REVIEW Open Access
The ole o ca dio ascula magne ic
esonance in ako subo synd ome
Rui Plácido
1,2
, Be na do Cunha Lopes
1
, Ana G. Almeida
2
and Ca los E. Rochi e
1*
Abs ac
Tako subo synd ome (TS) is a ansien o m o le en icula dys unc ion associa ed wi h a dis inc i e
con ac ion pa e n in he absence o signi ican co ona y a e y disease igge ed by s ess ul e en s. Se e al
aspec s o i s clinical p o ile ha e been desc ibed bu i s ill emains di icul o quickly es ablish he diagnosis
a admission.
Ca dio ascula magne ic esonance (CMR) has achie ed g ea imp o emen s in he las yea s, which in u n
has made his imaging echnology mo e a ac i e in he diagnosis and e alua ion o TS. Wi h i s supe io
issue esolu ion and dynamic imaging capabili ies, CMR is cu en ly he mos use ul imaging echnique in
his se ing.
In his e iew, we p opose o comp ehensi ely de ine he ole o CMR in he e alua ion o pa ien s wi h TS
and o summa ize a se o c i e ia sui able o diagnos ic decision making in his clinical se ing.
Keywo ds: Ca dio ascula magne ic esonance, Tako subo synd ome
Backg ound
Tako subo synd ome (TS) is a clinical condi ion ha was
i s ly desc ibed in 1990 by Sa o e al.[1], ea u ing a e e s-
ible le en icula (LV) dys unc ion wi h symp oms simila
o hose o acu e co ona y synd omes ypically wi hou sig-
ni ican epica dial co ona y lesions [2]. The de ining hall-
ma k o his en i y is he eco e y o unc ion occu ing
wi hin days o weeks o he index clinical p esen a ion.
The p ecise incidence o TS is unknown, bu s udies
e ealed a p e alence o 1–2 % o pa ien s p esen ing
wi h suspec ed acu e co ona y synd omes [3].
The pa hophysiological o igin o his synd ome e-
mains elusi e and di e en e iologies a e cu en ly con-
side ed, wi h an excess o ca echolamines p ecipi a ed by
a si ua ion o s ess being o no e. Ac i a ion o he sym-
pa he ic ne ous sys em p ecipi a es ca diac ad ene gic
s imula ion and a subsequen change in con ac ili y and
elec ophysiological s a us o he myoca dium [4].
Speci ically, ca diac ad ene gic dys unc ion cha ac e izes
he acu e phase o he synd ome wi h g adual
no maliza ion du ing he ollowing weeks o mon hs.
TS has gene ally been ega ded as a ela i ely benign
disease [5]. Howe e , i is cha ac e ized by subs an ial
mo bidi y and mo ali y [6], and complica ions a e mo e
equen han p e iously hough , ocu ing in up o 50 %
o pa ien s [7–9].
Ad anced imaging modali ies a e becoming inc eas-
ingly impo an in ca dio ascula disease managemen ,
p o iding essen ial ools owa ds he diagnos ic and
p ognos ic wo k-up o pa ien s wi h TS. Excluding
signi ican co ona y s enosis is pa o i s diagnos ic al-
go i hm. This e alua ion is mos o en done in asi ely,
as he pa ien usually p esen s wi h a clinical pic u e
closely esembling ha o acu e myoca dial in a c ion,
bu could be pe o med by co ona y compu ed omog-
aphy angiog aphy in selec ed cases.
Ca dio ascula magne ic esonance (CMR) has become
a p ima y ool o non-in asi e assessmen o pa ien s wi h
TS (Fig. 1), o e ing a unique combina ion o sa e y, de-
ailed ana omical isualiza ion and issue cha ac e iza ion
da a, wi h in e -obse e consis ency and quan i a i e
accu acy.
