Re iew
na u e publishing g oup
S a ins, 3-hyd oxy-3-me hylglu a yl coenzyme A educ ase
inhibi o s, a e e ec i e in lowe ing blood choles e ol as well
as in p ima y and seconda y p e en ion o ca dio ascula
e en s. App oxima ely 25 million people wo ldwide a e ak-
ing s a ins.1–3 S a ins a e sa e and e ec i e in he majo i y o
pa ien s, bu hey a e associa ed wi h muscle oxici y, which,
al hough a e, can be se ious and po en ially li e h ea ening.4,5
The clinical spec um o s a in-induced myo oxici y a ies
g ea ly om asymp oma ic ele a ions o c ea ine kinase (CK)
wi hou muscle pain, o muscle pain o weakness wi h aised CK
le els, myosi is wi h biopsy-p o en muscle in lamma ion, and,
inally, habdomyolysis wi h muscle symp oms, high CK, and
po en ial o acu e kidney inju y (Table 1).
6,7
Whe he s a ins
cause muscle pain in he absence o CK changes emains con-
o e sial. Al hough milde o ms o myo oxici y end o be sel -
limi ing and disappea a e cessa ion o he apy, hey can lead
o poo quali y o li e, poo d ug compliance, and, consequen ly,
he ailu e o p e en ca dio ascula e en s.
Because s a in-induced myo oxici y is uncommon, i is neces-
sa y o pool and analyze da a om a ious sou ces, including
mul icen e clinical ials and obse a ional s udies, o s udy
gene ic e iology. In addi ion, elec onic medical eco ds linked
o biological eposi o ies o o indi idual pa ien s ha e been
use ul in iden i ying and ec ui ing pa ien s wi h d ug-induced
oxici ies.6,8–12 The Pheno ype S anda diza ion P ojec was
s a ed a ew yea s ago o acili a e such mul icen e esea ch
collabo a ions on a ious ypes o se ious ad e se d ug eac-
ions (ADRs). Pheno ype consensus pape s on d ug-induced
li e inju y, skin inju y, and o sade de poin es ha e been pub-
lished by mul idisciplina y g oups o in e na ional expe s.
13–15
The aim o his a icle is o p o ide consensus de ini ions o
s a in-induced myo oxici y pheno ypes, o acili a e c oss-s udy
compa isons om he exis ing coho s, o aid in he ec ui -
men o e ospec i e and newly diagnosed pa ien s wi h
s a in-induced muscle damage, and o de ine he pheno ype
o genomic da a analyses.
We con ened an in e na ional expe wo kshop on s a in-
induced myo oxici y in Decembe 2013 in Li e pool, UK, o
ag ee on de ini ions and a minimum se o c i e ia o help in
iden i ica ion and ec ui men . We epo on he delibe a ions o
his wo kshop: i s , we summa ize e idence o he clinical and
biochemical pheno ypes ha ha e been epo ed, and second,
we epo on ou sugges ed s anda diza ion o he e minology
and pheno ypes o s a in-induced muscle oxici y.
Recei ed 25 Ma ch 2014; accep ed 27 May 2014; ad ance online publica ion 2 July 2014. doi:10.1038/clp .2014.121
Clinical Pha macology & The apeu ics
10.1038/clp .2014.121
Re iew
2July2014
96
4
25Ma ch2014
27May2014
S a ins a e widely used lipid-lowe ing d ugs ha a e e ec i e in educing ca dio ascula disease isk. Al hough hey a e
gene ally well ole a ed, hey can cause muscle oxici y, which can lead o se e e habdomyolysis. Resea ch in his a ea
has been hampe ed o some ex en by he lack o s anda dized nomencla u e and pheno ypic de ini ions. We ha e used
nume ical and desc ip i e classi ica ions and de eloped an algo i hm o de ine s a in- ela ed myo oxici y pheno ypes,
including myalgia, myopa hy, habdomyolysis, and nec o izing au oimmune myopa hy.
1Depa men o Molecula and Clinical Pha macology, TheWol son Cen e o Pe sonalised Medicine, Ins i u e o T ansla ional Medicine, Uni e si y o Li e pool,
Li e pool, UK; 2Depa men o Clinical Biochemis y, Newcas le upon Tyne Hospi als NHS Founda ion T us , Royal Vic o ia In i ma y, Newcas le upon Tyne, UK;
3Clinical T ial Se ice Uni , Ox o d, UK; 4Cen e o Musculoskele al Resea ch/NIHR Manches e Musculoskele al Biomedical Resea ch Uni , Uni e si y o Manches e ,
Manches e , UK; 5MRC/ARUK Ins i u e o Ageing and Ch onic Disease, Facul y o Heal h & Li e Sciences, Uni e si y o Li e pool, Li e pool, UK; 6Zo a Biosciences L d,
Tieo ie 2, Espoo, Finland; 7The Medical P oduc s Agency, Uppsala, Sweden; 8Depa men o Medical Sciences, Clinical Pha macology and Science o Li e Labo a o y,
Uppsala Uni e si y, Uppsala, Sweden; 9Di ision o Pha macoepidemiology and Clinical Pha macology, U ech Uni e si y, Facul y o Science, U ech Ins i u e o
Pha maceu ical Sciences, U ech , The Ne he lands; 10Duke Ins i u e o Genome Sciences and Policy, Du ham, No h Ca olina,USA; 11Ca dio ascula Heal h Resea ch
Uni , Depa men s o Medicine, Epidemiology and Heal h Se ices, Uni e si y o Washing on, Sea le, Washing on, USA; 12G oup Heal h Resea ch Ins i u e, G oup
Heal h Coope a i e, Sea le, Washing on, USA; 13Medical Resea ch Ins i u e, Ninewells Hospi al and Medical School, Dundee, UK. Co espondence: A Al i e ic
(Ana.Al i [email p o ec ed].uk)
Pheno ype S anda diza ion o S a in-Induced
Myo oxici y
AAl i e ic1, DNeely2, JA mi age3, HChinoy4, RGCoope 5, RLaaksonen6, DFCa 1, KMBloch1,
JFahy1, AHanson1, Q-YYue7, MWadelius8, AHMai land- an De Zee9, DVoo a10, BMPsa y11,12,
CNAPalme 13 and MPi mohamed1
Open
see REVIEW page 477
470 VOLUME 96 NUMBER 4 | OCTOBER 2014 | www.na u e.com/cp
Re iew
CONSENSUS PROCESS
The PREDICTION-ADR conso ium, unded by he EU
Se en h F amewo k P og amme, o ganized a join mee ing on
10 Decembe 2013 in Li e pool wi h a mul idisciplina y eam
o expe s on s a in- ela ed myopa hy and angio ensin con e -
ing enzyme-inhibi o angioedema. An in e na ional eam wi h
known expe ise in he a ea was assembled by in i a ion. The
g oup comp ised clinical and basic pha macologis s, in e nis s,
heuma ology and myopa hy expe s, immunology and clinical
chemis y scien is s, alle gis s, egula o y agency ep esen a-
i es, and manage s o elec onic medical eco d da abases.
