practice 54, 137-142

Ferreira, A. J. A., Correia, J. H. D. & Jaggy, A. (2002) Thoracolumbar disc disease in 71 paraplegic dogs: influence of rate of onset and duration of clinical signs on treatment results. Journal of Small Animal Practice 43, 158-163

Freeman, M. D., Woodham, M. A. & Woodham, A. W. (2010) The role of the lumbar multifidus in chronic low back pain: a review. PM & R: The Journal of Injury, Func-tion, and Rehabilitation 2, 142-146

Fujiwara, A., Tamai, K., An, H. S., et al. (2000) The relationship between disc degeneration, facet joint osteoarthritis, and stability of the degenerative lumbar spine. Journal of Spinal Disorders 13, 444-450

Henderson, A. L., Hecht, S. & Millis, D. L. (2015) Lumbar paraspinal muscle trans-verse area and symmetry in dogs with and without degenerative lumbosacral stenosis. Journal of Small Animal Practice 56, 618-622

Holloway, A., Dennis, R., McConnell, F., et al. (2009) Magnetic resonance imag-ing features of paraspinal infection in the dog and cat. Veterinary Radiology & Ultrasound 50, 285-291

Paravertebral muscle changes in dogs with IVDE

Journal of Small Animal Practice • Vol 60 • June 2019 • © 2019 British Small Animal Veterinary Association 373

Jinkins, J. R. (2004) Acquired degenerative changes of the intervertebral seg-ments at and suprajacent to the lumbosacral junction. European Journal of Radiology 50, 134-158

Kaiser, S. M., Harms, O., Konar, M., et al. (2016) Clinical, radiographic, and mag-netic resonance imaging findings of gastrocnemius musculotendinopathy in various dog breeds. Veterinary and Comparative Orthopaedics and Traumatology 29, 515-521

Kim, S. W. & Kim, S. S. (2011) Myonecrosis of paralumbar spine muscle. Spine 36, E1162-E1165

Kjaer, P., Bendix, T., Sorensen, J. S., et al. (2007) Are MRI-defined fat infiltrations in the multifidus muscles associated with low back pain? BMC Medicine 5, 2

Lerer, A., Nykamp, S. G., Harriss, A. B., et al. (2015) MRI-based relationships between spine pathology, intervertebral disc degeneration, and muscle fatty infiltration in chondrodystrophic and non-chondrodystrophic dogs. The Spine Journal 15, 2433-2439

McKee, M. (2000) Intervertebral disc disease in the dog 1. Pathophysiology and diagnosis. In Practice 22, 355-369

Osamura, N., Takahashi, K., Endo, M., et al. (2000) Lumbar paraspinal myone-crosis after abdominal vascular surgery: a case report. Spine 25, 1852-1854

Penning, V., Platt, S. R., Dennis, R., et al. (2006) Association of spinal cord com-pression seen on magnetic resonance imaging with clinical outcome in 67 dogs with thoracolumbar intervertebral disc extrusion. Journal of Small Animal Prac-tice 47, 644-650

Platt, S. R., McConnell, F., Garosi, L. S., et al. (2006) Magnetic resonance in the diagnosis of canine inflammatory myopathies in three dogs. Veterinary Radiol-ogy & Ultrasound 47, 532-537

Robertson, I. & Thrall, D. E. (2011) Imaging dogs with suspected disc herniation: pros and cons of myelography, computed tomography, and magnetic resonance. Veterinary Radiology & Ultrasound 52Suppl 1, S81-S84

Sehl, J., Gruber, A. D., Lazzerini, K., et al. (2017) Ischaemic paraspinal myopathy along with myelopathy due to putative fibrocartilaginous embolism in a dog. Veterinary Record Case Reports 5, e000400

Sukhiani, H. R., Parent, J. M., Atilola, M. A., et al. (1996) Intervertebral disk dis-ease in dogs with signs of back pain alone: 25 cases (1986-1993). Journal of the American Veterinary Medical Association 209, 1275-1279

Trampus, P., Goepfert, C., Welle, M., et al. (2018) Magnetic resonance imaging signal alterations in paraspinal muscles in dogs with acute thoracolumbar inter-vertebral disk extrusion. Frontiers in Veterinary Science 5, 16

Supporting InformationThe following supporting information is available for this article:

Movie S1. MRI images of the thoracolumbar spine of an Eng-lish Springer Spaniel, male neutered, 7y11m. Cineloop of T2W images in sagittal plane.

Movies S2. MRI images of the thoracolumbar spine of an English Springer Spaniel, male neutered, 7y11m. Cineloop of STIR images in sagittal plane plane.