* Co espondence: [email p o ec ed]
1
Ca dio ascula Magne ic Resonance and Compu ed Tomog aphy Sec o ,
Hea Ins i u e, InCo , Uni e si y o São Paulo Medical School, São Paulo,
B azil
Full lis o au ho in o ma ion is a ailable a he end o he a icle
© 2016 The Au ho (s). Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0
In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and
ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o
he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e
(h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed.
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68
DOI 10.1186/s12968-016-0279-5
In his e iew we p opose o explo e he u ili y o
CMR in he e alua ion o pa ien s wi h TS.
Ca dio ascula magne ic esonance
Du ing he ecen decade, CMR has become he diag-
nos ic ool o choice in e ia y ca e cen e s o pa ien s
wi h e idence o acu e nonischemic myoca dial inju y.
CMR has he unique abili y o nonin asi ely demon-
s a e myoca dial issue inju y h ough he p esence o
edema and/o delayed gadolinium con as washou in
myoca dial ex acellula ma ix, p o iding dis inc i e in-
sigh s in o he pa hogenesis and issue pa hology o se -
e al ca diac condi ions.
As expe ience and knowledge o TS has inc eased, he
p oposed Mayo diagnos ic c i e ia o he synd ome ha e
e ol ed [10]. These c i e ia highligh he essen ial ole o
CMR, no only o mo phological cha ac e iza ion bu also
o he exclusion o o he en i ies ha o he wise would no
be uled ou , pa icula ly myoca di is and myoca dial in-
a c ion wi h non-obs uc i e co ona y a e ies (MINOCA),
a he e ogeneous en i y wi h many po en ial e iologies.
Table 1 summa izes he main applica ions o CMR in
pa ien s wi h suspec ed TS. So a , a ious CMR c i e ia
ha e been used in se e al small g oups o TS as well as
in ela i ely la ge popula ion [11].
Diagnos ic a ge s and p o ocol
In a clinical se ing o e alua ion o TS, CMR appea s
as he one echnique o p o ide a comp ehensi e assess-
men , including accu a e and ep oducible measu emen
o bi en icula unc ion and olumes, he possibili y o
se ial e alua ion, assessmen o complica ions, myoca -
dial pe usion and issue cha ac e iza ion (Table 1). Ca -
diac speci ic sequences a e usually implemen ed in 1.5 T
and 3 T scanne s. Scanne s o 1 T migh ha e a ailable
ca diac sequences, and al hough limi ed, pa icula ly on
he signal- o-noise a io, could also be used i ha is he
only a ailable scanne , wi h po en ial bene i specially o
he pa ien wi h claus ophobia o la ge body size.
S eady s a e ee p ecession cines, phase con as se-
quences, black-blood T2-W iple IR, i s -pass pe usion,
ea ly gadolinium enhancemen (EGE) and la e gadolinium
enhancemen (LGE) sequences a e he mos widely used
sequences in TS. O he CMR echniques such as T1 and
T2 mapping echniques and ea u e/ issue acking could
be use ul in he nea u u e hough a p esen limi ed o
esea ch pu pose.
A p oposed s anda d CMR p o ocol in TS is depic ed
in Table 2.
Con ac ile unc ion
Cine sequences in CMR a e a c i ical componen in he
diagnosis o TS and i s a ian s. The wo d ako subo
means ‘oc opus po ’in Japanese, as he LV assumes a simi-
la shape. The ‘ ypical o m’p esen s wi h con ac ion ab-
no mali ies in he en icula wall, equally a ec ing he
an e io , in e io , and la e al walls, ha ex end beyond a
single epica dial ascula e i o y. This ‘ci cum e en ial
pa e n’can be conside ed a hallma k o TS [12]. Ano he
inding sugges i e o TS is he p esence o basal segmen s
hype kinesia, con ibu ing o he cha ac e is ic mo phology
(Figs. 2 and 3; see Addi ional iles 1 and 2). Howe e , o he
con ac ion pa e ns du ing he acu e phase o TS may be
p esen . A ecen epo sugges ed ha as many as 40 % o
Fig. 1 Recommended diagnos ic algo i hm o ako subo synd ome
Table 1 Applica ions o CMR in pa ien s wi h suspec ed TS
Quali a i e and quan i a i e assessmen o egional wall mo ion
abno mali ies.