A d a manusc ip ci cula ed be o e he mee ing con ained a
b ie li e a u e e iew on s a in- ela ed myopa hy and a desc ip-
ion o main pheno ypes. The expe g oup was asked o con-
side he minimum se o c i e ia o pa ien ec ui men . An
in oduc o y alk was gi en o s a he discussions. In addi ion,
an algo i hm was designed o aid s anda diza ion o nomencla-
u e and pheno ypes, and ecommenda ions we e made abou
he ypes o da a o be collec ed om each pa ien (Figu e 1).
A e he mee ing, he e ised manusc ip comp ising he c i-
e ia (Table 1) and algo i hm was app o ed by all con ibu o s.
We did no discuss causali y assessmen o mally, as he chal-
lenges o his ask a e common o many ADRs.16 They equi e
es ablishing a empo al ela ionship, along wi h dechallenge and
echallenge in o ma ion, and, pa icula ly impo an o s a in-
ela ed myo oxici y, igilan in o ma ion on muscle inju y om
alls, auma, o igo ous exe cise. We also ecommend inde-
penden assessmen o causali y by an adjudica ion panel, using
me hodology ha has been success ully applied o e ospec i e
and p ospec i e ec ui men o pa ien s wi h a e ADRs, includ-
ing se e e d ug-induced skin inju y.15
INCIDENCE
The a e o s a in-induced myo oxici y has been es ima ed om
ad e se e en da a in case se ies and andomized con olled
ials.17–24 Howe e , es ima ing ad e se e en s om andomized
con olled ials may be challenging because (i) hey a e no
usually powe ed o cap u e low- equency ADRs; (ii) pa ien s
a e exposed o d ugs and moni o ed only o a sho pe iod o
ime when myo oxici y may no ye ha e de eloped; (iii) he diag-
nos ic c i e ia o eac ion e en s may a y ac oss di e en ials;
and (i ) hey may ha e s ic inclusion and exclusion c i e ia ha
may exclude some popula ions who a e a highe isk o ADRs.
The isk o habdomyolysis among hospi alized pa ien s ecei -
ing lipid-lowe ing d ugs has ecen ly been es ima ed in a eal-
wo ld clinical se ing. Claims da a om 9 million membe s o
i e US heal h plans we e used o con i m 42 cases in >470,000
pa ien s exposed o lipid-lowe ing d ugs. The isk o habdomy-
olysis o indi iduals on di e en s a in p epa a ions, including
combina ion he apy, and he isk o como bidi ies we e es ima ed
o be be ween 0.3 and 8.4 in 10,000 pa ien -yea s.25 The au ho s
es ima ed ha he isk in compa ison wi h a o as a in as a e e -
ence was signi ican ly highe when s a ins we e used in combina-
ion wi h cy och ome P450 (CYP)3A4 inhibi o s (odds a io: 7.1,
95% con idence in e al: 1.6–31.6) and o ce i as a in mono-
he apy (odds a io: 4.7, 95% con idence in e al: 1.1–21.1).25
DIAGNOSTIC CRITERIA
The diagnosis o s a in-induced myo oxici y is based on medi-
cal his o y, clinical examina ion, and labo a o y es s and can
be con i med by muscle biopsy (Supplemen a y Table S1
online). Muscle biopsies can e eal muscle ibe nec osis, ype
II ibe a ophy, and inc eased lipid s o es in muscle ibe s o
in lamma ion.
26–29
In clinical p ac ice, a p agma ic app oach is
adop ed: discon inua ion o he culp i d ug and a oidance o
i s u u e use.30 Moni o ing o s a in he apy o muscle oxici y
includes CK measu emen s, al hough ou ine labo a o y es ing
is ecommended only o symp oma ic pa ien s.31 The po en ial
ha m o in oducing ou ine CK moni o ing in all pa ien s who
ake s a ins may ou weigh he bene i s om se e al pe spec-
i es, including alse-posi i e esul s wi h po en ially ensuing
in asi e in es iga ions, a psychological e ec on pa ien s ha
may esul in educ ion o s a in use, and an inc eased cos o
he heal h-ca e sys em. A ecen s udy on pa ien s’ pe cep ions
Table 1 S a in- ela ed myo oxici y pheno ype classi ica ion
SRM classi ica ion Pheno ype Incidence De ini ion Re e ence
SRM 0 CK ele a ion <4× ULN 1.5–26% No muscle symp oms Re s. 1,20,34,35,67
SRM 1 Myalgia, ole able 190/100,000
Pa ien -yea s; 0.3–33%
Muscle symp oms wi hou CK ele a ion Re s. 1,19,21,50,68
SRM 2 Myalgia, in ole able 0.2–2/1,000 Muscle symp oms, CK <4× ULN, comple e
esolu ion on dechallenge
Re . 20
SRM 3 Myopa hy 5/100,000
Pa ien -yea s
CK ele a ion >4× ULN <10× ULN ± muscle
symp oms, comple e esolu ion on dechallenge
Re . 1
SRM 4 Se e e myopa hy 0.11% CK ele a ion >10× ULN <50× ULN, muscle
symp oms, comple e esolu ion on dechallenge
Re s. 20,69
SRM 5 Rhabdomyolysis 0.1–8.4/100,000
Pa ien -yea s
CK ele a ion >10× ULN wi h e idence o enal
impai men + muscle symp oms o CK >50× ULN
Re s. 4,6,25,44,45
SRM 6 Au oimmune-media ed
nec o izing myosi is
~2/million pe yea HMGCR an ibodies, HMGCR exp ession in muscle
biopsy, incomple e esolu ion on dechallenge
Re s. 51,70
Nume ic classi ica ion was de eloped by an expe g oup; desc ip i e nomencla u e was adap ed om he ecommenda ion o he Ame ican College o Ca diology/Ame ican
Hea Associa ion/Na ional Hea , Lung, and Blood Ins i u e Clinical Ad iso y Boa d.3,7
CK, c ea ine kinase; HMGCR, 3-hyd oxy-3-me hylglu a yl-coenzyme A educ ase; SRM, s a in- ela ed myo oxici y; ULN, uppe limi o no mal.
ClInICAl PhARMACology & TheRAPeuTICS | VOLUME 96 NUMBER 4 | OCTOBER 2014 471
Re iew
o s a in he apy has demons a ed educed adhe ence o s a in
he apy in hose conce ned abou ADRs.32,33
CLINICAL PRESENTATIONS OF STATIN-INDUCED MUSCLE
TOXICITY
S a in-induced muscle oxici y can p esen in many di e en
ways, bu he ecognized pheno ypes and deg ees o se e i y
a e classi ied in o se en di e en ca ego ies, as ep esen ed in
Table1 (s a in- ela ed myo oxici y (SRM) 0–SRM 6).
Muscle symp oms
S a in-induced muscle oxici y may p esen wi h a wide a ie y
o symp oms, including a igue, muscle pain, muscle weakness,
muscle ende ness, and c amps. These symp oms a e usually
p oximal and symme ical bu may be gene alized. They can be
agg a a ed by igo ous exe cise and some imes by addi ion o a
new medica ion.2 Pain ole ance a ies g ea ly in di e en indi-
iduals, and many s udies ha e elied on sel - epo ing o muscle
symp oms. I pa ien s can ole a e mild muscle pain, s a ins a e
no usually discon inued (SRM 1, Table 1).
Plasma CK ele a ion
Asymp oma ic se um CK ele a ions (SRM 0) and muscle pain
wi hou an inc ease in CK (SRM 1) ha e been he wo mos
commonly (occu ing in up o 33% o pa ien s) desc ibed
ea u es o s a in-induced oxici y.20,34,35 Wi h he inc easing
use o elec onic medical eco ds, i may be possible o use CK
ele a ions o iden i y pa ien s wi h s a in myo oxici y. Howe e ,
because he use o CK as a sc eening es o muscle inju y is no
ecommended, he iden i ica ion o an isola ed CK ele a ion in
an elec onic (o pape -based) eco d may indica e suspicion
o he clinician o muscle oxici y, e en i symp oms a e a i-
ably eco ded. CK le els a e o en used as a c ude es ima e o
se e i y, bu he co ela ion be ween muscle symp oms and CK
le els is impe ec , and he clinical in e p e a ion o CK le els
is complex.36 The e is conside able a iabili y in he inclusion
c i e ia and CK le els in he li e a u e, pa icula ly in gene ic
suscep ibili y s udies. Some au ho s in es iga e pa ien s wi h
sel - epo ed myalgia and CK le els om 1 o 3× he uppe limi
o no mal (ULN) (SRM 2), whe eas o he s apply mo e s ingen
c i e ia wi h CK ele a ions >4× ( e . 1) (SRM 3) o >10× he
ULN (SRM 4) o myopa hy and ≥50× he ULN o habdo-
myolysis (SRM5),9,35,37–42 which a e he c i e ia used in some
ecen indus y- unded s udies. To p e en inclusion o pa ien s
wi h CK ele a ions om causes o he han s a in myo oxici y,
we ha e adop ed he >4× he ULN o myopa hy (SRM 3) and
>10× he ULN o se e e myopa hy (SRM 4). Al hough hese
cu o poin s ha e been commonly used in clinical ials and in
se e al guidelines, hey a e somewha a bi a y. Howe e , u u e
analyses based on hese cu o poin s (<4× he ULN, 4–10× he
ULN, and >10× he ULN) will help in de ining he sensi i i y
and speci ici y o di e en gene ic ma ke s, and p o ide a i s
s ep in u u e e inemen o cu o le els. Some in es iga o s ha e
also used alanine amino ans e ase ele a ion o iden i y mus-
cle inju y;35 his may be o use when measu ed in combina ion
wi h CK. Isola ed ele a ion o alanine amino ans e ase, in he
Figu e 1 Algo i hm o de ining he ype o s a in- ela ed myo oxici y. New nomencla u e was in oduced o e lec he pheno ype classi ica ion and se e i y
(SRM 0 o SRM 6) o s a in- ela ed oxici y. CK, c ea ine kinase; HMGCR, 3-hyd oxy-3-me hylglu a yl-coenzyme A educ ase; LLD, lipid-lowe ing d ug; SRM, s a in-
ela ed myo oxici y; ULN, uppe limi o no mal.
Check enal unc ion
u ine myoglobin
Rhabdomyolysis
No
neg
pos
Symp oms esol e
a e s a in -
wi hd awal?
No S a in
discon inua ion?
Yes
Nec o izing au oimmune
myopa hy SRM6
Yes
Nons a in
myopa hy
Pa ien ea ed wi h s a ins (>2 weeks)
CK >4X
ULN <10X ULN
± symp oms = SRM3
CK >10X
ULN <50X ULN
± symp oms = SRM4
CK >10X ULN
symp oms o CK <50 ULN
SRM5
Myalgia Myopa hy
A e he muscle- ela ed
symp oms ole able?
S a in he apy wi h clinical and
CK moni o ing
S a in challenge
o al e na i e LLD
Muscle biopsy
i a ailable
Se um an i-HMGCR
an ibody es
Yes No
Muscle pain and weakness un ela ed o auma
No muscle symp oms, CK ele a ed
Ele a ed CK <4X ULN
No symp oms = SRM0
Ele a ed CK <4X ULN
+ symp oms = SRM2
No mal CK
+ symp oms = SRM1
472 VOLUME 96 NUMBER 4 | OCTOBER 2014 | www.na u e.com/cp
Re iew
absence o an inc ease in CK le els, should aise he suspicion o
li e inju y, which can also a ely occu wi h s a ins.43
Rhabdomyolysis
Rhabdomyolysis (SRM 5), he mos se ious ADR associa ed
wi h s a ins,4,6,25,44,45 is cha ac e ized by muscle nec osis,
elease o myoglobin in o he bloods eam, and some imes
acu e enal ailu e.4,6,26,45,46 Muscle symp oms a e accom-
panied by ma ked CK ele a ion, ypically g ea e han 50×
he ULN. Pigmen neph opa hy wi h b own u ine is ypically
e iden due o myoglobinu ia and is consis en wi h se um
c ea inine ele a ion.7
HMGCR au oan ibodies
A numbe o s udies (e.g., e .46) ha e epo ed he occu -
ence o an au oimmune-media ed nec o izing myopa-
hy (SRM 6) a e s a in exposu e, wi h symp oms ha
con inued a e d ug wi hd awal. Se um au oan ibodies
agains 3-hyd oxy-3-me hylglu a yl-coenzyme A educ ase
(HMGCR), he pha macological a ge o s a ins, ha e been
epo ed in pa ien s wi h s a in-induced au oimmune myo-
pa hy.