P ecise quan i ica ion o igh and le en icula unc ion.
Tissue ca ac he iza ion, p o iding ma ke s o e e sible
(in lamma ion, edema) and i e e sible (nec osis, ib osis) inju y.
Assessmen o addi ional abno mali ies –pe ica dial e usion,
pleu al e usion and en icula h ombi.
Depic ion o mechanical complica ions.
Di e en ial diagnosis
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 2 o 12
Table 2 Summa y o CMR p o ocol o pa ien s wi h suspec ed TS
P o ocol Sequence Planes Use ulness and Cu en U iliza ion
Scou Balanced S eady S a e F ee
P ecession (bSSFP)
Non-ECG-ga ed
T ansaxial, co onal and sagi al
co e ing he en i e ho ax
S anda d o all CMR s udies
Easily iden i y pleu al and pe ica dial
e usions
Edema Black blood T2-weigh ed
( as spin echo) iple-in e sion
eco e y (IR)
Sho -axis plane co e ing he LV
Slice hickness = 8 mm
Recommended o di e en ial
dis inc ion om myoca di is o acu e
MI. Usual inding is inc ease SI in
mid-apical segmen s
T1-Mapping Modi ied Look-Locke (MOLLI),
Sho ened Modi ied Look-Locke
In e sion Reco e y (ShMOLLI),
sa u a ion eco e y single sho
acquisi ion (SASHA), o he s
Sho -axis plane co e ing he LV
wi h speci ic TI = 100–5000 ms,
collec ed using bSSFP eadou s
Slice hickness = 8 mm
Resea ch ool ha may se e as a
complemen a y echnique o T2-
weigh ed imaging. Quan i a i e means
o de ec myoca dial edema wi hou
he need o e e ence ROIs
T2-Mapping T2-p epa ed single-sho SSFP
sequence, Mul iecho FSE
(MEFSE), o he s
Ma ching sho -axis T1 Applica ion unde esea ch e alua ion.
T2 alues mo e closely co ela e wi h
ee wa e issue con en o e T1-based
echniques in suspec ed myoca dial
in lamma ion. I may o e a mo e s able
and uly quan i a i e al e na i e o
edema de ec ion in cases when
con en ional T2-weigh ed imaging ails,
specially in hin and apidly mo ing walls
Mo phology and Func ion bSSFP Sho -axis plane co e ing en i e LV
Long-axis - 3 slices each plane
(2CH, 4CH and LVOT)
Slice hickness = 6–8mm
In e slice gap = 2–4mm
Manda o y o all CMR s udies
in es iga ing TS.
I will gi e in o ma ion on he hallma k
o he disease, egional abno mal
con ac ili y no ela ed o co ona y
e i o y
Quan i a i e T acking
Techniques o Myoca dial
Mo ion and S ain
Myoca dial Tissue Tagging
(SPAMM o o he s) o a pos -
p ocessing o egula bSSFP
cine images
Sho -axis plane co e ing en i e LV
Long-axis - 3 slices each plane
(2CH, 4CH and LVOT)
Slice hickness = 6–8 mm In e slice
gap = 2–4mm
Tagged o no agged images equi e
speci ic so wa es o analysis. On bSSFP
images is a no el echnique wi h high
po en ial o ansla ing in o ou ine
clinical p ac ice allowing acking o
issue oxel mo ion o cine-CMR images
o assess myoca dial s ain, eloci ies
and displacemen . Po en ially use ul o
de ec ion o subclinical ca diac
in ol emen in TS, o p e ious TS in
he eco e y phase.