47
Muscle biopsies can also be diagnos ically use ul o
he de ec ion o HMGCR exp ession on he cell su ace o
egene a ing muscle ibe s.47 In 51 pa ien s wi h sel -lim-
i ed s a in in ole ance, an i-HMGCR an ibodies we e no
obse ed in a single indi idual.48 This sugges s ha sel -
limi ed s a in myopa hy has a dis inc biological mechanism
om s a in-induced au oimmune myopa hy. In addi ion,
a o as a in and sim as a in p omo e a p oin lamma o y
o Th1 esponse in ac i a ed pe iphe al blood mononuclea
cells by inc easing he numbe o in e e on-γ-sec e ing T
cells.49 I should be no ed, howe e , ha many au oimmune
myopa hy cases a e likely o be excluded om s udies because
wi hd awal o he s a ins does no alle ia e symp oms and
Table 2 Repo ed isk ac o s o de eloping s a in-induced myopa hy
Risk ac o Desc ip ion Re e ence
Pa ien ac o s
Ad anced age >80 yea s) Re . 7
Female gende
Re . 42
Low body mass
index
Como bidi ies Un ea ed hypo hy oidism Re . 50, Supplemen a y
e s. 71 and 72
Low i amin D le el Supplemen a y e . 73
online
Ch onic enal insu iciency (GFR 60 ml/min), especially when associa ed wi h diabe es Re . 7,25
In ec ion, li e impai men , hype ension Re . 25
Alcohol abuse Re . 7
Physical exe cise Supplemen a y e s.
74–76
Su ge y The Ame ican Hea Associa ion ecommends empo a y cessa ion o s a ins be o e majo su ge y Re . 7
His o y o s a in
myopa hy
Pe sonal o amily his o y o s a in myopa hy Re . 50
D ug ac o s
Highe s a in dose Inc eased equency o myopa hy Supplemen a y e s. 77
and 78
In e ac ing d ugsaCYP3A4 enzyme inhibi o s (pa icula ly impo an o CYP3A4 subs a es sim as a in, lo as a in,
and a o as a in): dil iazem, e apamil, cla i h omycin, eli h omycin, e y h omycin, i aconazole,
cyclospo ine, p o ease inhibi o s ( i ona i , indina i , and saquina i ), amioda one, and usidic acid
Re s. 59–62,
Supplemen a y e s.
79–81
CYP2C9 enzyme inhibi o s (wi h e ec s on lu as a in, a CYP2C9 subs a e): omep azole
and luconazole
Supplemen a y e s.
82–84
OATP1B1 inhibi ion (wi h e ec s on sim as a in, p a as a in, lo as a in, and osu as a in):
gem ib ozilb
Re s. 53–57
Supplemen a y e s. 85
and 86
Die - ela ed
in e ac ions
G ape ui juice (>200 ml daily) inc eases le els o sim as a in, a o as a in, and lo as a in
Supplemen a y e s. 71–87 a e p o ided wi h Supplemen a y Table S1 online.
CYP, cy och ome P450; GFR, glome ula il a ion a e; OATP, o ganic anion- anspo ing polypep ide.
aOnly a selec ion o d ug in e ac ions wi h s a ins has been p esen ed. Reade s should e e o mo e ex ensi e e iews o a ulle lis o in e ac ing d ugs (Kellick e al.,
2014 (Supplemen a y e . 87), h p://www.mh a.go .uk/Sa e yin o ma ion/D ugSa e yUpda e/D ugSa e yUpda esea ch esul s/index.h m, and h p://pha macis sle e .
he apeu ic esea ch.com/pl/A icleDD.aspx?nidchk=1&cs=&s=PL&p =2& p =31&dd=280606&pb=PL&ca =4803&segmen =4421&AspxAu oDe ec CookieSuppo =1).bThe
main mechanism o he in e ac ion is men ioned in he able, bu i is impo an o no e ha in some cases he e may be mul iple mechanisms. Fo example, wi h gem ib ozil, in
addi ion o OATP1B1 inhibi ion, CYP2C8 and UGT1 inhibi ion may also con ibu e. UGT, u idine diphospha e glucu onyl ans e ase.
ClInICAl PhARMACology & TheRAPeuTICS | VOLUME 96 NUMBER 4 | OCTOBER 2014 473
Re iew
hus s a ins may be discoun ed as he causal agen s o myo-
pa hy pa hogenesis.
TIME TO ONSET
In he PRIMO (P edic ion o Muscula Risk in Obse a ional
Condi ions) s udy, he median ime o onse in he 832 o 7,924
pa ien s who de eloped muscula symp oms was one mon h
a e s a in ini ia ion.50 A s udy o 45 pa ien s epo ed a mean
du a ion o s a in he apy be o e myopa hic symp oms o 6.3
mon hs, wi h a maximum o 9.8 mon hs.19
A la ge s udy, using a case-c osso e design o de e mine
s a in myo oxici y in wo p ima y-ca e da abases comp ising
93,831 pa ien s, sugges ed ha mos cases occu wi hin he
i s 12 weeks o s a in exposu e.