Fi s -pass pe usion Sa u a ion- eco e y imaging
wi h bSSFP eadou
Gd con as - i s -pass bolus: 0.1
mmol/kg a 4–5 mL/s
Immedia ely a e –2nd Gd bolus
o LGE: + 0.1 mmol/kg
3–6 slices acqui ed in sho axis
plane o LV
Slice hickness = 8 mm
Images a es can help on iden i ying
h ombus o p e ious ch onic myoca dial
in a c ions wi h eplacemen ib osis.
EGE <2 min a e 2nd Gd bolus
2D segmen ed IR g adien echo-
in e sion ime se a 500–550 ms
a 1,5 T (iden i y h ombus). Single-
sho o PSIR e sions can be an
al e na i e he e. Less adop ed, he
adi ional non-ga ed ee-b ea hing
T1w FSE images p e and pos Gd
bolus, wi h myoca dial SI pe se o
in ela ion o skele al muscle can
be used.
Sho -axis plane co e ing LV
(especially mid-apical segmen s)
Long-axis –1 o mo e slice each
plane (2CH, 4CH and LVOT)
Slice hickness = 8 mm
In e slice gap = 2
A su oga e o capilla y leakage and
hype emia in he myoca dium. Few
da a on he li e a u e on he indings
o hese echniques in TS.
LGE 5–10 min a e 2nd Gd bolus
2D segmen ed IR g adien echo
wi h o wi hou Phase-Sensi i e
IR (PSIR)
Single-sho o 3D e sions o LGE
can also be used.
Sho -axis plane co e ing LV
(especially mid-apical segmen s)
Long-axis –1 o mo e slice each
plane (2CH, 4CH and LVOT)
Slice hickness = 8 mm
In e slice gap = 2
The usual inding in TS is absence o
signi ican myoca dial LGE by isual
analysis. Quan i a i e analysis using
so wa es wi h a a ie y o h esholds
echniques can de ec small amoun s
o pa chy LGE.
Gd gadolinium
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 3 o 12
pa ien s ha e a mid- en icula a ian [13], wi h
mid- en icula akinesis and apical spa ing (Figs. 4
and 5; see Addi ional ile 3). Basal akinesis wi h mid-
en icula and apical spa ing also has been epo ed
[14]. No di e ences in demog aphic, clinical, angio-
g aphic, labo a o y pa ame e s, o ou come we e
ound be ween ypical e sus a ypical TS. The e a e
some cases epo ing a dynamic pa e n o wall mo-
ion abno mali y in he same pa ien [15].
These aspec s can be associa ed wi h a sys olic en-
icula dys unc ion, se e e in some pa ien s, p oducing
clinical symp oms and signs o hea ailu e.
Bo h basal hype kinesis and mid-apical akinesis can
p oduce a dynamic obs uc ion in he LV ou low ac
associa ed o no wi h sys olic an e io mo ion o he
mi al lea le s and/o unc ional mi al al e egu gi a-
ion (Fig. 6; see Addi ional ile 4). Fo an accu a e meas-
u emen o mi al egu gi a ion olume and LV ou low
ac g adien , phase con as echniques can be used.
I is known ha wall mo ion analysis wi h CMR has a
pi o al ole in clinical p ac ice. A he p esen , image
analysis is mos commonly pe o med quali a i ely.
Howe e , he diagnos ic accu acy o quali a i e assess-
men has been shown o be conside ably ope a o de-
pendan [16].
Fea u e/ issue acking CMR is a no el echnique ha
allows quan i ica ion o mo ion and s ain using s and-
a d s eady-s a e ee-p ecession cine sequences o
ou ine en icula mo pho- unc ional p o ocol. So a ,
quan i a i e da a on egional myoca dial unc ion in TS
ha e been explo ed by echoca diog aphy. Heggemann e
al. [17] desc ibed abno mal global and egional s ain
pa e ns du ing he acu e phase o pa ien s wi h TS ha
imp o ed o e ime. Fu he mo e, sub le abno mali ies
o egional LV unc ion seemed o pe sis in o he ea ly
ollow-up pe iod as sugges ed by he p esence o pos -
sys olic sho ening in mo e han hal o LV segmen s.
The au ho s concluded ha long- e m ollow-up is
needed o cla i y whe he hese abno mali ies will u -
he imp o e.
Rega ding ea u e/ issue acking CMR, u he s udies
a e undoub edly needed o de e mine i s ole in bo h
Fig. 2 Typical apical balloning in ako subo synd ome. Cine CMR 4-chambe iew (a-la e dias ole; b-la e sys ole). See Addi ional ile 1
Fig. 3 Typical apical balloning in ako subo synd ome. Cine CMR 2-chambe iew (a-la e dias ole; b-la e sys ole). See Addi ional ile 1
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 4 o 12
clinical and esea ch en i onmen s (Figs. 7 and 8; see
Addi ional iles 5, 6 and 7).
Myoca dial edema
An impo an hallma k o in lamma o y cell inju y is he
inc eased pe meabili y o cellula memb anes. Whe eas
ini ial memb ane de ec s a e o unc ional na u e, lead-
ing o Na + in lux and subsequen in acellula edema, a
mo e se e e inju y allows o a ne e lux o wa e and
ansmemb anous leakage o la ge molecules such as
oponin, e en ually leading o loss o cellula unc ions.
Tissue T2 weigh ed (T2-W) imaging is a me hod ha
assesses magne ic image con as , which is di ec ly a -
ec ed by he change in issue wa e con en , esul ing in
a high signal in ensi y o edema ous issue [18]. Black-
blood T2-W iple-IR [19] p o ide con as be ween e-
gional edema and no mal myoca dium due o he dual
supp ession o he a and lowing blood signal. When
global myoca dial in ol emen is suspec ed, a quan i a-
i e app oach should be p e e ed, classically by calcula -
ing he SI a io be ween myoca dium and skele al
muscle −T2 SI a io −and a cu o alue o ≥1.9 is
conside ed signi ican , o by newe and p omising
pa ame ic mapping echniques, especially T2 mapping.
In TS one o he mos cha ac e is ic inding is myoca -
dial edema, a di ec esul o he inju y p ocess. I as-
sumes a di use o ansmu al dis ibu ion in bo h apical
and mid planes, a loca ion ha is no es ic ed o a pa -
icula ascula e i o y. Those a eas o edema ypically
ma ch he dys unc ional en icula con ac ion a ea ob-
se ed wi h cine CMR sequences, which a e usually glo-
bally ci cum e en ial, bu es ic ed o a po ion o he
LV, such as in ol ing all segmen s o apical po ion, o al-
e na i ely mid o basal po ions (Figs. 9 and 10).
These ea u es can be used o di e en ia e TS om
acu e myoca dial in a c ion, in which edema usually has
a co ona y dis ibu ion on he LV wall.
In acu e myoca di is pa ien s, T2-W iple IR sequence
shows high signal in ensi y o en wi h a mul i ocal and
he e ogeneous dis ibu ion in egions wi h LGE, ha
could be ansmu al, bu equen ly wi h highe signal
in ensi y on mid-wall o subepica dial myoca dial laye .
Ano he indica o o he in lamma o y p ocesses in TS
is he sligh associa ion be ween edema and pe ica dial
e usion [20] (Figs. 11 and 12).
Fig. 4 Mid- en icula a ian o ako subo synd ome. Cine CMR 4-chambe iew (a-la e dias ole; b-la e sys ole). See Addi ional ile 3
Fig. 5 Mid- en icula a ian o ako subo synd ome. Cine CMR 2-chambe iew (a-la e dias ole; b-la e sys ole). See Addi ional ile 3
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 5 o 12

Ini ial T2 hype in ensi y educes o e he ollowing
weeks, whe eas T2 hype in ensi y o segmen s a ec ed
by myoca dial in a c is main ained o mo e han 2 o
3 mon hs a e symp om onse [21, 22].