21
The au ho s ecommended
a 26-week cu o o enable ewe misclassi ica ions o exposed
cases.21 A 26-week cu o seems sensible o hose pa ien s
who ha e been on a s able dose o s a in mono he apy; how-
e e , his may need o be ex ended. I is impo an o no e
ha some pa ien s can de elop s a in myo oxici y ei he a e
a s a in dose inc ease o a e he concomi an adminis a ion
o an in e ac ing d ug, which may occu any ime du ing he
li e cycle o s a in use. Fu he mo e, au oimmune myopa hy
may ake longe o de elop, wi h a ime o onse as long as
3yea s.51
RISK FACTORS
Concomi an medica ions
Fib a es. A signi ican body o e idence sugges s an inc eased
isk o s a in myopa hy, pa icula ly habdomyolysis, in pa ien s
aking s a ins in combina ion wi h ib a es.4 Analysis o he
US Food and D ug Adminis a ion Ad e se E en Repo ing
Sys em be ween 1998 and 2002 was used o de e mine epo -
ing a es o habdomyolysis in pa ien s aking eno ib a e and
gem ib ozil in combina ion wi h s a ins.52 Gem ib ozil, a ib ic
acid de i a i e, inc eases sys emic exposu e o ac i e sim as a-
in acid by inhibi ing bo h glucu onida ion and o ganic anion-
anspo ing polypep ide (OATP)1B1 anspo e –media ed
up ake in o he li e .53,54
O e all a es o habdomyolysis o any s a in medica ion
use s cop esc ibed eno ib a e o gem ib ozil we e 4.5 and 8.7
pe million p esc ip ions, espec i ely. When s a i ied o hose
pa ien s ecei ing ce i as a in only, he a es inc eased o 140
and 4,600 pe million p esc ip ions wi h eno ib a e/ce i as a in
and gem ib ozil/ce i as a in combina ion he apy, espec i ely.52
Al hough he e ec o ib a e coadminis a ion on s a in bio-
a ailabili y exis s o he majo i y o s a ins,53,55–57 he inc ease
in myopa hy isk is especially p onounced wi h ce i as a in.5,58
Al hough ce i as a in was wi hd awn in 2001, he iden i ica ion
o p edisposing pha macogenomics ac o s may s ill be o use
o elucida ing p edisposing ac o s o muscle oxici y wi h he
s a ins mo e commonly used nowadays.
CYP3A4 inhibi ion. CYP3A4 inhibi o s such as azole an i un-
gals, p o ease inhibi o s, amioda one, cyclospo ine, calcium
channel blocke s, and mac olide an ibio ics, o name a ew,
inc ease isk o myopa hy o s a ins ha unde go CYP3A4
me abolism (sim as a in, a o as a in, and lo as a in).59–62 In
addi ion, some oods such as g ape ui juice, which con ains
u anocouma ins, i e e sibly inhibi CYP3A4 in he gu . The
e ec is educed gu wall me abolism o s a ins (pa icula ly
sim as a in) and inc eased sys emic exposu e, which can lead
o ad e se e ec s.60,63
Como bidi ies
A lis o como bidi ies ha a e associa ed wi h an inc eased isk
o de eloping s a in- ela ed myo oxici y is shown in Table 2.
They include hypo hy oidism, ch onic enal insu iciency, in ec-
ion, impai ed li e unc ion, hype ension, physical exe ion,
and diabe es.
Exe cise
I is commonly belie ed ha igo ous exe cise inc eases he isk
o s a in-induced myopa hy. Indeed, i is hough ha myopa hic
symp oms may occu in 25% o s a in use s who exe cise, as
compa ed wi h a popula ion incidence es ima ed a 1–5% o
hose who exe cise bu do no ake s a ins.64 In p o essional
a hle es, i is es ima ed ha as many as 75% o hose aking
s a ins may de elop muscula symp oms;65 howe e , his may
be o e es ima ed.
CONCLUSIONS
Gi en he high p e alence o s a in use wo ldwide and hei sig-
ni icance in he p e en ion o ca dio ascula and ce eb o ascu-
la disease, ex ensi e esea ch on he p edic ion and p e en ion
o se ious ad e se e ec s such as s a in- ela ed myo oxici y is
jus i ied. Imp o ed pa ien ole abili y and adhe ence o s a ins
is c ucial because i educes he incidence and he cos o any
heal h-ca e sys em o ea ing ca dio ascula disease. La ge
p ospec i e s udies o pa ien coho s ea ed wi h s a ins a e
equi ed o iden i y new gene ic suscep ibili y bioma ke s o
s a in- ela ed myo oxici y ha could be implemen ed in o
clinical p ac ice. To da e, one o he p oblems in compa ing
he esul s o obse a ional s udies on s a in myo oxici y has
been he lack o pheno ype classi ica ion and s anda diza ion o
nomencla u e. In addi ion, gi en he a i y o he mos se e e
pheno ypes, a small sample size o se e al s udies has hampe ed
gene ic bioma ke disco e y.
We ha e adap ed a p e iously desc ibed consensus app oach15
o de ine pheno ypic c i e ia ha can be used in an e o o
s anda dize s a in- ela ed myo oxici y pheno ypes using he
nume ical and desc ip i e nomencla u e gi en in Table 1. Ou
s anda diza ion was based on expe opinion om a mul idis-
ciplina y g oup and he li e a u e. The ollowing a e key c i e ia
o be used in he deep pheno yping o pa ien s wi h suspec ed
s a in-induced muscle inju y:
• A CK le el ha is >4× he ULN in he p esence o absence
o clinical symp oms o he de ini ion o myopa hy. We
ha e adop ed he >4× he ULN le el p agma ically, as we
eel, based on he li e a u e, ha his cu o p o ides he
igh balance in p e en ing inclusion o pa ien s wi h CK
ele a ion due o no mal a ia ion o o he causes, while
474 VOLUME 96 NUMBER 4 | OCTOBER 2014 | www.na u e.com/cp

Re iew
simul aneously ensu ing ha we do no unnecessa ily
exclude aluable pa ien s in s udies in es iga ing gene ic
ac o s p edisposing o s a in myo oxici y. A CK >10× he
ULN should be ca ego ized as habdomyolysis i accompa-
nied by enal impai men . The adop ion o he classi ica-
ion shown in Table 1 may help in mo e clea ly de ining
he a ious pheno ypes ha a e ec ui ed in di e en
s udies. Pape s epo ing gene ic ac o s p edisposing o
s a in-induced muscle inju y should show he e ec size
o he gene ic polymo phism based on he deg ee o CK
ele a ion, as has been done in ecen s udies.9,66
• Clinical symp oms need o be ca e ully eco ded in he
pa ien s; hese should include no only he muscle symp-
oms bu also whe he he e is any in ol emen o he
kidneys, which migh indica e habdomyolysis. Causali y
assessmen as o whe he he s a in was esponsible should
include de ails o he empo al ela ionship be ween
onse o s a in use and he occu ence o myo oxici y, he
e ec o dechallenge, and he e ec o any echallenge.
Dechallenge may no always be success ul, o example,
in pa ien s wi h au oimmune myopa hy, and should no
necessa ily be used o exclude s a ins as e iological agen s.
• Apa om CK, measu emen o alanine amino ans e ase,
u ine myoglobin le els (when clinically indica ed), and
enal unc ion may be use ul. In pa ien s wi h a suspec ed
au oimmune myopa hy, he measu emen o an i-HMGCR
an ibodies and muscle biopsy should be conside ed.
• P edisposing ac o s, including in e ac ing d ugs, como -
bidi ies, and exe cise, should be e alua ed in all pa ien s.
Fac o s such as auma ha lead o muscle inju y i espec-
i e o s a in use need o be excluded.
• The ime o onse can be a iable and can be delayed as
long as 3 yea s in pa ien s wi h au oimmune myopa hy.
Howe e , in gene al, mos cases o s a in-induced myo-
oxici y occu wi hin 6 mon hs o 1 yea o s a in onse ,
an inc ease in he dose o he s a in, o he concomi an
adminis a ion o an in e ac ing d ug.
We ha e also de eloped an algo i hm ha will help assign
pheno ypes o indi idual pa ien s based on clinical and bio-
chemical pa ame e s (Figu e 1).
We hope he c i e ia desc ibed in his a icle will help clini-
cians and esea che s o ca ego ize pheno ypes in pa ien s wi h
s a in- ela ed myo oxici y in o de o acili a e esea ch in his
a ea.
SUPPLEMENTARY MATERIAL is linked o he online e sion o he pape a
h p://www.na u e.com/cp
ACKNOWLEDGMENTS
This p ojec has ecei ed unding om he Eu opean Union’s Se en h
F amewo k P og amme o esea ch, echnological de elopmen , and
demons a ion unde g an ag eemen 602108. This wo k was p esen ed
a he Pheno ype S anda diza ion Wo kshop, Li e pool, UK, 10 Decembe
2013.
CONFLICT OF INTEREST
The au ho s decla ed no con lic o in e es .
© 2014 Ame ican Socie y o Clinical Pha macology and The apeu ics
1. Law,M. & Rudnicka,A.R. S a in sa e y: a sys ema ic e iew. Am. J. Ca diol. 97,
52C–60C (2006).
2. Sa hasi am,S. S a in induced myo oxici y. Eu . J. In e n. Med. 23, 317–324
(2012).
3. S one,N.J. e al. 2013 ACC/AHA Guideline on he T ea men o Blood
Choles e ol o Reduce A he oscle o ic Ca dio ascula Risk in Adul s: A Repo
o he Ame ican College o Ca diology/Ame ican Hea Associa ion Task Fo ce
on P ac ice Guidelines. Ci cula ion 2013; e-pub ahead o p in 12 No embe
2013.
4. G aham,D.J. e al. Incidence o hospi alized habdomyolysis in pa ien s
ea ed wi h lipid-lowe ing d ugs. JAMA 292, 2585–2590 (2004).
5. S a a,J.A., Chang,J. & G een,L. Ce i as a in and epo s o a al
habdomyolysis. N. Engl. J. Med. 346, 539–540 (2002).
6. Floyd,J.S., Heckbe ,S.R., Weiss,N.S., Ca ell,D.S. & Psa y,B.M. Use
o adminis a i e da a o es ima e he incidence o s a in- ela ed
habdomyolysis. JAMA 307, 1580–1582 (2012).
7. Pas e nak,R.C., Smi h,S.C.J , Bai ey-Me z,C.N., G undy,S.M., Cleeman,J.I. &
Len an ,C.; Ame ican College o Ca diology; Ame ican Hea Associa ion;
Na ional Hea , Lung and Blood Ins i u e. ACC/AHA/NHLBI Clinical Ad iso y
on he Use and Sa e y o S a ins. Ci cula ion 106, 1024–1028 (2002).
8. Ai hal,G.P. & Daly,A.K. P eemp ing and p e en ing d ug-induced li e inju y.
Na . Gene . 42, 650–651 (2010).
9. Ca ,D.F. e al. SLCO1B1 gene ic a ian associa ed wi h s a in-induced
myopa hy: a p oo -o -concep s udy using he clinical p ac ice esea ch
da alink. Clin. Pha macol. The . 94, 695–701 (2013).
10. Chang,C.H., Kusama,M., Ono,S., Sugiyama,Y., O ii,T. & Akazawa,M.
Assessmen o s a in-associa ed muscle oxici y in Japan: a coho s udy
conduc ed using claims da abase and labo a o y in o ma ion. BMJ Open 3,
(2013).
11. Pi mohamed,M., Ai hal,G.P., Beh ,E., Daly,A. & Roden,D. The
pheno ypes anda diza ion p ojec : imp o ing pha macogene ic s udies
o se ious ad e se d ug eac ions. Clin. Pha macol. The . 89, 784–785
(2011).
12. Sai,K. e al. De elopmen o a de ec ion algo i hm o s a in-induced
myopa hy using elec onic medical eco ds. J. Clin. Pha m. The . 38, 230–235
(2013).
13. Ai hal,G.P. e al. Case de ini ion and pheno ype s anda diza ion in d ug-
induced li e inju y. Clin. Pha macol. The . 89, 806–815 (2011).
14. Beh ,E.R. e al. The In e na ional Se ious Ad e se E en s Conso ium (iSAEC)
pheno ype s anda diza ion p ojec o d ug-induced o sades de poin es. Eu .
Hea J. 34, 1958–1963 (2013).
15. Pi mohamed,M. e al. Pheno ype s anda diza ion o immune-media ed
d ug-induced skin inju y. Clin. Pha macol. The . 89, 896–901 (2011).
16. Agbabiaka,T.B., Sa o ić,J. & E ns ,E. Me hods o causali y
assessmen o ad e se d ug eac ions: a sys ema ic e iew. D ug Sa . 31,
21–37 (2008).
17. Cham,S., E ans,M.A., Denenbe g,J.O. & Golomb,B.A. S a in-associa ed
muscle- ela ed ad e se e ec s: a case se ies o 354 pa ien s. Pha maco he apy
30, 541–553 (2010).
18. deLemos,J.A. e al.; In es iga o s. Ea ly in ensi e s a delayed conse a i e
sim as a in s a egy in pa ien s wi h acu e co ona y synd omes: phase Z o
he A o Z ial. JAMA 292, 1307–1316 (2004).
19. Hansen,K.E., Hildeb and,J.P., Fe guson,E.E. & S ein,J.H. Ou comes in 45
pa ien s wi h s a in-associa ed myopa hy. A ch. In e n. Med. 165, 2671–2676
(2005).
20. Kashani,A. e al. Risks associa ed wi h s a in he apy: a sys ema ic o e iew o
andomized clinical ials. Ci cula ion 114, 2788–2797 (2006).