To no e, e en hough T2-W iple IR is no mally used
in CMR o de ec edema, some au ho s ha e sugges ed
he use o T2 mapping echnique o diagnose i , which
seems o be mo e insensi i e o mo ion a i ac s [23].
Despi e inc easing e idence on he use o T2-W iple
IR, i is a ecognized mo ion and blood pool a i ac
p one sequence. Pa ame ic mapping echniques, espe-
cially T2 mapping, a e eme ging as a mo e obus
al e na i e o edema app ecia ion and quan i ica ion in
di e se clinical scena ios. Fu u e esea ch explo ing he
ole o T2 mapping in TS is equi ed.
Myoca dial pe usion
In mos cases, LV segmen s show no pe usion de ec s
and indings on hese echniques a e usually no mal in
pa ien s wi h TS. Howe e , in some pa ien s CMR
i s -pass pe usion s udies may appea no mal a
basal segmen s bu demons a e subendoca dial pe u-
sion de ec s, especially mo e apically [24]. Acco ding
o he mos accep ed mechanism o TS, which p oposes a
igo ous neu ohumo al discha ge p ecipi a ed by emo-
ional s ess, a se e e mic o ascula dys unc ion could be
expec ed, leading o seconda y myoca dial s unning.
Th ombus assessmen
EGE sequence, ideally pe o med wi hin 2 min a e
con as agen in usion, is op imal o iden i ica ion o
adhe en endoca dial h ombus on mid-apical seg-
men s wi h se e e hypokinesis o akinesis. Th ombus
is ecognized as a low signal in ensi y (no gadolinium
up ake), nea ly black, ha makes con as wi h he
in e media e signal o myoca dium and blood pool,
when de in e sion ime o he IR g adien echo se-
quence is se o 500–550 ms.
In some cases, LV h ombus can be de ec ed by cine
CMR (Fig. 13), al hough hese sequences a e less sensi-
i e o his pu pose.
Myoca dial nec osis and ib osis
LGE depends on di e ences in he olume o dis i-
bu ion o gadolinium in no mal and pa hologically al-
e ed myoca dium [25] and se e al mo phological
changes wi hin he issue may con ibu e o a la ge
olume o dis ibu ion.
Fig. 6 Cine CMR in ho izon al long-axis iew demons a ing he
sys olic “je ”in le en icula ou low ac (whi e a ow)wi hassocia ed
sys olic an e io mo ion o he mi al lea le s (g een a ow) and unc ional
mi al al e egu gi a ion ( ed a ow). This dynamic obs uc ion
can de elop as a esul o dyskine ic apical and mid en icula segmen s
wi h hype dynamic unc ion o basal segmen s. See Addi ional ile 4
Fig. 7 2-D longi udinal s ain analysis by issue- acking CMR (So wa e CVI42 Ve sion 5, Ci cle Ca dio ascula Imaging Inc., Calga y, Canada)
a end-sys ole in ho izon al long-axis iew (a), 4-chambe iew and 2-chambe iew (c)inapa ien wi h he‘classical o m’o ako subo
synd ome. Acco ding o he colou -scale (a he le on each panel), ed ep esen s he mos posi i e longi udinal s ain and da k-blue
ep esen s he mos nega i e longi udinal s ain (no mal sys olic longi udinal s ain is nega i e ep esen ing sho ening). A end-sys ole,
mid-apical balloning is ep esen ed on ed wi h a ‘ci cum e en ial pa e n’. See Addi ional iles 5, 6 and 7
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 6 o 12
In gene al, LGE indica es he ela i e changes o he
ex acellula and in acellula olumes. The e o e, any
changes o he in e s i ium, such as edema o ib osis,
inc ease he olume o dis ibu ion causing LGE [26].
Fu he mo e, myoca dial nec osis causes in acellula
accumula ion o gadolinium, which con ibu es g ea ly
o an expansion o he olume o dis ibu ion and LGE.
None heless, in edema ous iable myoca dium he a io
o he ex acellula and in acellula olumes is no sub-
s an ially al e ed [27]. The e o e, he pe i-in a c zone
in ol ing he edema ous iable myoca dium bu no he
in a c ed non- iable myoca dium, does no exhibi sig-
ni ican LGE on CMR.
The absence o LGE in TS pa ien s as documen ed by
CMR s udies has been desc ibed in many cases and is a
Fig. 8 Global longi udinal (a) and adial (b) s ain cu es o he le en icule by issue- acking CMR (So wa e CVI42 Ve sion 5, Ci cle Ca dio as-
cula Imaging Inc., Calga y, Canada)
Fig. 9 Myoca dial edema. T2-weigh ed iple-in e sion eco e y 2-
chambe iew showing ansmu al signal hype in ensi y in he
mid-apical segmen s o le en icle
Fig. 10 Myoca dial edema. T2-weigh ed iple-in e sion eco e y
sho -axis iew showing ansmu al signal hype in ensi y in he
mid-apical segmen s o le en icle
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 7 o 12
common diagnos ic c i e ion in mos CMR cen e s.
Howe e , se e al s udies challenged his no ion by
epo ing delayed hype enhancemen in TS pa ien s
[28, 29] (Fig. 14). Rol e al. [30] desc ibed ha he
LGE signal in ensi y ound in pa ien s wi h TS wi hin
24 h o admission was lowe han ha usually
documen ed in cases o myoca dial in a c ion o myo-
ca di is and was no longe p esen on ollow-up CMR
2 weeks la e . Also he ex en o LGE as ep esen ed
by he ela i e mass o enhancing issue is less han
usually epo ed in s udies o myoca dial in a c ion
[31]. Mo e ecen ly, in a la ge mul icen e coho
s udy, Ei el e al. [32] showed ha LGE was de ec ed
in only 9 % o pa ien s when using a h eshold o 3
s anda d de ia ion (SD) abo e he mean o emo e
myoca dium o de ine signi ican enhancemen . None
o hei pa ien s had e idence o LGE when using a
h eshold o 5 SD, which is usually p oposed as he
cu o o ib osis de ec ion in myoca di is and acu e
myoca dial in a c ion. This poin s owa d he ac ha
using di e en h eshold o LGE de ec ion in di e -
en s udies may be esponsible o he di e en p e a-
lence o LGE-posi i e pa ien s. The au ho s concluded
ha he absence o signi ican LGE (>5 SD) combined
wi h myoca dial edema and ma ked LV ballooning is
aunique ea u eo TS.
No wi hs anding wi h his, a pa hophysiological ex-
plana ion o LGE in TS has been in es iga ed. I is
known ha he le el o ex acellula ma ix p o eins
is ele a ed and he amoun o collagen-1 and ib o-
nec in is signi ican ly aised in he phase o se e e
con ac ile impai men , ollowed by no maliza ion in
unc ional eco e y phase [33]. In he s udy by Rol e
al. [30] pa ien s also unde wen LV endomyoca dial
biopsy in he acu e phase and a e eco e y, o
Fig. 11 Pe ica dial e usion (a ows) in a pa ien wi h ako subo
synd ome. Cine CMR 4-chambe iew (la e sys ole)
Fig. 12 Pe ica dial and pleu al e usions in a pa ien wi h ako subo synd ome. Cine CMR axial iew a he le el o he pulmona y al e
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 8 o 12
de e mine he amoun o ex acellula ma ix. All
pa ien s wi h TS showed a signi ican inc ease o
collagen-1 compa ed wi h con ol issue. Mo eo e ,
he amoun o collagen-1 was signi ican ly highe in
LGE posi i e pa ien s [de ined as an a ea o signal in-
ensi y ha was 2 SD o g ea e han he signal in en-
si y o nonenhanced myoca dium]. The la e g oup
showed no inc eased le els o ca diac bioma ke s,
leading o he hypo hesis ha LGE canno be a ib-
u ed o di e ences in sa colemmal in eg i y. Tha
inding was suppo ed by he ac ha signs o onco-
ic cell dea h unde elec on mic oscopy we e no
p esen on issue samples. Also, he e we e no di e -
ences in he ex en o edema on T2 weigh ed mea-
su emen s be ween g oups. The e o e, e idence ha
LGE in TS is caused by ei he nec osis o edema is
lacking, being plausible a p ocess o ansien ib osis
(o di use eac i e ib osis).