21. Molokhia,M., McKeigue,P., Cu cin,V. & Majeed,A. S a in induced myopa hy
and myalgia: ime end analysis and compa ison o isk associa ed wi h s a in
class om 1991-2006. PLoS One 3, e2522 (2008).
22. Oshima,Y. Cha ac e is ics o d ug-associa ed habdomyolysis: analysis o
8,610 cases epo ed o he U.S. Food and D ug Adminis a ion. In e n. Med.
50, 845–853 (2011).
23. P e e ,M.A. e al. Sa e y and ole abili y o p a as a in in long- e m clinical
ials: p ospec i e P a as a in Pooling (PPP) P ojec . Ci cula ion 105,
2341–2346 (2002).
24. Ridke ,P.M. e al.; JUPITER S udy G oup. Rosu as a in o p e en ascula
e en s in men and women wi h ele a ed C- eac i e p o ein. N. Engl. J. Med.
359, 2195–2207 (2008).
25. Czi aky,M.J. e al. Risk o hospi alized habdomyolysis associa ed wi h
lipid-lowe ing d ugs in a eal-wo ld clinical se ing. J. Clin. Lipidol. 7, 102–108
(2013).
ClInICAl PhARMACology & TheRAPeuTICS | VOLUME 96 NUMBER 4 | OCTOBER 2014 475
Re iew
26. Gilad,R. & Lampl,Y. Rhabdomyolysis induced by sim as a in and
ke oconazole ea men . Clin. Neu opha macol. 22, 295–297 (1999).
27. Me iggioli,M.N., Ba boi,A.C., Rowin,J. & Coch an,E.J. HMG-CoA Reduc ase
Inhibi o Myopa hy: Clinical, Elec ophysiological, and Pa hologic Da a in Fi e
Pa ien s. J. Clin. Neu omuscul. Dis. 2, 129–134 (2001).
28. Phillips,P.S. e al.; Sc ipps Me cy Clinical Resea ch Cen e . S a in-associa ed
myopa hy wi h no mal c ea ine kinase le els. Ann. In e n. Med. 137, 581–585
(2002).
29. Pie ce,L.R., Wysowski,D.K. & G oss,T.P. Myopa hy and habdomyolysis
associa ed wi h lo as a in-gem ib ozil combina ion he apy. JAMA 264,
71–75 (1990).
30. Hohenegge ,M. D ug induced habdomyolysis. Cu . Opin. Pha macol. 12,
335–339 (2012).
31. Elhayany,A., Mishaal,R.A. & Vinke ,S. Is he e clinical bene i o ou ine
enzyme es ing o pa ien s on s a ins? Expe Opin. D ug Sa . 11, 185–190
(2012).
32. Fung,E.C. & C ook,M.A. S a in myopa hy: a lipid clinic expe ience
on he ole abili y o s a in echallenge. Ca dio asc. The . 30, e212–e218
(2012).
33. Lems a,M., Blackbu n,D., C awley,A. & Fung,R. P opo ion and isk indica o s
o nonadhe ence o s a in he apy: a me a-analysis. Can. J. Ca diol. 28,
574–580 (2012).
34. Blaie ,O., Lishne ,M. & Elis,A. Managing s a in-induced muscle oxici y in a
lipid clinic. J. Clin. Pha m. The . 36, 336–341 (2011).
35. Link,E. e al.; SEARCH Collabo a i e G oup. SLCO1B1 a ian s and s a in-
induced myopa hy–a genomewide s udy. N. Engl. J. Med. 359, 789–799
(2008).
36. Wilke,R.A. e al.; Clinical Pha macogenomics Implemen a ion Conso ium
(CPIC). The clinical pha macogenomics implemen a ion conso ium: CPIC
guideline o SLCO1B1 and sim as a in-induced myopa hy. Clin. Pha macol.
The . 92, 112–117 (2012).
37. B unham,L.R. e al. Di e en ial e ec o he s4149056 a ian in SLCO1B1 on
myopa hy associa ed wi h sim as a in and a o as a in. Pha macogenomics J.
12, 233–237 (2012).
38. Donnelly,L.A. e al. Common nonsynonymous subs i u ions in
SLCO1B1p edispose o s a in in ole ance in ou inely ea ed indi iduals
wi h ype 2 diabe es: a go-DARTS s udy. Clin. Pha macol. The . 89, 210–216
(2011).
39. Donnelly,L.A. e al. Robus associa ion o he LPA locus wi h low-densi y
lipop o ein choles e ol lowe ing esponse o s a in ea men in a me a-
analysis o 30 467 indi iduals om bo h andomized con ol ials and
obse a ional s udies and associa ion wi h co ona y a e y disease ou come
du ing s a in ea men . Pha macogene . Genomics 23, 518–525 (2013).
40. Linde,R., Peng,L., Desai,M. & Feldman,D. The ole o i amin D and
SLCO1B1*5 gene polymo phism in s a in-associa ed myalgias.
De ma oendoc inol. 2, 77–84 (2010).
41. Ma cian e,K.D. e al. Ce i as a in, gene ic a ian s, and he isk o
habdomyolysis. Pha macogene . Genomics 21, 280–288 (2011).
42. Voo a,D. e al. The SLCO1B1*5 gene ic a ian is associa ed wi h s a in-
induced side e ec s. J. Am. Coll. Ca diol. 54, 1609–1616 (2009).
43. Newman,C., Tsai,J., Sza ek,M., Luo,D. & Gibson,E. Compa a i e sa e y o
a o as a in 80 mg e sus 10 mg de i ed om analysis o 49 comple ed ials
in 14,236 pa ien s. Am. J. Ca diol. 97, 61–67 (2006).
44. An ons,K.A., Williams,C.D., Bake ,S.K. & Phillips,P.S. Clinical pe spec i es o
s a in-induced habdomyolysis. Am. J. Med. 119, 400–409 (2006).
45. Black,C. & Jick,H. E iology and equency o habdomyolysis.
Pha maco he apy 22, 1524–1526 (2002).
46. Needham,M., Fabian,V., Kneze ic,W., Panegy es,P., Zilko,P. & Mas aglia,F.L.
P og essi e myopa hy wi h up- egula ion o MHC-I associa ed wi h s a in
he apy. Neu omuscul. Diso d. 17, 194–200 (2007).
47. Mammen,A.L. e al. Au oan ibodies agains 3-hyd oxy-3-me hylglu a yl-
coenzyme A educ ase in pa ien s wi h s a in-associa ed au oimmune
myopa hy. A h i is Rheum. 63, 713–721 (2011).
48. Mammen,A.L. e al. Ra i y o an i-3-hyd oxy-3-me hylglu a yl-coenzyme
A educ ase an ibodies in s a in use s, including hose wi h sel -limi ed
musculoskele al side e ec s. A h i is Ca e Res. (Hoboken). 64, 269–272 (2012).
49. Cowa d,W.R., Ma ei,A., Yang,A., Vasa-Nico e a,M.M. & Chow,S.C. S a in-
induced p oin lamma o y esponse in mi ogen-ac i a ed pe iphe al blood
mononuclea cells h ough he ac i a ion o caspase-1 and IL-18 sec e ion in
monocy es. J. Immunol. 176, 5284–5292 (2006).
50. B ucke ,E., Hayem,G., Dejage ,S., Yau,C. & Bégaud,B. Mild o mode a e
muscula symp oms wi h high-dosage s a in he apy in hype lipidemic
pa ien s– he PRIMO s udy. Ca dio asc. D ugs The . 19, 403–414 (2005).
51. G able-Esposi o,P., Ka zbe g,H.D., G eenbe g,S.A., S ini asan,J., Ka z,J. &
Ama o,A.A. Immune-media ed nec o izing myopa hy associa ed wi h s a ins.
Muscle Ne e 41, 185–190 (2010).
52. Jones,P.H. & Da idson,M.H. Repo ing a e o habdomyolysis wi h
eno ib a e + s a in e sus gem ib ozil + any s a in. Am. J. Ca diol. 95, 120–122
(2005).
53. Backman,J.T., Ky klund,C., Ki is ö,K.T., Wang,J.S. & Neu onen,P.J. Plasma
concen a ions o ac i e sim as a in acid a e inc eased by gem ib ozil. Clin.
Pha macol. The . 68, 122–129 (2000).
54. Shi a a,Y. & Sugiyama,Y. Pha macokine ic and pha macodynamic al e a ions
o 3-hyd oxy-3-me hylglu a yl coenzyme A (HMG-CoA) educ ase inhibi o s:
d ug-d ug in e ac ions and in e indi idual di e ences in anspo e and
me abolic enzyme unc ions. Pha macol. The . 112, 71–105 (2006).
55. Ky klund,C., Backman,J.T., Ki is ö,K.T., Neu onen,M., Lai ila,J. &
Neu onen,P.J. Plasma concen a ions o ac i e lo as a in acid a e ma kedly
inc eased by gem ib ozil bu no by beza ib a e. Clin. Pha macol. The . 69,
340–345 (2001).
56. Ky klund,C., Backman,J.T., Neu onen,M. & Neu onen,P.J. Gem ib ozil
inc eases plasma p a as a in concen a ions and educes p a as a in enal
clea ance. Clin. Pha macol. The . 73, 538–544 (2003).
57. Schneck,D.W. e al. The e ec o gem ib ozil on he pha macokine ics o
osu as a in. Clin. Pha macol. The . 75, 455–463 (2004).
58. Ozdemi ,O., Bo an,M., Gökçe,V., Uzun,Y., Koçak,B. & Ko kmaz,S. A case
wi h se e e habdomyolysis and enal ailu e associa ed wi h ce i as a in-
gem ib ozil combina ion he apy–a case epo . Angiology 51, 695–697 (2000).
59. Ray,G.M. An i e o i al and s a in d ug-d ug in e ac ions. Ca diol. Re . 17,
44–47 (2009).
60. So okin,A.V., Duncan,B., Pane a,R. & Thompson,P.D. Rhabdomyolysis
associa ed wi h pomeg ana e juice consump ion. Am. J. Ca diol. 98, 705–706
(2006).
61. Zhou,S., Chan,E., Li,X. & Huang,M. Clinical ou comes and managemen o
mechanism-based inhibi ion o cy och ome P450 3A4. The . Clin. Risk Manag.
1, 3–13 (2005).
62. Zhou,S. e al. Mechanism-based inhibi ion o cy och ome P450 3A4 by
he apeu ic d ugs. Clin. Pha macokine . 44, 279–304 (2005).
63. Mazokopakis,E.E. Unusual causes o habdomyolysis. In e n. Med. J. 38,
364–367 (2008).
64. Di ks,A.J. & Jones,K.M. S a in-induced apop osis and skele al myopa hy. Am.
J. Physiol. Cell Physiol. 291, C1208–C1212 (2006).
65. Sinzinge ,H. & O’G ady,J. P o essional a hle es su e ing om amilial
hype choles e olaemia a ely ole a e s a in ea men because o muscula
p oblems. B . J. Clin. Pha macol. 57, 525–528 (2004).
66. O’Mea a,H. e al. Elec onic heal h eco ds o biological sample collec ion:
easibili y s udy o s a in-induced myopa hy using he Clinical P ac ice
Resea ch Da alink. B . J. Clin. Pha macol. 77, 831–838 (2014).
67. Bays,H. S a in sa e y: an o e iew and assessmen o he da a--2005. Am. J.
Ca diol. 97, 6C–26C (2006).
68. Moghadasian,M.H., Mancini,G.B. & F ohlich,J.J. Pha maco he apy o
hype choles e olaemia: s a ins in clinical p ac ice. Expe Opin. Pha maco he .
1, 683–695 (2000).
69. Hea P o ec ion S udy Collabo a i e G oup. MRC/BHF Hea P o ec ion S udy
o choles e ol lowe ing wi h sim as a in in 20,536 high- isk indi iduals: a
andomised placebo-con olled ial. Lance 360, 7–22 (2002).
70. Mohassel,P. & Mammen,A.L. S a in-associa ed au oimmune myopa hy and
an i-HMGCR au oan ibodies. Muscle Ne e 48, 477–483 (2013).
This wo k is licensed unde a C ea i e Commons
A ibu ion-nonComme cial-noDe i s 3.0
unpo ed license. The images o o he hi d pa y ma e ial
in his a icle a e included in he a icle’s C ea i e Commons
license, unless indica ed o he wise in he c edi line; i he
ma e ial is no included unde he C ea i e Commons
license, use s will need o ob ain pe mission om he
license holde o ep oduce he ma e ial. To iew a copy o
his license, isi h p://c ea i ecommons.o g/licenses/
by-nc-nd/3.0/
476 VOLUME 96 NUMBER 4 | OCTOBER 2014 | www.na u e.com/cp