Di e en ly, Sacha e al. [34] and Ma échaux e al.
[35] epo ed cases o con ac ion-band nec osis de-
ec ed a au opsy, a possible explana ion o LGE in
pa ien s wi h TS.
Some au ho s in es iga ed he clinical implica ions
o LGE (using di e en h esholds) in TS. Ei el e al.
[32] showed no ela ion be ween he occu ence o
LGE and clinical p esen a ion, mo ali y, age, sex,
ECG pa e n, o ype o s ess igge in 158 pa ien s
wi h TS. Na use e al. [36] epo ed, in a coho o
20 pa ien s wi h TS, no signi ican ly di e ences in
age, sex, conges i e hea ailu e, peak c ea ine
kinase, no ad enaline ele a ion, ECG abno mali ies,
ejec ion ac ion, ype o TS and ime om onse o
symp oms o CMR in he LGE posi i e g oup. How-
e e , he e was a g ea e p e alence o ca diogenic
shock (38 s 0 %, p= 0,049) and longe ime o
no malize elec o and echoca diog aphic changes.
They also epo ed a posi i e co ela ion be ween he
LGE a ea a io ( olume o enhanced issue/ o al ol-
ume o LV myoca dium) in he sub-acu e phase (2–7
days a e p esen a ion) and he du a ion o no maliza ion
o bo h ECG ( = 0.738, p= 0.037) and echoca dio-
g am ( =0.762, p= 0.028).
The u ili y o LGE is undeniable in he se ing o
ocal myoca dial ib osis and nec osis, whe e di e en
signal in ensi y h eshold me hods a e used o de ec
enhancemen . Howe e , when myoca dial inju y is
sub le and he e a e no con iden ly ecognizable
e e ence egions o no mal myoca dium, ib osis
may go unde ec ed on con en ional LGE h eshold
me hods. In con as , T1 mapping sequences p o ide
a di e en app oach o issue cha ac e iza ion, pe -
mi ing he measu e o he in insic elaxa ion imes
o myoca dium p e and pos gadolinium in usion, no
dependen on ela i e signal in ensi ies. The e o e, i
has he po en ial o imp o e he sensi i i y o CMR
issue cha ac e iza ion, especially when di use and
sub le ib osis is he possible mechanism, like in TS.
Righ en icula in ol emen in TS
An impo an issue ha has ecen ly been seen wi h
he use o CMR and i s cine sequences is a signi i-
can pe cen age o pa ien s (up o one- hi d) wi h
igh en icula in ol emen , which in u n was as-
socia ed wi h longe hospi aliza ion, wo se ma ke s
o hea ailu e, pa icula ly pleu al e usion (Fig. 12)
and olde age. Consequen ly, bi en icula ballooning
Fig. 14 La e gadolinium enhancemen CMR sho -axis iew depic ing
pa chy highe signal in ensi y in he mid-apical segmen s o le
en icle, in a pa ien wi h ako subo synd ome wi hin 48 h o
admission. These indings p obably e lec a p ocess o di use
eac i e ib osis
Fig. 13 Apical h ombus (dashed ci cle), one o he complica ions o
ako subo synd ome. Cine CMR ho izon al long-axis iew
Plácido e al. Jou nal o Ca dio ascula Magne ic Resonance (2016) 18:68 Page 9 o 12